PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
This PDF file contains the front matter associated with SPIE Proceedings Volume 13447, including the Title Page, Copyright information, Table of Contents, and Conference Committee information
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Considering the problem concerning voltage beyond limit caused by the photovoltaic generators in low-voltage distribution network system, a power-voltage coordinated manage method about inverters is offered to deal with voltage beyond limit by photovoltaic generators. This method can give priority to voltage adjustment through reactive power belonging to the PV, if the reactive regulation ability from the PV reaches its limit while the voltage continues to exceed the limit, an active power reduction control based on sparrow search algorithm is further adopted to minimize the waste of photovoltaic active power and avoid voltage beyond limit. The simulation outcomes confirm the potency as well as advantages belonging to the offered method in dealing with the beyond-limits about voltage created by the PV.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This study aims to explore the application of mechanical vibration control technology in improving environmental safety in mining. By introducing vibration machinery and screening machine technology, combined with the concept of intrinsic safety technology, a mechanical vibration control system suitable for mining is proposed. Based on advanced algorithm design, the system monitors and dynamically adjusts the vibration parameters in mining operations in real time through multi-dimensional system simulation. The simulation process covers the analysis of key indicators such as vibration screening efficiency, mechanical stability and environmental impact, and verifies the effectiveness of the system through precise data. The research results show that this technology has significant advantages in improving the screening efficiency and safety performance in the mining process, especially in reducing dust diffusion and noise pollution. By optimizing the parameter configuration of the vibration screening machine, the system can ensure efficient operation while greatly improving the intrinsic safety of the operation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In addressing the issues of ambiguity and volatility in collecting value at a specific moment as testing data during automated laboratory testing report generation, the method is proposed by borrowing from manual operational procedures by adopting a sliding window to continuously collect data, and the ADF (Augmented Dickey-Fuller) method is utilized to automatically determine the stability of the data, and the maximum-minimum difference method should be applied to set limits on the fluctuation range, and calculate the average value as the testing data. The morphological algorithm is employed to highlight the features of text in testing picture to reduce the impact of color and texture features of the testing sample, and OCR (Optical Character Recognition) technology is adopted to identify the text and its orientation. By reconstructing the coordinate axes to determine the quadrant where the text is located, combined with prior knowledge, the correct orientation of the testing picture is judge, which overcomes problem that the characteristics of the testing picture are complex and the text orientations is not unified, which lead to the overall orientation is difficult to judge. Consequently, this is convenient to insert the testing report, improves work efficiency, and alleviates the "slow to test" issue in the testing industry.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the rapid development of industrial automation, the application of machine vision technology in assembly lines is becoming more and more extensive. This study mainly discusses the application of machine vision technology in industrial automation assembly lines, focusing on the visual inspection system based on an edge detection algorithm. By using the edge detection algorithm in image processing technology, the accuracy of the workpiece's position, shape and size is realized, effectively improving the automation level and work efficiency of the assembly line. This paper designs a complete machine vision system, including image acquisition, preprocessing, feature extraction and subsequent detection algorithm. This paper conducts detailed algorithm simulation and model testing to verify the system's effectiveness. The simulation results show that the designed edge detection algorithm has high accuracy and stability in the industrial automation environment; the detection accuracy reaches 0.01mm, and the system error is controlled within 0.5%. This study provides strong technical support for the future intelligent development of industrial automation assembly lines.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Based on the data of Beijing-Tianjin-Hebei region from 2011 to 2019,this paper measures the medical efficiency of Beijing-Tianjin-Hebei region in recent ten years by using the super efficiency SBM model, and uses the ML index decomposition method to conduct a dynamic analysis of medical efficiency, and further discusses the calculation results through convergence analysis, the medical efficiency in Beijing-Tianjin-Hebei region was studied. The study found that the overall medical efficiency in Beijing-Tianjin-Hebei region showed an increasing trend, but the increase was not large, and there was greater room for progress; the convergence of ML index changes is also high, indicating the overall development of Beijing-Tianjin-Hebei region.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In recent years, the field of automation engineering is rapid advancements and technological breakthroughs, resulting increasing demand for qualified and innovative professionals. To meet these requirements, higher education and vocational education in engineering education sector are constantly seeking effective teaching models that foster innovation and nurture talent in automation engineering and other engineering fields. One such teaching model that has gained prominence is the Engineering Practice Innovation Project (EPIP) teaching model. This paper aims to promote and evaluate the effectiveness of the EPIP teaching model in fostering innovation and nurturing talents in automation engineering education. The study investigates the impact of the EPIP model on the trainee’s ability to apply theoretical knowledge to practical projects, on their learning outcomes, and the development of their creativity and innovative skills. Preliminary findings indicate that the EPIP teaching model positively impacts on students' learning outcomes and innovation skills. Moreover, the collaborative nature of the EPIP model encourages teamwork, thinking out of box, and increase engineering practice skills, essential for success in the automation engineering industry. Furthermore, the EPIP model promotes innovation by providing students with opportunities to explore novel ideas and develop creative solutions and also plays a crucial role in talent development in automation engineering. The EPIP teaching model is designed to bridge the gap between theory and practice by providing trainees with practical experience in solving real-world problems. The results shows that EPIP teaching model for fostering innovation and talent in automation engineering education is effective. Trainees feedback is used to analyses the effectiveness fostering innovation and talents in automation engineering education through Engineering Practice Innovation Project (EPIP) teaching model. More than 90% of the trainee’s positive perception that through EPIP teaching models greatly motivate their learning, innovation, talent skill in the course. The trainee’s positive perception for EPIP that to ensures their continued growth and success in the engineering industry. It involves trainees working on industry-relevant projects, collaborating with professionals, and applying their knowledge to develop innovative solutions. This approach not only enhances trainees’ technical skills but also cultivates creative, critical thinker, and innovators.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the rapid development and widespread application of cloud computing, container technology has become a hot topic for enterprises and research institutions in recent years. This article proposes an automated container arrangement and resource optimization algorithm design based on cloud computing technology to address these issues. This algorithm can predict and adjust resources based on real-time and historical data of applications, ensuring efficient resource utilization and good application performance. The experimental results show that our optimization algorithm has improved by about 16% compared to the optimal adaptation strategy. Due to the more precise resource allocation of the algorithm in this article, the response time of the application has not been significantly affected. In the testing of system throughput, it can be found that our algorithm has improved throughput by about 15% compared to random strategies. Finally, energy consumption was tested and the results showed that our algorithm reduced energy consumption by approximately 28% due to higher resource utilization. These results demonstrate that our algorithm outperforms existing strategies in multiple key metrics.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In the rapidly evolving landscape of cybersecurity, the need for sophisticated intrusion detection is paramount. Traditional methods often struggle to keep pace with the complexity and diversity of emerging threats. This paper presents an enhanced Bayesian network classifier (EBNC), integrating a semi-lazy learning framework and multi-conditional entropy to enhance the accuracy and efficiency of intrusion detection. The EBNC dynamically constructs class-specific local classifiers for each testing instance, thereby optimizing the decision-making process. It also employs multi-conditional entropy to quantify causal and conditional dependencies implicated in each testing instance. Experiments conducted on the NSL-KDD dataset demonstrate that the EBNC outperforms alternative algorithms with respect to the 0-1 loss, bias, F1 score, as well as training and classification time. These results highlight the EBNC's efficiency and accuracy in intrusion detection, demonstrating its adaptability to new threats and its capacity for rapid, informed responses in the expanding field of cybersecurity.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Motor imagery electroencephalogram recognition is a key area in brain-computer interfaces, with applications in human-computer interaction, rehabilitation, and virtual reality. Traditional methods often overlook the brain's topological characteristics and the correlations between EEG channels, resulting in suboptimal decoding. To address this, we propose an adaptive spatio-temporal graph neural network (ASTGNN), which constructs a brain graph topology and adaptively learns the topological adjacency matrix, exploring both common and individual electrode connections. The spatial features are extracted using a graph convolutional network, while a gated position-aware self-attention mechanism captures movement information and global dependencies, enhancing temporal feature extraction. Experiments show that ASTGNN significantly improves recognition, achieving an average accuracy of 87.36%.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The research and application of big data processing and analysis have been quite mature, but more and more fields have put forward demands for real-time analysis and rapid response of fast and massive distributed big data. Distributed big data framework performance needs to be optimized. To this end, the article studies and analyzes the platform architecture and data computing model of typical batch processing technology Hadoop, memory computing technology Spark, and stream computing technology Storm, and summarizes the similarities and differences of these three big data processing technologies. Then, on the basis of studying the big data platform and traditional cardinality estimation algorithm, a HyperLogLog algorithm application model based on the streaming platform is proposed, and the cardinality calculation is performed in the Storm processing engine to achieve the performance of the distributed big data processing framework optimization. The results show that the HyperLogLog cardinality estimation algorithm of the Storm stream computing platform can be used to optimize the performance of the distributed big data processing framework.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the gradual deepening of the development of mineral resources, people's demand for accurate detection of mineral components is also increasing. High-precision electron probe technology has become an important tool in mineral composition detection due to its efficient micro-chemical composition analysis. In this paper, we will discuss the application of high-precision electron probes (EPMA) in the detection of mineral composition, focusing on the key technical factors that affect the accuracy of its analysis. Based on the comprehensive application of theoretical analysis and empirical experiments and data analysis methods, some of the operating parameters are optimized, and the content changes of various components such as iron, zinc, cadmium and other components in minerals under different geological conditions are analyzed. The results show that the adjustment of the analysis parameters of the high-precision electron probe can significantly improve the accuracy and reliability of the mineral composition detection process, and provide strong scientific support for the evaluation of mineral resources and geological exploration.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper introduces a facial expression recognition method based on a residual network that incorporates a channel-space attention mechanism module. The method enhances the ResNet50 architecture and integrates label smoothing as a regularization technique. These innovations effectively reduce the impact of noise in facial expression recognition tasks, improve the model's generalization ability, and significantly boost recognition accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
False propaganda, as one of the unfair competition behaviors, seriously damages consumer rights and disrupts market order. This article aims to design a false propaganda detection technology based on LLM and BERT for dishonest behavior caused by false propaganda. By annotating false propaganda text data with LLM and training a false propaganda detection model based on BERT, the rapid development of the false propaganda detection model and effective cost savings in manual annotation have been achieved. The results show that this method has good performance and efficiency, and can meet practical engineering needs. It has certain application and reference value for model development in market supervision fields such as false propaganda.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Military threat target image detection is one of the research difficulties in the field of image processing and target recognition. The characteristics of small target, complex background and similar features make it difficult to identify unmanned intelligent equipment. Aiming at this problem, this paper proposes a new threat target detection algorithm based on two-stage deep network. The algorithm can identify multi-threat target images. Faster R-CNN has been improved using two methods: feature pyramid network and Mosaic image enhancement. The accuracy rate is 7.3% higher than that before improvement, which provides new ideas and methods for military threat target image detection.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Active illumination in underwater imaging encounters two significant degradation issues: color distortion and uneven lighting, which complicate the analysis of underwater images and videos. This study introduces a network for enhancing underwater low-light images that utilizes techniques for color correction and brightness estimation. The network comprises four sub-networks: the Image Decomposition Network, which utilizes an encoder-decoder architecture with dilated convolutions for global luminance estimation, generating illumination and reflection images; the Illumination-Guided Enhancement Block, which performs channel-wise concatenation of the illumination and reflection images; the Noise Suppression and Fusion Model, which reduces noise in the image using the illumination map; and the Color Restoration Network, which handles color correction. Results from experiments conducted on real-world data show that the proposed method effectively mitigates issues related to uneven lighting and color deviation in underwater low-light images, surpassing conventional methods in both objective evaluation metrics and subjective visual quality.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Target detection is a core task in computer vision, but traditional methods based on RGB images often encounter challenges such as computational complexity and high resource consumption. In contrast, event cameras have garnered significant attention due to their advantages, including high dynamic range, real-time performance, low data redundancy, and fast sensing. This paper introduces a novel target detection method grounded in event-time surface feature fusion. The proposed neural network architecture integrates four key modules: (1) extraction of biologically-inspired time-surface features from event cloud data by leveraging temporal information within local spatial neighborhoods; (2) utilization of a self-attention mechanism within the event cloud feature extraction network to capture point features; (3) fusion of event time-surface features with point features; and (4) sampling and grouping techniques to generate voting centers and clusters, with a proposal module providing final object localization and recognition. Experimental results demonstrate that this event-time surface-based method effectively harnesses raw event data, significantly improving detection accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In recent years, with the tomato market demand continues to expand, China's tomato cultivation area increases year by year, its consumption of manpower, material and financial resources are also expanding, especially in the manpower link is in the tomato picking this link in the consumption of too much, and nowadays the aging population is serious, the lack of labor force to go to the picking, in order to solve the problem, this paper is based on the YOLOv5s model for the rapid recognition of ripe tomato fruit in a complex environment. In order to solve this problem, based on YOLOv5s model, this paper carries out an in-depth study on the rapid recognition of ripe tomato fruits in complex environments, so as to deploy the control module on the picking robot at a later stage. Firstly, fruit image datasets of tomatoes with different maturity (green and red) are collected, then labeled using labeling for deep neural network model training, and finally validated on a test set. The experimental results show that Mature tomatoes and Immature tomatoes precision rate is 97.1% and 98%, recall rate is 94.8% and 95.9%, and the average precision mean is 97.6% and 98.5%, respectively. The results demonstrate that the target detection algorithm can better meet the detection requirements of real picking tomatoes under the complex situations of fruit superposition, background interference, different densities, and foliage occlusion.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Aiming at the problem of low degree of automation in the separation of grape disease leaf identification and fruit identification in the cultivation process of grapes, we design a detection program for simultaneous identification of grape disease leaf and fruit. By collecting the disease data of grape leaves and fruits individually or mixed, and transferring the collected pictures to the detection system, YOLOv5s algorithm will automatically recognize the diseases of fruits and leaves in the pictures and feed back the recognized pictures, the training set is 2870 pictures of 7 kinds of diseases. The test set is 711 photos of various diseases. According to the validation test of the test set, the model has an average accuracy of 95.6% for labeling the seven diseases of fruits and leaves respectively, which effectively improves the efficiency of disease recognition in the cultivation process.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Named Entity Recognition (NER) aims to locate and identify entities with specific meaning in text. The NER problem can usually be regarded as a type of sequence labeling problem. The key to solving this type of problem lies in determining the boundaries and categories of entities. However, due to the fuzziness of entity boundaries and limitations at the labeling level, most existing NER models introduce vocabulary information loss and The problem of entity boundary recognition error. To this end, a named entity recognition method based on the boundary-aware attention mechanism is proposed. By introducing pointer annotation to construct the boundary position vector, a sequence annotation layer that fuses the boundary position information is established to fully exploit the boundary characteristics of the entity. On the basis of integrating the boundary position vector, the lattice structure of the text information is converted into a planar structure composed of spans, a dynamic position encoding strategy is designed based on pointer annotation, and then the semantics of the label of the entity annotation and the entity boundary position are learned based on the generative adversarial network Similarity, and improve entity recognition performance by introducing information of entity boundary pointers in the weight calculation of the attention mechanism. Experimental results on weibo and Chinese-Literature-NER data sets show that the proposed method has obvious advantages in accuracy and F1 index compared with the baseline method, verifying the effectiveness of the attention mechanism based on boundary awareness in named entity recognition.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The composition distribution detection of processed tobaccos is one of the main difficulties in cigarette design and processing. The radio of the four kinds of tobacco influences the quality of cigarettes. Therefore, it is very important to detect tobacco shreds on the tobacco processing in order to ensure the cigarette consistency of formulation. Aiming at solving the problems of poor efficiency and low accuracy of traditional method and utilizing the advantages of deep leaning network in object detection, a method based on Yolov7 for tobacco shred detection is proposed. The collected images of four categorized tobacco shreds are manually marked, and model is trained through Yolov7 network. The mean average precision of tobacco shred detection in this work reached 0.986, and the tobacco shred detection speed was also vastly increased compared to traditional manual detection method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Residential historical buildings carry rich historical information and cultural connotations, as well as being an important part of China's architectural engineering. However, over time, cracks have developed in the wooden structures of these buildings, posing a threat to the structural integrity and long-term preservation of the buildings. Timely detection and repair of these cracks is therefore critical. In this study, we propose and apply an improved YOLOv8-n model for real-time detection of cracks in wooden members of residential historic buildings. First, we collected a large image dataset of cracks in wooden structure of residential historic buildings and labeled the cracks. Then, we improved the YOLOv8-n model by introducing the coordinate attention mechanism and the Ghost convolution module so that the model can pay more attention to the details and features of the crack region. Finally, we completed the training and testing of the YOLOv8-n model before and after the improvement on the constructed wood member crack dataset. The experimental results show that the improved YOLOv8-n model achieves significant performance improvement in the detection of cracks in wood members of residential historic buildings compared with the original YOLOv8-n model, with an F1 score of 94.7%, an average accuracy of 96.2%, and a detection speed of 89.3fps. In conclusion, compared with the traditional contact detection method, our proposed improved YOLOv8-n inspection model is a contactless nondestructive inspection method. The method provides fast and accurate detection of wood structure cracks and accurate crack localization information, providing strong support for the restoration and preservation of crack-damaged wood structures in residential historic buildings.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Garbage target detection plays a crucial role in environmental protection and urban management. With the continuous development of computer vision technology, deep learning methods have become one of the main tools for garbage target detection. YOLOv7 is an advanced object detection model known for its high speed and accuracy, but it still faces challenges in garbage target detection. In order to enhance the performance of YOLOv7[1] in garbage target detection, this study proposes an approach based on improved YOLOv7.Firstly, we introduce the GAMAttention[2] attention module to enhance the model's ability to detect small targets. Secondly, we incorporate the Decoupled Head[3] to improve the model's robustness to targets in complex backgrounds. Additionally, we optimize the model's loss function to further enhance detection accuracy. The proposed garbage target detection method based on improved YOLOv7 has made significant progress in improving detection performance, providing more effective tools and technical support for environmental protection and urban management. This approach is expected to be widely promoted in practical applications, addressing challenges and issues in the field of garbage target detection.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper, we propose a lightweight facial expression recognition algorithm based on joint model compression. Firstly, we propose a self-distillation method based on facial expression recognition, which not only considers the overall output of the model but also focuses on the prediction results of each facial expression category, thereby improving the performance and generalization ability of the facial expression recognition model. Secondly, in order to further compress the facial expression recognition model, we apply a channel pruning method based on transformed L1 regularization. This pruning method imposes non-convex constraints on the parameters of the facial expression recognition model through transformations of the parameters in the L1 norm penalty term, achieving more accurate sparse regularization and effectively preventing overfitting of the model. Lastly, the pruned network is quantified to low-bit to obtain the final lightweight facial expression recognition model. The proposed method achieves 73.25% on the FER2013 dataset, with a parameter compression rate of approximately 50%, which achieves an effective lightweight.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Convolutional neural networks(CNNs) based object detection methods are prone to be interfered with by background noise and cannot make full use of semantic information in the positive sample-choosing phase. To overcome these issues, first, we proposed an external module called “Semantic Information Attention Module (SIAM)”. The SIAM module can reduce the influence of features from non-target areas. Furthermore, we proposed a new positive sample selection strategy “Heat IOU”, and the “Heat IOU” standard will give candidate samples with more semantic information more weight in the positive sample selection phase. Note that SIAM and Heat IOU can be easily combined with existing convolutional backbones, which provide a flexible plug-and-play way to spur the object detection system to focus on the zones in the image with abundant semantic information in forward propagation and positive sample choosing phase. Experimental results demonstrate that our SIAM and Hot IOU successfully improve the performance of the object detection task. The code is available at: https://github.com/OZERO98/SIAM
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As one of Chinese famous classical works, Hongloumeng has been translated by different languages and accepted by most scholars in the world. At present, there are two famous translating versions which are broadly accepted since their translations are loyal and closely related to source text, and translators are married couple of Yang Hsien-yi, Gladys and British translator Hawks. In order to comprehensively study characteristics of these two translating versions, this paper applies corpus search software Antconc 3.2.1w to study translating characteristics of vocabulary, syntax and context so as to make comparative analysis and deeply understand their translating characteristics. Based on above analysis, this paper discovers that Yang Hsien-yi and Gladys prefer literal language and their translation is more loyal to source text but it is difficult for western readers to deeply understand the content. However, in terms of Hawks’ translation, it is widely accepted by western readers because of its colloquial language but the content is less loyal to source text than Yang Hsien-yi and Gladys’ translation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In response to the problems that landslides occur in remote locations, are difficult to monitor and landslide target detection is not easily deployed at the embedded end, a landslide detection device based on the Jetson nano edge device is developed. The technology detects information of landslide feature objects through YOLOV5m technology, and is able to output landslide area information as well as geographic location information while accurately detecting landslide targets. A dataset of 25,000 landslide images is used to build the data set, and the training and validation sets are divided 9:1. A deep learning network is used to extract landslide features and build a landslide target detection model. After that, the train.py file is placed on the cloud server for training, and the best.pt file is migrated to Jetson Nano and tested on the embedded platform. The experimental results show that the average running time of single frame of YOLOV5mmodel in the embedded device is 100ms, and the detection accuracy can be maintained above 80%, which can achieve accurate detection and information acquisition of landslide target on Jetson Nano device, and lay the foundation for the development of edge device module for landslide detection later.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The advancement of deep learning technology has greatly improved object detection in a variety of settings. The detection accuracy of multiple ship targets under a complicated background is still poor in the field of ships. In this paper, a SHIP-YOLOV5S model for multi-scale ship target detection and recognition is proposed by improving the YOLOv5 model. Especially the introduction of the Swin Transformer module, enabling the model to capture richer context and global information, and improve the detection performance of multi-scale ship targets. SimAM is simultaneously integrated into the model to enhance the detection accuracy of the model by highlighting the more important feature information and suppressing the less important feature information. So as to improve the detection accuracy of the model. The improved Ship-YOLOv5s model was compared and validated on the collection-produced ship dataset ShipData. The results show that the improved Ship-YOLOv5s model has a detection accuracy of 82.7%, a recall of 77.9%, and a mAP of 80.7% for ship targets, which are 2.1%, 1.0%, and 0.5%, respectively, compared to the YOLOv5 model. It shows that the improved model has more excellent ship target recognition and detection performance, which lays the foundation for subsequent applications in the fields of harbor management, navigation safety, and maritime search and rescue.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Because the detection of malicious files is not yet popular, there are often various "backdoors" for files on most forums. For Internet users who publish other people's privacy, bad comments or pictures in cyberspace, and upload and download malicious files in various forums and communities, this paper proposes that through the use of naive Bayesian algorithm and AC computer algorithm, we can quickly and accurately identify malicious content in text, pictures, and software. Through the secondary development of API, we can monitor hundreds of millions of data in real-time, It greatly alleviates the problem of detecting malicious files in the current environment.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Manufacturing of large steel bar reinforcement semi-finished parts becomes more popular as assembly construction demands of bridge tower grow. However, the popularity of industrial device like welding robot that is usually welcomed in mass production is hindered by the traditional usage and cost it brings. In this paper, we proposed a newly welding method for spatial irregularly arranged steel bar intersections by using vision-guided multi-axis motion robot, wherein composition of robot system and welding methods are introduced. Finally, the practicability and effectiveness are verified through a typical bridge tower engineering application.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The development of modern agriculture is inseparable from the promotion of science and technology, and agricultural robot has become one of the important directions of agricultural technology innovation. Intelligent harvesting robots can effectively reduce the labor force of manual harvesting and improve harvesting efficiency. This article aims to design an intelligent picking machine based on Raspberry Pie and Arduino, explore its effectiveness in agricultural production, and provide new, efficient, and low-cost technical means for agricultural production. The significance of this study lies in improving the level of agricultural production automation, promoting the transformation and upgrading of agricultural methods, and laying the foundation for achieving agricultural modernization, technological advancement, and sustainable development. This paper will introduce an intelligent picking robot scheme based on Raspberry Pie and Arduino control, design an intelligent picking machine, design an intelligent picking machine, which can automatically detect and identify fruits, and pick fruits through mechanical devices, including hardware design, control logic and application scenarios, hoping to provide some reference value for the development and application of agricultural robot.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to tackle the challenge of inadequate comprehension by cooperative robots during assembly tasks, we present an innovative algorithm that leverages Graph Convolutional Neural Networks. The primary objective of this algorithm is to extract essential information relevant to assembly tasks from assembly process cards. To achieve this, we also implement an Optical Character Recognition algorithm, which effectively identifies textual information contained within the cards cells. By incorporating the Graph Convolutional Neural Network, we can then capture the underlying topological relationships among the key information elements. This comprehensive approach not only facilitates the prediction of key cell locations but also help robots to autonomously comprehend assembly tasks through an automated data mining process.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The hot-rolled steel plates from the same furnace in the steel plant need to be cut as samples for quality inspection according to national regulations. Currently, the handling of samples is still manual, and manual handling is time-consuming and labor-intensive. Therefore, this article designs a sample handling robot system. The robot workstation system is constructed through hardware and software design. The system mainly consists of ABB six axis robots, end effectors, conveyor chains, template temporary storage racks, template collection boxes, etc. Use SolidWorks to create end effector models and import them into RobotStudio to establish workstations. Utilizing RobotStudio software to achieve path simulation and optimization of robots, solving the problems of collision and mold piercing.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Rigid robots have made significant progress in design, manufacturing, and control. In contrast, soft robots, based on flexible materials, exhibit robust environmental adaptability, excellent biological interaction capability, and outstanding safety advantages. In the field of soft robotics, magnetic actuation technology demonstrates outstanding overall performance, featuring fast response, programmability, remote controllability, and powerful driving force. Currently, magnetic control in soft robots primarily focuses on the study of magnetic elastomers. Despite possessing a certain loadbearing capacity, there is room for improvement in deformation capabilities, i.e., environmental adaptability. Magnetorheological fluid (MRF), as a novel smart material, exhibits Newtonian fluid characteristics in the absence of magnetic field and Bingham body characteristics when subjected to a magnetic field. Soft robots employing this unique property of MRF hold potential applications in biomedical and flexible interaction fields. The study presents the design of an innovative soft robotic system centered around magnetorheological fluid as the principal smart material. A control system has been engineered to align with the specific operational environments of the robot, and a tangible experimental platform has been established to facilitate testing and validation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the continuous development of technology, mechatronics integration technology plays a more important role in intelligent logistics and warehouse management, promoting the continuous innovation and development of the logistics industry. This article explores the role of mechatronics integration technology in intelligent logistics and warehouse management. The integrated application of mechatronics technology has applied sensor technology and automation control technology in automated warehousing equipment, and information processing technology has been applied in logistics transportation processes. Core technologies such as sensor technology and automation control technology have been applied in warehousing management, effectively improving logistics efficiency and reducing operating costs. And analyze and illustrate the specific application of mechatronics technology in enterprises and the significant effects it brings based on practical cases. This article provides an in-depth exposition and exploration of the development and application of mechatronics technology in intelligent logistics and warehousing management. The application of mechatronics technology has also encountered challenges in technology, talent, and management. Therefore, corresponding countermeasures and suggestions are proposed to further develop mechatronics technology in intelligent logistics and warehouse management. This is an important issue in current application. Mechatronics integration technology will continue to make new breakthroughs in wider applications and higher levels of development, driven by market demand and policy environment in future technological innovation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The national secret algorithm is used to transform Hyperledger Fabric platform. In this paper, the existing standard protocol Hyperledger Fabric1.4 based on federated chain is replaced by encryption algorithms such as SM2, SM3 and SM4. Then, the blockchain network, Fabric-CA, and Fabric-SDK are reconstructed respectively. This project designs a security mechanism suitable for encrypted blockchain transaction information. SM4 symmetric cryptography is used to encrypt the user's private information, so that only traders can obtain private information. The validity period of bidding documents is tested by SM2 public key cryptography. Finally, this paper will complete the improvement of the encryption algorithm on Hyperledger Fabric platform. The practical application and performance verification of the designed super ledger Fabric platform are carried out. Experiments show that the algorithm proposed in this paper has successfully completed the reconstruction of state secrets on the super ledger Fabric software platform. Its operation efficiency and performance have reached the practical requirements.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper a transmission line inspection system based on multi-rotor unmanned aerial vehicle is designed. Using the images captured by the camera on the Linux platform the data is sent from the bridge to the base on the surface. The main platform of this system is a tablet computer that supports both Windows and Linux. An embedded system with Linux as the core is designed which uses digital HD camera to collect images and then sends images to the surface by using 2.4GHz network bridge to achieve wireless transmission of high-definition images. The overall hardware architecture is composed of multi-rotor UAV components core control microcontroller high-definition video module wireless bridge and land base station. The modular layered isolation method is adopted to ensure the efficiency and accuracy of the system. The experimental results show that this method has strong real-time performance and high precision. It can transmit high-definition image at a small cost and help the fast detection of power grid.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the increase in the number of motor vehicles, the number of stolen cars has also been increasing year by year, and vehicle networking anti-theft has become one of the hot topics and research topics. This article analyzes the behavioral characteristics of stolen vehicles based on several common situations of car theft, and then proposes the overall requirements for anti-theft systems. A car intelligent anti-theft system combining multiple sensors and the Internet of Things is proposed by comparing and designing the performance of several key technical solutions according to the requirements. Secondly, the hardware system design was carried out using modular design principles, including sensor acquisition and transmission node design and vehicle controller design. On this basis, we designed a frame timeslot ALOHA algorithm based on time grouping to optimize information collection and improve system efficiency. After simulation testing, this solution can provide new ideas and technical paths for automotive anti-theft design.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Straw power generation can reduce pollutant emissions and alleviate the energy crisis. However, the development of straw power generation supply chain is affected by various risk factors. Therefore, this study clarified the internal structure and feedback mechanisms of the risk factor system, and constructed the system dynamics model of the straw power generation supply chain to explore the impact of risk factors on the supply chain in terms of cost. Subsequently, simulations were carried out using the Vensim PLE software. The results indicate that environmental risk has the greatest impact on the total cost of straw power generation supply chain, followed by market risk and information risk, with technical risk having the least impact. Finally, Risk control strategies were proposed for the most influential risks, and the system dynamics simulation was carried out. The results show that the effect of information risk control strategies are the most effective, followed by market risk and environmental risk. This study provides a theoretical foundation for the risk control in straw power generation supply chain and contributes to the sustainable development of the straw power generation industry.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The engine equipped by hovercraft is divided into propulsion gas turbine and hovering gas turbine. Due to its large variety and quantity of propulsion devices, the air intake system of its power plant has the characteristics of multi-branches and complex structure. The flow field and pressure loss of multi-branch air intake system of hovercraft are numerically simulated in this paper. Under the same flow pressure, the inlet flow rate of the hovering gas turbine is the largest, followed by the flow rate of the gas turbine on both sides, and the flow rate of the gas turbine in the middle is the smallest. During high-speed hovering navigation, all gas turbines are in a high power state, and the overall flow rate of the intake system is high.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As an important place for displaying natural history specimens, researching and publicizing scientific knowledge, nature museums face the problems of single exhibition mode, poor science popularization effect, and insufficient audience participation. The article discusses the application of XR technology in natural museums and its advantages, which brings innovation to traditional museum exhibitions by breaking the physical space limitations and providing immersive experiences. This article takes the meteorite ore exhibition as the theme of the museum as an example, based on XR technology to carry out the research on the museum interaction system. It provides a new immersive exploration experience for visitors. In this experience, visitors are no longer passive receivers, but active explorers, who in the process of participation, not only can get fun, but also in-depth learning of natural science knowledge, which effectively enhances the educational and attractive nature of the exhibition.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This article mainly studies the energy security and diversified value evaluation system under the coupling of multiple factors such as economy, energy, and environment. It not only helps governments and departments at all levels to timely grasp the dynamic level of energy security, assist in the formulation of energy development policies, but also helps to clarify the interests of new equipment operators such as power sources, power grids, power users, energy storage, and energy service providers. Firstly, the definitions of energy security and diversified values by major domestic and foreign organizations were analyzed, and the relevant research theories on energy security and diversified values were clarified. The specific connotations of energy security and diversified values were clarified. Framework for energy security and diversified value system was proposed, and an evaluation index system for energy security and diversified values was constructed. Key measures to enhance Chongqing's energy security and value were proposed, and a radar chart based on the economic, safety, and environmental contradiction triangle was presented. The contradiction triangle will accompany the development process of the new power system for a long time and will continuously coordinate and handle relationship among safety, economy, and environmental protection in the future. This study indicates new power system needs to accelerate construction of energy storage, promote energy storage facilities to provide services to aspects of power generation, transmission and distribution, encourage integrated development of wind and solar energy storage stations, support the layout of grid side energy storage at key nodes, and support development of user side energy storage.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Compared with simulation technology, the prominent feature of digital twin technology is its complex interaction ability with the external information, which enables the digital twin function of the physical objects through complex interaction. Therefore, a complete digital twin system is formed, and the digital twin model can truly and effectively play its role. To achieve this, it is necessary to establish data interfaces that effectively connect with external information, physical field calculations, the visual displays, etc., to realize the system integration of digital twin models and verify their engineering feasibility. This article analyzes the functional requirements of the system integration, designs its functional framework, constructs functional modules of physical hardware layer, twin model layer, and human-computer interaction layer, and realizes the integration of valve side bushing digital twin model system with data exchange capability, data transmission capability, and coordination between various functional modules. Among generated parameters, E-field strength inside 220kV reduced ratio bushing is controlled within 3kV/mm, and the composite insulator umbrella skirt is installed outside the core. Total length of the bushing is 2465mm, the length of the bushing tail is 810mm, and the length of the hollow composite insulator is 1250mm.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
According to wind power output curve, it can be seen that due to the significant influence of climate on wind power output, there is high degree of randomness and fluctuation in output, and there is no obvious characteristic pattern. From the historical wind power output curve of typical days during the peak season, it can be seen that in most years, wind power output is relatively high from 21:00 to 2:00 in the evening and relatively low from 14:00 to 17:00 in the afternoon. the wind power output shows the certain characteristic pattern: the cumulative wind power output is higher from21:00to2:00 in the evening, and relatively lower from 14:00 to 17:00 in the afternoon. The 1500MWdistributed photovoltaics were connected to the Chenjiaqiao Yuping area in the Chongqing load center area. Three permanent N-1 and N-2fault calculations were conducted on the 500KV AC line in Chongqing, and it was found that the power grid could maintain stable operation. After the fault, the voltage of each 500kV bus was restored to 0.9p.u. There was no disconnection of distributed power sources in the Chenjiaqiao Yuping area. Three permanent N-1 and N-2 fault calculations were conducted on each 220kV AC line within the Chenjiaqiao Yuping area power grid, and it was found that the power grid could maintain stable operation. After the fault, the voltage of each 220kV bus in the Chenjiaqiao Yuping area was restored to 0.9p.u., there was no disconnection of distributed power sources. The paper provides control modes, open-loop control and closed-loop PI control, which is used for simulation analysis of frequency regulation in wind farms.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The airport baggage handling system ranks the highest in the percentage of airport complaints every year, and the smoothness of its transportation affects the operation of the airport and the travel of passengers to a certain extent. For the main operating airports in China, the questionnaire survey, fieldwork method, and overall research method were used to analyze the baggage handling systems of 42 terminals in 37 airports, and field investigations were conducted on baggage handling systems at 5 of them. Through data analysis and comparison, the main operation status of the airport baggage handling system was understood, for which the factors affecting the smoothness of the baggage handling system were proposed, the causes of baggage jam were identified; and the design methods and improvement suggestion measures to solve the smoothness of baggage transportation were given from 4 perspectives: baggage acceptance, baggage transfer process, baggage transfer terminal, and system operation control. The research results provide important data references for improving the smoothness of baggage handling systems in China and also provide relevant guidance for the design, manufacturing, and installation of the baggage handling system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper presents a wearable human-machine co-robotic control system composed of a wearable bionic exoskeleton suit and a bionic robotic arm, integrated with various sensors including stretchable flexible sensors. The system aims to achieve collaborative control of the bionic robotic arm through the wearable exoskeleton suit, demonstrating a novel human-machine co-robotic technology. The real-time upper limb joint motion data of the operator is captured by multiple sensors arranged on the bionic exoskeleton suit. After signal processing, motion planning, and actuator control, the control of the bionic robotic arm is achieved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Due to the continuous implementation of the dual carbon strategy, the capacity of new energy grid-connected generation has been increasing year by year, while the capacity of energy storage equipment is also growing steadily. To enhance the operational proficiency and control expertise of dispatchers, this paper utilizes a multi-agent simulation model to construct a grid operation simulation model and develops a simulation training system software for the application of grid control strategies. By networking servers, network equipment, and end-user computers, the simulation software has been equipped with the necessary hardware infrastructure for operation. The developed grid control strategy simulation training system, which includes the integration of distributed new energy sources and energy storage devices, enables simulation of various scenarios in grid operation, providing a simulated practice platform for personnel to enhance their work capabilities.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper introduces the main control characteristics of full-cushion hovercraft, analyzes the control system of American LCAC hovercraft, introduces the control command mechanism and actuator of the hovercraft, analyzes the subsystems and main functions of the control system, and explains the control signal flow of the control system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Tolerance technology is the third-generation network security technology commonly used in the world. It is derived from the category of information survival and endogenous security technology. A scholar from Carnegie Mellon University gave this survival technology a definition: the so-called "invasion survival technology." It is the ability of the system to perform its own tasks within a limited time when external attacks, failures and accidents have occurred. It assumes that we cannot detect intrusions to the system completely and correctly. When external intrusions or failures occur suddenly, we can use tolerance technology to solve the problem of survival of the system to ensure the confidentiality, integrity, and integrity of the information system. reliability and non-repudiation. In the current Internet security incidents, countless experiences and lessons tell us that it is not enough to rely on blocking and defensing for network security. Based on the existing network defense-in-depth architecture and the common behavior and parameter baselines of current emergency response systems, this paper designs an intrusion-tolerant system based on the behavioral baseline and network-in-depth architecture, which realizes intrusion tolerance and malicious behavior blocking under high confrontation intensity. To adapt to the maximum recovery and necessary tolerance under the attack technology of international APT organizations, and to protect the integrity and availability of the system to the greatest extent.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to protect the safety of LNG carrier, onboard equipment, and personnel, it is necessary to monitor the liquid level, temperature, and pressure of LNG carrier. In case of emergency, the carrier side and shore side equipment should be turned off and alarmed. Therefore, this article designs a LNG carrier safety monitoring system. Firstly, the overall architecture of the system was designed, consisting of a management layer, a control layer, and a device layer; subsequently, its control functions were designed, including ESD emergency shutdown trigger/reset, cargo tank protection, sea/port mode switching, ESD testing control, ESD shielding control, and override control; then, based on the algorithm configuration software and graphic editing configuration software of Hollysys MACS V6.5 system, the control program was written and the human-machine interface was drawn; secondly, a functional test platform is built to test the functions of the system. The results show that the system can control each state of the LNG ship, monitor each parameter in real time, and perform emergency shutdown in emergency situations. Finally, due to the overload operation of the output equipment for a long time, in order to prevent the failure of the output equipment and lead to major accidents, this paper proposes a fault prediction based on BP neural network, so as to achieve accurate fault prediction and ensure the safe operation of LNG ships.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to solve the problems such as high maintenance cost and difficult spare parts optimization of the complex electronic information system, considering advantages of the generalized stochastic Petri Net (GSPN) in characterization and analysis of dynamic change process of complex system, the process and method of spare parts optimization based on Monte Carlo simulation are studied to achieve the best balance between the rate of mission success and the cost of maintenance.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Currently, VR technology is constantly developing and widely applied in the field of medical education. Many experiments in medical courses have evolved from simple animated demonstrations to virtual simulation experiments. The current lack of intuitive clarity and ease of operation in teaching human anatomy experiments is a common problem. The 3D anatomical structure teaching model has the advantages of three-dimensional and highly simulated teaching, which can effectively reproduce the three-dimensional structure and three-dimensional entity information of tissues and organs. This article introduces the basic principles of VR technology and its application in anatomy experimental teaching. By comparing the traditional teaching methods and VR integrated teaching methods of 60 students in the anatomy course, the advantages of applying VR anatomical structure teaching mode to human anatomy experimental teaching were analyzed. As a result, the combination of VR anatomical structure teaching and traditional experimental entity specimen teaching solved many difficulties in the teaching process of this specialized course, and gave full play to their respective advantages, effectively overcame the shortcomings of traditional experimental teaching and VR anatomical structure teaching model, and greatly mobilized students' subjective initiative. Make it easier for students to master the complex knowledge of human anatomy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the continuous development of computer network technology, artificial intelligence stands out. Although its technology is not mature enough, artificial intelligence is also a kind of human intelligence technology presented through ordinary computer programs. Artificial intelligence technology can not only help people process computer data, but also gradually turn services into humanized and intelligent ones, bringing convenience to people's life and work. Therefore, this paper analyzes the definition and characteristics of AI, and discusses the application of AI in computer network technology in the era of big data.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To address the issues affecting the detection effectiveness of carton label images in cigarette carton logistics, such as dense and overlapping targets and significant disparities in shape and aspect ratio, this paper proposes an enhanced QR code target detection algorithm based on YOLOv3. By integrating deformable convolution for feature extraction, adopting the Soft-nms algorithm for more flexible non-maximum suppression, and incorporating the focal loss function, the modified YOLOv3 algorithm demonstrates improved performance. The algorithm's efficacy is validated through comparative experiments on a specific dataset for cigarette carton labels, showing a 3.37% increase in F1-score sample model identification rate and a 2.2% improvement in target detection accuracy. These enhancements offer a targeted and effective solution for increasing the accuracy of cigarette carton label recognition.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper proposes an intelligent recommendation method for electronic commerce. This paper aims to study the problem that the MAE value in the existing e-commerce recommendation algorithm is too large, which affects the prediction accuracy of the system. In terms of hardware, PC104 industrial computer is used as the processing CPU. DSP and memory chip MPC565 are used as processing center. The clock signal, reset circuit and power supply circuit are designed by means of signal fiber compensation. The software calculates the personalized characteristics of e-commerce according to its characteristics, and then uses the association rules algorithm to process the personalized characteristics data. Use Java tools to convert it into code, use different functions of the code to achieve intelligent recommendation and then complete the design of the software part. The experiment proves that the e-commerce intelligent recommendation model constructed by using individual characteristics has less sample set and higher prediction accuracy and practicability than the conventional e-commerce intelligent recommendation method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper proposes a new DEP model based on dynamic network. The method adopts the distribution search based on the natural classification level of products and divides them into tree structures. This index field is extensible and is the most basic and important link in the DEP architecture. Some key functions such as security management and reliability evaluation are also completed. The core functions of the system such as security control and trust evaluation are also realized by this layer. At the same time, with the assistance of third parties, complete the entire transaction process. This method adopts the text comparison method based on fuzzy inference, when the user searches the product, it can not only find the exact product, but also give the product close to the demand. This will greatly improve the transaction opportunities of enterprises, tap potential customers, and enhance personalized services. Experiments show that DEP itself has good expandability, can effectively reduce the query route and communication overhead, and can achieve high optimization efficiency.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
National traditional culture and traditional technology are the products of historical precipitation and indispensable precious resources. The establishment of national cultural database is of great significance for the protection and inheritance of cultural heritage. Aiming at the specific functional requirements of traditional cultural digital protection system for image segmentation and retrieval, this paper focuses on the common algorithms of image segmentation and feature matching. This paper analyzes and compares the applicability and processing effect of each algorithm in ethnic patterns. EGBIS image segmentation algorithm and shape context feature matching algorithm are used to achieve high-precision image processing. This paper designs and implements a national culture retrieval system that meets the application needs. Combined with the image characteristics of ethnic patterns and the application requirements of the system, this paper defines the concepts of source images, primitives and hyperprimitives, and establishes a relational database based on them. This study has laid a certain theoretical foundation for the study of digital inheritance of national culture.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The protection mode of digitalization effectively transforms various national traditional cultures into digital products, providing a way of thinking for the protection of national traditional cultures. This mode still lacks the mode of how to mine the data of national traditional culture and analyze the internal relationship structure after facing the massive data, which can guide the digital protection of national traditional culture under the complex social conditions in the future. This paper discusses the significance of digitalization for the protection of traditional national culture under the environment and ideology of big data development. The possibility of protecting traditional national culture based on big data was analyzed and discussed. This paper designs and establishes the digital perception analysis system of traditional culture. The system uses grey clustering algorithm to identify the characteristics of cultural products. The experimental results show that this method can improve the efficiency of traditional cultural product classification. This method has certain reference value for the research of digital protection system of traditional cultural industry.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In view of the problem that low-speed running rolling bearing fault characteristics are prone to noise annihilation, a hierarchical adaptive threshold denoising method based on (TQWT) is proposed, and the method is combined with envelope spectrum analysis for fault analysis and diagnosis of low-speed bearings. First, the collected bearing vibration signal is TQWT decomposed to obtain the decomposed wavelet coefficient; then construct the hierarchical adaptive threshold function using Sigmoid function to threshold the high frequency coefficient of TQWT; finally, combined with the high frequency wavelet coefficient and low frequency wavelet coefficient to reconstruct the signal to obtain the denoising bearing vibration signal. Through the experimental platform collected in low speed bearing fault vibration signal experimental analysis, experimental results show that the method has achieved good denoising effect, in reducing noise interference at the same time effectively retain the bearing fault feature information, denoising the signal envelope spectrum can clearly present fault spectrum characteristics, can observe the fault characteristics of multiple frequency peak, and peak interference near the few, effectively improve the low speed running bearing fault diagnosis rate;
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper, according to the living habits of domestic ordinary families, room layout and other specific circumstances, put forward four basic points of intelligent family pension plan to help families to achieve intelligent home care. First of all, the system not only provides rich ZigBee intelligent devices for the elderly to meet their needs in daily life, but also integrates emergency keys and Kinect devices into the system, so that when the elderly have accidents, they can send alarm information to the outside world to ensure their safety. Secondly, an intelligent pension system based on health data mining is proposed. Firstly, the heartbeat signal and respiratory signal of the human body are collected by the non-contact vital sign sensor, and the weight signal of the human body is collected by the pressure sensor. The collected physiological parameters of the human body are sent to the remote server based on the wireless module. Then, by comparing the efficiency of classification and clustering methods in heart rate and respiration rate data mining, a health data mining algorithm is implemented based on K-means clustering. Considering the reliability and stability of the system, the upstream and downstream data between the intelligent device and the cloud server during the operation of the system are all governed by the unified communication protocol rules. Finally, users can interact with the system through Web pages, smart speakers and mobile terminals, so as to give full play to the diversified manipulation characteristics and increase the elderly home care intelligent life experience.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Ornithopters, characterized by their high energy efficiency, minimal target visibility, enhanced flexibility, and compact design, represent a critical trajectory in the evolution toward more intelligent, coordinated, biomimetic, and concealed aircraft systems. This review begins with an examination of the current development status of typical aircraft in regions such as the USA, Russia, and Europe. It then delineates the principal technical characteristics of ornithopters, focusing on energy consumption, target detectability, agility, and system control. In light of the rapid evolution in combat approaches, the discussion extends to pivotal aspects of ornithopter utilization in practice, which primarily include situational awareness, command response, decision-making, and posture control. The paper concludes by projecting future trends in ornithopter development, emphasizing advancements in intelligent decision-making, endurance enhancement, system stability, and collaborative operation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
By design state, pillars under helicopter platform could be vibrated by vortex. Circular structure is firstly simulated to be checked with DNV-RP-C205 to validate the method. And then methods of decreasing frequency of vortex induced vibration and increase inherent frequency of pillars are to be studied to get how they are improved. The methods of decrease vibration of pillars induced by vortex to be plausibly in engineer is studied.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Retinal vessel morphological changes correlate closely with ocular and cardiovascular diseases, aiding in their evaluation, screening, and diagnosis. In order to extracted the information from retinal vessels, a retinal vessel segmentation method based on adaptive fuzzy local information C-means clustering is proposed. After using contrast-limited adaptive histogram equalization for image enhancement, the features of retinal vessels are extracted by B-COSFIRE filters. Finally, the segmentation of fundus vessels is achieved by adaptive fuzzy local information C-means clustering algorithm. On DRIVE and STARE datasets, the method proposed achieve the average sensitivity of 0.6841 and 0.7585, the average accuracy of 0.9458 and 0.9463, respectively. The proposed method provides good segmentation performance, and its segmented vascular network has good integrity and continuity. Compared with the feature space FCM method, our method has significantly improved the sensitivity of detecting retinal blood vessels.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Energy security, environmental issues, and economic issues are the driving factors for studying the reduction of energy consumption and greenhouse gas emissions. To solve the problems of urban centralized heating systems, this article designs and implements an intelligent monitoring system based on the current mainstream Internet of Things communication technology NB IoT. The monitoring system can maintain device information in the cloud platform and monitor energy consumption in real-time. Management personnel can issue device control instructions through the monitoring system. On this basis, we designed an RBF neural network prediction model to predict the water supply temperature of the secondary network, achieving closed-loop control of the water supply temperature of the secondary network. Through simulation experiments on the established predictive model and control algorithm, the results show that the adopted control algorithm can effectively control the system model due to changes in amplification factor and time constant compared to conventional PID control, verifying the feasibility and effectiveness of the adopted control method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to apply to a wider range of scenarios, multicast packets are required from the infrastructure. The gateway nodes in the network are more complex than regular nodes, both in terms of joining multicast groups and forwarding packets between networks. The current detection of duplicate packets generally follows the same principle and varies a bit more in terms of resource consumption. In this paper, we propose an algorithm for data duplicate packet detection using a data structure, balanced binary tree, and compare it with conventional routing algorithms for detecting duplicate packets and analyze that the linear mapping relationship between priority and waiting time leads to the transmission of duplicate packets, which increases energy consumption and reduces the throughput of the network.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Crime prediction has been challenging, partly due to the difficulty in selecting predictive variables. Previous studies theoretically and empirically confirmed the relationship between many variables and crime. This study identified a total of 19 theory-driven variables and applied machine learning tree models for crime prediction. To refine the variable pool, we extracted the subsets of the variables from the variable pool by using feature selection processes, which was based on Gain and Permutated Variable Importance Measure (PVIM) methods. We finally found historical cases, percent nonlocal population, and ambient population (25-44) variables ranked high. Thus, our results call more attentions on these variables for crime prediction.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In automated production lines, flexible workshop scheduling is essential for optimizing resource allocation and ensuring efficient production. This study develops an efficient scheduling strategy to enhance the distribution of processing and handling tasks. Addressing the issues of local optima and slow convergence in traditional Genetic search algorithms, we introduce an improved Genetic algorithm (GA) that dynamically adjusts the crossover and mutation rates. Additionally, it sorts individuals based on their fitness during the crossover process, significantly improving the quality and efficiency of solutions. Simulation results demonstrate that the improved algorithm significantly outperforms traditional methods in task completion, reducing the time from 198 to 146 unit times, a decrease of 26%. Furthermore, the algorithm exhibits faster convergence and superior optimization capabilities.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the rapid advancement of aerospace electronics, the traditional single-satellite application paradigm is transitioning towards collaborative networking involving multiple satellites. Furthermore, the continuous improvement in on-board computing capabilities raises concerns in the aerospace sector regarding the effective utilization of satellite-based processing resources for on-orbit edge computing. This study draws inspiration from mobile edge computing design principles and proposes a tailored computing payload management software design for satellite edge computing. By integrating traditional on-orbit Payload Management with mobile edge computing technology, the design facilitates task scheduling across heterogeneous computing resources within satellite constellations. Moreover, validation scenarios are devised to demonstrate the effectiveness of edge computing enabled by the computing payload management software in improving the timeliness of satellite services. Validation results indicate a 31.9% reduction in satellite service latency under the proposed edge computing scenario leveraging the computing payload management software.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Deep learning-based helmet detection is a common aspect of industrial safety production. The existing hard hat data set covers a large number of open-air scenes, and lacks low-light scenes such as underground mines and tunnel projects. Therefore, it cannot meet the detection requirements under low light. To develop a dataset better suited for this purpose, we first collected multiple images in harsh light to supplement the data. Then, we analyzed and compared the network structure and model features of the current phase-based helmet detectors, including the YOLO series, SSD, RetinaNet, CenterNet, and EfficientDet. Experimental results show that the YOLO series algorithm is the most accurate model, and its detection model YOLOv7 also shows the best performance, accuracy score is 94.57%, a recall rate is 94.83%, and mean average precision (mAP) is 97.7%. The model has good applicability and robustness for helmet detection in different scenarios.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As the carrier of information assets, the importance of accounting information systems is increasing with the development of informatization. Meanwhile, information systems are facing more security threats. How to effectively ensure the security of information systems and prevent information assets from being stolen or damaged has become an inevitable challenge. Therefore, conducting research on the security issues of information systems is of great significance. In response to the data security risks in accounting information systems, this article first establishes a risk indicator system and proposes an information system risk assessment model based on adaptive fuzzy neural networks, which scores the risk indicator attributes during the assessment process. In addition, we propose a fine-grained weight adaptive adjustment algorithm in the model, which provides a quantitative method for the risk indicators of accounting information systems. The feasibility and applicability of the model have been verified through simulation testing.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Link flood attacks (LFA) represent a significant cybersecurity threat due to their ability to cause extensive network disruptions. Traditional traffic scrubbing methods, relying on static rules, often struggle to counter dynamic LFAs effectively. This paper introduces the dynamic traffic scrubbing (DTS) method, leveraging programmable data planes. DTS integrates moving target defense principles into traffic scrubbing, dynamically deploying and adjusting scrubbing rules to thwart attackers. Game theory is employed to analyze the distributed denial of service attack and defense strategies. Experimental results demonstrate that DTS effectively mitigates dynamic LFAs with minimal overhead, showcasing its robustness in defending against sophisticated cyber threats.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Vehicular ad-hoc networks (VANETs) are essential in mobile ad-hoc network applications, utilizing vehicle location data. However, safeguarding the privacy and security of this data is challenging, especially with untrusted roadside units. To address this, we propose a novel differential privacy-based location data publishing mechanism. It consists of three key components: dynamic group leader selection, data aggregation, and privacy-preserving data publishing. The group leader selection algorithm is based on directional thresholds, hop cycles, and lifespans, assuming semi-trusted vehicles. Additionally, the differential privacy-based algorithm uses map grid segmentation, privacy budget recovery, and a filtering mechanism to ensure location privacy. Comparative experiments show that our scheme outperforms existing models in total hops, group leader lifetime, and mean absolute error.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Microservice architecture, renowned for its flexible scalability and loose coupling, has seen widespread adoption. However, this architecture has also become a prime target for network attackers, with low-rate denial of service (LDoS) attacks posing significant threats to microservice systems. Due to their covert nature, traditional passive defense methods often fall short in countering LDoS attacks. To address this challenge, this paper introduces a microservice-oriented quick shuffling method (QSM), based on the moving target defense (MTD) principle. The QSM framework leverages the dynamic and flexible properties of microservices to implement a rapid shuffling mechanism, ensuring the protection of innocent clients. Experimental results demonstrate the effectiveness of QSM in robustly defending microservice systems against LDoS attacks.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Intelligent Transportation Systems (ITS), particularly vehicular ad-hoc networks, play a crucial role in improving road safety and traffic efficiency. However, transmitting vehicle location and identity data poses significant privacy risks, potentially allowing attackers to exploit this information. In this study, we developed a user-centric privacy model based on information entropy, enabling vehicles to monitor real-time location privacy levels. We also proposed a decision algorithm for pseudonym changes, dynamically identifying a cooperative set of vehicles for pseudonym alteration. Comparative experiments using SUMO and NS-3 simulations show that our scheme enhances location privacy while reducing pseudonym change frequency. This research provides valuable theoretical and technical foundations for ITS.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In oilfield downhole overhaul operations, lead impression analysis is employed to detect casing damage and assess the internal condition of downhole equipment or pipelines. However, current lead impression analysis primarily relies on manual labor, resulting in inefficiencies and significant errors. This research suggests a deep learning-based approach for lead impression analysis in order to solve this problem, utilizing deep neural networks to predict lead impression types and thereby evaluate the internal state of downhole equipment or pipelines, ultimately aiding in the repair of downhole casings. Initially, a lead impression image dataset is constructed; subsequently, a lead impression recognition model, LeadMoldNet, based on convolutional neural networks, is designed. Finally, extensive experiments validate the effectiveness and performance of the LeadMoldNet model.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The urban metro rail transit system generates ground vibration during operation, and these vibrations will propagate through the ground and lead to vibration of nearby buildings. Therefore, the finite element numerical model is constructed with the example of subway tunnel passing through the surface building, under the influence of structural thickness ratio TR-BL and distance ratio DR-SD, the vibration response characteristics of surface building infrastructure to descending train dynamic load are analyzed. The results indicate that: the vibration response of the neighboring infrastructure structure is obviously positively correlated with TR-BL and DR-SD, in which the floor beam has a more obvious marginal effect with the increase of both, i.e., when TR-BL<0.7 and DR-SD<3, the vibration response can be effectively weakened; The distribution pattern of peak acceleration across different sections of the infrastructure framework exhibits a general trend of progressive attenuation moving from left to right, where the maximum values of response acceleration likewise undergo a gradual decline from left to right. Additionally, the vibration responses of floor beams exhibit a weakening trend, as compared to those of frame columns, as they extend towards the right. Compared with the frame columns, the ground beams are more significantly affected by the vibration of train operation; the changes of the structural thickness ratio and distance ratio do not affect the time-varying law of the structural vibration response, but only affect the response intensity value.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Security incident handling is an important task in the security protection of critical information infrastructure, and how to effectively implement specific work at each stage is directly related to the security of critical information infrastructure and even the security of the country. Based on the characteristics and security protection requirements of critical information infrastructure security incidents, this paper studies the work plan of critical information infrastructure security incident handling from the aspects of incident disposal preparation, incident monitoring and discovery, incident assessment and decision-making, incident reporting and disposal, and continuous improvement, so as to effectively guide the development of security incident handling work of critical information infrastructure operators.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In the impact simulation condition, output overload and short circuit condition of circuit starting and sudden load, the five-axis CNC machining materials have the problem of five-axis CNC machining material overpunching, which may lead to the five-axis CNC machining material damage or mistriggering, five-axis CNC machining protection. Common used open ring frequency modulation or duty cycle ratio flow limiting strategy, five axis CNC machining waveform control effect is poor. Therefore, for the unstable output voltage under the overload condition, we adopt the five-axis CNC machining material optimization design method based on the simulation model. First, the physical model of the simulation system is introduced, and the basic equation of the simulation model is described, and the motion of electrons and anions under the joint action of magnetic field is calculated. Secondly, the dynamic modeling process of five-axis CNC machining materials is analyzed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The development of network technology has laid a solid foundation for the network transmission of integrated media. The amount of information in integrated media is large, and it has the characteristics of real-time, interactivity, continuity, and synchronization. How to ensure and improve the service quality of media has become an urgent problem that integrated media technology needs to solve. In this paper, a real-time communication model of streaming media transmission based on edge computing is proposed. Based on the above principles, the core modules and network architecture of the model were derived, and the structure and transmission scheme of the model were studied and designed. This article will help ensure the real-time, continuity, and stability of integrated media communication by optimizing the system architecture design.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The application of multi-sensor (millimeter wave radar, LiDAR, binocular camera, etc.) fusion perception technology to the monitoring of cross sections of transmission lines is proposed in response to the difficulty of defect elimination and repair after accidents in the critical section of "three spans".The factor graph model is constructed based on the transmission line status information collected by multiple sensors, and the fusion effect of multiple types of signals is achieved by integrating multi-sensor information variables and adopting a global optimization method to improve credibility. Through simulation, the information fusion effects of weighted average method, multi Bayesian estimation method, and factor graph algorithm were compared. The simulation results show that the factor graph algorithm has higher credibility and better performance in fusing multiple types of signals. It can maintain the stability of the system and enhance its robustness, thereby achieving the goal of accurate identification and warning of transmission line crossing safety risks and improving the safety operation level of
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The present research addresses the challenge of optimizing control in the wastewater treatment process, presenting a refined control model rooted in the particle swarm optimization (PSO) algorithm. Through a comprehensive examination of the nonlinear and interdependent multivariable dynamics inherent in wastewater treatment systems, a tailored control model suitable for real-world operational conditions is developed. In the algorithm design phase, the PSO algorithm is employed to fine-tune control parameters, enhancing both the system's stability and overall treatment performance. The creation of the simulation framework integrates several crucial elements of the wastewater treatment process, thereby ensuring the model's robustness and practical applicability in diverse operational scenarios. The results from the simulation highlight that the proposed optimization control model offers considerable benefits in minimizing energy usage while boosting treatment effectiveness, demonstrating strong adaptability and global convergence characteristics. Subsequent data analysis corroborates the model's precision and practicality, delivering valuable technical insights for the intelligent optimization of wastewater treatment processes.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To address the challenges posed by the complexity and dynamic nature of monitoring within a mining environment, this research introduces an intelligent surveillance system grounded in wireless sensor networks (WSN). This system employs a network node localization algorithm to enable real-time tracking of critical factors in the mining context, including gas levels, temperature and humidity, geological shifts, and more. In order to enhance the accuracy of monitoring and bolster system reliability, an optimized network node placement strategy is developed in this study, incorporating an adaptive algorithm to dynamically fine-tune the data routing paths. During the system simulation phase, a virtual model of the mining environment is created to evaluate the wireless sensor network's monitoring performance across various scenarios. The simulation findings demonstrate that the proposed system offers notable advantages in terms of precision in data collection and energy efficiency of nodes. Detailed data analysis further substantiates the system's capability to enhance both the effectiveness and safety of environmental monitoring in mines. This research not only delivers technical backing for mine environment monitoring but also serves as a reference for the future deployment of wireless sensor networks in intricate settings.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This study aims to build a logistics system monitoring and optimization model based on intelligent algorithm to improve the efficiency and accuracy of logistics security management. The model focuses on item leakage events in the logistics process, and applies linear programming optimization algorithm to design risk assessment and emergency response strategies through comprehensive consideration of several factors such as item characteristics, leakage scale, environmental impact and emergency evacuation path. The model simulates different leakage scenarios, generates specific emergency evacuation plans, and optimizes the evacuation path and time. Simulation results show that the proposed model can significantly improve the efficiency of emergency response, reduce the possibility of secondary disasters, and ensure safety. Data analysis further verifies the robustness and applicability of the model in different scenarios. This study provides a new idea for intelligent logistics management, which not only has important theoretical significance, but also provides technical support for practical safety management.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A study was carried out on the damage to small unmanned structures caused by shock waves, and an equivalent three-dimensional model was established using the Coyote UAV as a typical small unmanned aircraft. A variety of explosion shock wave on the unmanned aircraft damage factors for simulation analysis and calculation, the shock wave on the unmanned aircraft structure of the form of damage, combat shock wave on the “coyote” unmanned aircraft damage numerical calculations for the relevant combat unit to combat the “coyote” The results of the numerical calculation of the damage to the Coyote UAV by the shock wave of the combat unit provide reference data for the relevant combat unit to strike the Coyote UAV, which lays a certain foundation for the subsequent research.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As multipurpose self-elevating platforms evolve towards deeper waters, larger scales, and multifunctionality, the modularization and generalization of platform design have become key challenges. This paper focuses on drilling and well-servicing operations in the Bohai Bay, considering the geological and environmental conditions of the target operational area. Systematic demonstrations were conducted for the configuration of the drilling and workover system, the jacking system, and the dynamic positioning (DP) system. A comparative analysis was performed on aspects such as calm sea water resistance, Cantilever, the Drill Floor, and Leg structural forms and strength analysis. Empirical formula methods were applied to reasonably forecast foundation bearing capacity and penetration depth. Additionally, a dynamic positioning capability analysis was conducted for the target Liftboat using hydrodynamic calculation methods. The study concludes by identifying key technologies and processes for the overall design of the next-generation multipurpose Liftboat with large operational air gap.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The service ecosystem has been created as a logic collection of different services, which is a typical Cyber-Physical System. With the rapidly developed service-oriented architecture (SOA) technology, it increasingly attracted more attention by providing users more flexible services. Clarifying the relations between services and the structure of the service ecosystem is of great importance to construct a productive service ecosystem. However, the current researches mostly study the optimize service discovery algorithms, service compositions and service recommendations, these works are mainly based on the analysis of the service relations from a component view and ignore the system structure constructed by the services and their relations. In this paper, we focus on the system structure of the Programmable-web service ecosystem and propose an model of the service ecosystem. We have analyzed the statistics of the actual service ecosystem based on the service data and find that the structure of the service system shows scale-free characteristics. The results show that the proposed model could capture the statistics of the actual network and be applicable to model the Programmable-web service ecosystem.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to meet the application requirements of reliable remote information interaction between acquisition equipment and ground control console, a transmission protocol with cyclic redundancy check and feedback retransmission functions is proposed on the FPGA platform. In the process of testing, due to the metastable phenomenon of FPGA transmission across the clock domain, the flag bit of the feedback instruction is not received correctly, resulting in incorrect data. The protocol is further optimized, in the embedded software protocol module of the receiver and the transmitter, the data is aligned using the response operation, making the data transmission link more stable. The test results show that the reliability of the optimized CRC feedback retransmission protocol with self-error detection function is improved, and it has a wide range of engineering application scenarios.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The development of air cooling technology in the northern region, where water resources are relatively scarce, has played a vital role in the development of the power industry. However, the circulating water system often causes corrosion of metal materials due to high concentration of impurity ions or pH, which affects the safe operation of indirect air-cooled units. In this paper, the authors firstly explain the reasons for the increase in pH of circulating water; then, analyze the possible hazards caused by this phenomenon; finally, put forward a treatment measure of CO2 into the circulating water to deal with this problem.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Warships have become an important combat force in the maritime battlefield environment, and how to carry out anti-warship missions based on existing warshipboard equipment has become the focus of the current combat application. This paper takes the foreign large-caliber naval gun ammunition as the research object, and analyses its destructive effect on warship targets. First of all, based on numerical simulation method to carry out naval gun munitions power modelling, and then analyze the effect of the fighting part of the role of the medium and the location of the explosion point on the warship target damage power. Simulation results show that: large-calibre naval artillery munitions have the ability to effectively damage the port side of the warship, and its destructive effect is mainly related to the explosive medium, explosion location and fuse action mechanism and other factors, so the proposed reasonable combat method in the field of anti-warship combat can achieve better performance, and effectively support the naval artillery weapon to carry out anti-warship tasks.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Underwater imaging frequently suffers from reduced clarity, loss of texture, and color inaccuracies, primarily caused by the interaction of light with water and the presence of various suspended particles. These challenges pose significant difficulties for underwater applications. Existing enhancement methods struggle to fully address the varying scattering intensities across different spatial locations and lack efficient integration of attention mechanisms with convolutional features at multiple scales. To tackle these issues, we present a new multi-branch reconstruction framework for underwater images, which blends attention-driven features. Our design incorporates a Swin Transformer module to adjust convolutional outputs through attention mechanisms at multiple scales and employs a dark channel prior for position encoding, reducing the impact of scattering variations in different spatial regions. Additionally, we leverage a large visual language model to boost the enhancement process by utilizing its image encoding and semantic insights to guide the network. Experimental evaluations show that our network excels in maintaining accurate color representation and preserving texture details.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Optical flow estimation in low-light scenes remains a challenging task due to the presence of significant imaging noise, which adversely affects the accuracy and robustness of the estimation. This paper proposes a novel approach for optical flow estimation that leverages Multi-Level Noise Modeling. Initially, the impact of imaging noise on optical flow estimation accuracy is analyzed, leading to the construction of a noisy training dataset specifically tailored for low-light scene optical flow estimation using Multi-Level Noise Modeling techniques. Subsequently, a noise-resistant optical flow estimation network is introduced, designed explicitly for low-light scenarios to improve precision in high-noise environments. The key innovation of this method lies in developing a parameterizable Multi-Level Noise Model and employing implicit feature-supervised training for optical flow estimation under standard lighting conditions, thereby avoiding the need for explicit low-light image enhancement. Experimental results demonstrate that the proposed method exhibits superior noise resistance and robustness across various noise levels, particularly under extreme conditions, where it surpasses existing mainstream methods in foreground-background differentiation and contour edge accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Given the shortcomings of the traditional Golden Jackal Optimization (GJO) algorithm, including limited accuracy and slow convergence speed in solving mobile robot path planning problems, an improved adaptive Golden Jackal Optimization (IAGJO) method is proposed. Firstly, a nonlinear adaptive energy strategy is presented to adjust the energy decay pattern, balancing global and local search. Secondly, an enhanced position update mechanism based on Cauchy and Gaussian mutation increases population diversity and guide the search based on optimal individuals, thereby facilitating efficient exploration of unknown regions and avoiding local optima. Finally, the IAGJO algorithm is applied to mobile robot path planning (MRPP), demonstrating that the IAGJO achieves shorter path lengths and higher search efficiency, exhibiting significant advantages over existing methods.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Icing on the wings of an aircraft increases drag and reduces lift, leading to a decrease in the aerodynamic performance and stability margin of UAV, which affects flight security. In order to solve the problem of icing state identification of UAV, this paper establishes a simulation model of icing on the wing of a certain type of UAV, analyses the impact of wing icing on the dynamic response of UAV, conducts simulation experiments with multiple state points and different degrees of icing, establishes a deep-learning neural network for icing state identification, and carries out the training and testing of three different strategies based on the simulation data. The results show that the icing identification method proposed in this paper has high accuracy in the prediction of icing degree and can provide a reference for the design of natural icing degree identification and prediction system for UAV.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper addresses the issue of matching two columns of data during account reconciliation by designing an appropriate combinatorial optimization algorithm. In the core computing region, we employ a backtracking search strategy to reduce both space and time complexity. The algorithm evaluates all possible combinations to identify the optimal match. Experiments demonstrate that the program exhibits robust anti-interference performance, making it suitable for real-world applications and thereby reducing the workload of reconciliation personnel.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Efficient algorithms are an important representation and fundamental prerequisite for the interactive capability of digital twin models. The internal influencing factors and evolution characteristics of the ultra-high pressure valve side casing are complex, resulting in a huge amount of calculation for the twin model. The existing literature lacks a fast calculation method to meet the requirements of fast calculation. The application of classical intrinsic orthogonal decomposition (POD), snapshot POD, and singular value decomposition (SVD) POD to the digital twin reduction of the valve side casing has the problem of low efficiency and cannot meet the fast calculation needs of complex parameters in the digital twin model of the valve side casing. This article proposes the spatiotemporal non-uniform POD reduction algorithm suitable for the digital twin model of ultra-high pressure valve side casing. It decomposes into different regions in space and uses non-uniform sampling in time, which can effectively balance the accuracy and timeliness of the calculation of the digital twin model of ultra-high pressure valve side bushing. When calculating model at 0 seconds and 2.5 seconds, the errors of the spatiotemporal heterogeneity POD reduction calculation results were 0.006% and 0.002%, respectively, indicating that the reduction calculation results can meet the accuracy requirements; The calculation time has been reduced from the 2min12s and 3min45s to 6.78s and 9.61s, both of which have decreased by about 95%, significantly improving the calculation efficiency.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Mathematics plays a crucial role in articulating the rationale behind artificial intelligence across various tiers. It furnishes a numerical structure for delineating cognition, sentiment, and advanced intellect. Core mathematical principle said in unraveling the underpinnings of correlation and causation within AI, facilitating innovative data exploration techniques characterized by multi-faceted, multi-central, and multi-dimensional strategies. Moreover, delving into brain-inspired computation and expansive simulations enriches comprehension of the driving mechanisms and logical connections within AI. Mathematics offers indispensable instruments for crafting novel AI frameworks and steering inventive breakthroughs in this domain.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Soybean protein is one of the important components of modern human diet. How to accurately and quickly determine the location of soybean protein is the main topic of research scholars. In order to understand the soybean genome, improve human dietary health behavior, explore effective methods of soybean protein subcellular localization prediction with different deletion degrees, and improve the level of soybean protein subcellular localization prediction, this paper aims to understand the current situation of machine learning algorithms and applications and, according to the research and application characteristics of protein localization technology in the new era, this paper mainly discusses the application advantages of four kinds of machine learning algorithms in predicting soybean protein localization, so as to provide technical support for soybean protein data prediction analysis.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Under the influence of complex theory, fractal, as a nonlinear geometry, reveals the appearance and operation law of the real world, and has the ability to accurately quantify the complexity of the system, and has been widely used in many disciplines in recent years. Especially in the field of architecture, which is closely related to geometry, fractal geometry, as a guiding form and rule system, can greatly expand the scope of architectural design and related theoretical research. Starting with self-similarity, the most essential morphological feature of fractal, through deconstruction and analysis of classical fractal geometric patterns, the three elements of fractal geometry are extracted, which are “initial graphics, generation logic and iteration times”, and applied to the generation and analysis strategies of architectural skin, space and structure. The research conclusion shows that self-similarity can be used as a generation method in the design of architectural skin, space and structure, with the advantages of simple and fast generation process and rich and diverse generation results. This construction method can break through the limitations of direct imitation of fractal patterns in previous architectural design, and enrich design diagrams from the perspective of generation strategy. It is found that the morphological mechanism, algorithmic logic, and fractal dimension in fractal geometry theory can have guiding significance for architectural design form, design thinking and design quantification. But fractal is also a science with mathematical rigor, and the combination with architectural design needs to be refined and optimized to a certain extent, to better expand the feasibility of its application in architecture.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Under the development trend of economic globalization, the employment problem of college students has been highly valued by all sectors of society. At present, the employment of college students has changed from elite employment to mass employment, and the market environment is facing new contradictions and problems, both the government and the society should put forward solutions from the mode of talent training. The training mode of talents in colleges and universities comes from the suggestions of subject experts, the needs of individual learning and the needs of social development, which are contradictory and unified and indispensable. Therefore, on the basis of understanding the employment of college students in the new era, this paper focuses on the platform of higher education personnel training in the era of big data, and then proposes effective measures for college students' employment according to the needs of social development.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Political party platforms lay the groundwork for campaigns and messaging in US presidential elections. We conducted an extensive data mining analysis of the syllabic content in every Democratic and Republican party platform from 1856 to 2012. Our techniques revealed marked differences in the syllabic patterns of platforms from the two major parties over different historical eras. Since the rise of radio broadcasting, significant distinguishing syllabic distributions have been identified between Democratic and Republican platforms through our statistical analysis. Voting data was then analyzed to determine if correlations exist between platform syllabic patterns uncovered through our data mining work and presidential election outcomes. Overall, this research applied natural language processing and statistical methods to surface important insights from the vast text data contained in decades of US political party platforms.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Large-scale expansion is observed in the development of offshore wind turbines, with increasing complexity in offshore convergence. The conventional prim algorithm and kruscal algorithm suffer from issues such as significant disparity in loop load, instability of the algorithm, and limited search efficiency. Consequently, they are not suitable for addressing dynamic programming problems. Nevertheless, a dynamic solution is imperative to effectively address the interdependence between system cost and topological structure in offshore wind power collection. Hence, based on the spatial location distribution characteristics of wind farms and considering the random fluctuation characteristics of wind power, this paper proposes a hybrid greedy genetic algorithm combining local search and hybrid repair optimization operators. Additionally, a topology optimization technique is developed with the objective of maximizing overall economic advantages. After the analysis of the real case in Jiangsu Province, it is evident that the proposed optimization model yields a topological structure cost that is 8.38% lower than both HVDC and FFTS schemes. This outcome holds significant technical and economic value, offering valuable insights for offshore wind farm planning and construction.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper, the use of a negative air outlet Angle super bending diffuser blade, the compressor rotor blade design, to achieve the given pressure ratio, flow, compressor ultra-high load, ultra-low speed design. In this paper, the design load coefficient of 1.7771, much higher than the conventional load coefficient 0.4. In the process of two-dimensional blade design, the automatic optimization design method is adopted. Computer numerical simulation results show that the designed compressor stage can reach the given design point flow rate and pressure ratio, and has a high stage efficiency, And because the rear section of the rotor blade channel convergence is not easy to produce flow separation, it has a large surge margin.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The design scheme of this teaching environment gives full play to the function of VRML, taking the classroom as the research object, focusing on the design of the switch of the induction door, the opening and closing of the laptop, the playback of computer video and the unfolding of the projector screen, etc., students can enter the virtual environment through the network, conveniently interact with the environment in three dimensions, deeply integrate VRML technology into teaching, build a situational teaching scene combining virtual and real, carry out contextualized, experiential and immersive teaching, and create a good educational environment for high-quality technical and skilled talents.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Smart classrooms, as an innovative form of teaching and learning, have received more and more attention from the education community for their powerful technical support and flexible teaching strategies. This paper introduces the connotation, key technologies, characteristics and applications of smart classrooms in teaching and learning, discusses the learning theories supported by smart classrooms, proposes a collaborative learning model based on smart classrooms, and evaluates the teaching efficacy of smart classrooms. In the smart classroom environment, the advantages of collaborative learning are more obvious, teachers can organize students to collaborate more conveniently, and students can communicate and share more conveniently.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper, aiming at the problems of low degree of digitalization and low measurement efficiency of the geometric morphology measurement of aerospace large-size curved structural parts, a measurement method based on area array scanning and visual measurement system is proposed. Through the establishment of digital collaborative measurement system, the coordinate unity of measurement field in multi-system cooperative measurement is resolved. The method is used to verify the surface structural parts in the field. The verification results show that the digital collaborative measurement method and system can realize the measurement of curved structural parts.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Federated learning is a distributed machine learning framework based on privacy protection. Heterogeneity is unavoidable in distributed data sets. When the data sets of different participants are not independent and identically distributed, the performance of the existing federated learning algorithm is inevitably affected. At the same time, the uneven distribution of participants' own data sets will also aggravate the deterioration of federated learning performance. Therefore, based on the inconsistent distribution of the training set and test set and uneven distribution of every participant training set labels, this paper proposes an SPFL algorithm to solve this problem and compares it with the existing FedAvg, FedProx, and FedDC under the same experimental conditions. Experimental results show that the accuracy of the SPFL model is significantly better than other algorithms when the participant data set is non-independent and uniformly distributed. The optimal accuracy of SPFL is at least 5% higher than the other three algorithms.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Aiming at the oscillation problem in the electric vehicle powertrain systems, an H∞ synthesis robust control strategy is proposed from the angle of acceleration control. The desired frequency characteristics of the system are obtained through H∞ loop-shaping, which ensures the high-frequency and low-frequency performance of the system, and the μ-synthesis ensures the robustness under the uncertainty of system parameters. By adding the coprime factor perturbation and the stability margin of the system into the H∞ loop-shaping, the oscillation existing in the system is suppressed. In the simulation analysis, the μ-analysis is used to verify the robustness and stability of the system, and the performance of the model uncertainty is intuitively displayed. The system has better closed-loop performance and robustness in the presence of parameter uncertainty and transmission shaft clearance. The designed acceleration controller also has a good control effect on the speed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the increasing number of network attacks, the network security of universities has been threatened unprecedentedly. This study explores a deep learning-based method for network intrusion detection in colleges and universities. Firstly, the development of network intrusion detection technology is reviewed, and the characteristics of university network environment are analyzed emphatically. Furthermore, the quality and representativeness of the data are ensured through sophisticated data preprocessing strategies. On this basis, an appropriate deep learning model is selected, and the configuration is optimized for the university network environment. After the system was deployed in real time, it was verified on multiple data sets, and the results showed that the model had high accuracy, precision and recall, showing strong generalization ability. The conclusion part confirms the effectiveness of deep learning technology in network intrusion detection in colleges and universities, and provides a new research direction and practical reference for the field of network security.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The distinctive attributes of blockchain technology, including decentralization, traceability, tamper resistance, transparency, security, and privacy, have paved the way for remarkable achievements across various domains, encompassing financial transactions, healthcare systems, agricultural intelligence, and energy upgrades. Notably, within the domain of blockchain-based supply chain finance, it has emerged as a potent solution to longstanding challenges, ranging from inaccurate risk assessment to information asymmetry and the mismatch between supply and demand service offerings. It has furnished the conventional supply chain ecosystem with a reliable and robust mechanism. This study addresses the multifaceted role and impactful applications of blockchain technology, elucidates the technical underpinnings of its implementation, and sheds light on the prevailing challenges that demand resolution within the sphere of supply chain finance. Furthermore, this paper also presents insights and prospects from an enterprise perspective, offering valuable guidance to future scholars venturing into this domain.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Rockfall, as a more common form of geological disaster in mountainous areas, is caused by geological structure, formation lithology, topography and other factors, and usually appears in natural slopes, artificially excavated road slope and other areas. According to the accumulated experience of research on rolling stone disaster movement of highway slope in recent years, in order to reduce the adverse impact of falling rocks on highway traffic safety, people will set up protective facilities around the slope to effectively intercept the stones falling down the hillside. From a practical point of view, the design must understand the falling speed, height and movement trajectory of the falling rock, and propose effective prevention and control measures according to the movement characteristics of the falling rock, so as to reduce the adverse effects caused by the rolling stone on the highway slope. Therefore, after understanding the characteristics of highway slope rockfall movement and the risk identification process of slope engineering, this paper constructs a highway rockfall risk assessment system based on big data, takes a certain highway slope rockfall movement as an example for simulation analysis, and finally puts forward effective prevention and control management measures.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Knowledge graph is a heterogeneous graph containing multi-relational information. Knowledge graph embedding has become one of the most effective methods in knowledge graph completion task. Using Graph Attention Network as a knowledge graph embedding model can effectively improve the quality of knowledge graph completion. At present, the knowledge graph embedding model based on Graph Attention Network uses the single-layer attention mechanism to calculate the importance of all local neighborhood information of a entity in the knowledge graph. However, we observe that this method of placing all neighborhood information in the same layer to calculate attention weights ignores the heterogeneity of the knowledge graph. Therefore, we proposes a hierarchical aggregation method for multi-relational neighborhood information. The local neighborhood information of a entity is divided into a relationship layer and an entity layer, and the multi-layer attention mechanism is used to calculate its contribution to the entity. This method can aggregate the local neighborhood information of entity in a finer granularity, so that entity can aggregate more effective information and ignore more invalid information, thereby improving the performance of the model. In addition, our proposed model performs superior compared to several state-of-art methods.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In view of the fact that the existing multi-view clustering algorithm does not dig deep into the local spatial information structure, resulting in inaccurate clustering results, this paper intends to make full use of the correlation and diversity between data to improve the clustering accuracy, and use the local similarity measure to mine the potential in the data. The structure is clustered and analyzed, and the atomic information is merged in a fuzzy manner to maintain the robustness of the algorithm. It is not sensitive to noise and will retain more view structure information. The algorithm (MvNIFC) in this paper verifies the effectiveness of the algorithm through experiments on various data sets, and compares several existing multi-view blur algorithms experimentally. The results show that the clustering accuracy and noise data processing are both improved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Aiming at the problems of slow detection speed and low accuracy in current human fall detection tasks, a fall detection algorithm based on Yolov7 was proposed. ODConv-ELAN module was constructed in Yolov7 backbone network to replace the original ELAN structure and enhance the ability of extracting target features. Secondly, the more advanced EIoU function is used as the new boundary frame loss function, which improves the convergence speed and efficiency of the prediction frame in the process of model training. Finally, CA attention mechanism is introduced into the output terminal of the network to improve the detection performance of human fall behavior. In addition, a fall detection data set in the campus environment was created. The accuracy P of the improved algorithm in this data set reached 94.34%, the recall rate R reached 92.34%, and the average accuracy mAP reached 94.65%, which realized the demand for more accurate human fall detection.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Based on the market demand, the company decides to independently develop a desktop cloud system for government, education and small and medium-sized enterprises. The system is designed for low-cost, easy-to-use, easy-to-manage and highperformance experience, aiming to solve the pain points of traditional PCs such as information security loopholes, high maintenance costs, inefficient management, etc., and to provide a cost-effective and easy-to-adopt desktop virtualization solution. The author has participated in the requirement study, technical solution design and R&D work, focusing on the system architecture design and part of the code development. Mainly through the optimization of the virtualization platform, management console, security mechanism and storage database and other modules, so that the desktop cloud system of security, reliability, manageability is better than similar products and easier to be accepted by customers and deployed to meet demands.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to better carry out railway construction in the western region and avoid safety problems caused by adverse geological disasters during tunnel excavation, advance geological forecast research and analysis is carried out. In view of the problem of insufficient accuracy of a single advanced geological forecasting method, the advantages, disadvantages and applicability of the current main advanced geological forecasting methods are compared. Propose a comprehensive advanced geological forecast implementation plan that combines multiple methods and complements multiple physical property parameters. Taking the advanced geological prediction project of a certain railway tunnel as an example. Based on the geological survey method, tunnel face geological sketching method, blasting source method and other forecasting methods, the on-site excavation face was monitored, the development of joint fissure water in front of the tunnel face was successfully predicted, and the surrounding rock grade and support grade were adjusted in a timely manner. Facts have proved that the combined forecasting of multiple methods has greatly improved the accuracy of advance forecasting of poor geology during team road construction.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to investigate the low-damage characteristics of electromechanical antiseismic support hangers under seismic conditions, vibration table tests on the antiseismic support hangers of non-structural components were conducted. The test subjects were seismic support hangers for pipelines designed and installed according to the "Code for Seismic Design of Buildings" GB 50011-2010. The structural system mainly consists of anchoring devices, reinforced suspension rods, antiseismic connection components, pipeline connection components, and antiseismic braces. To study the influence of different seismic levels on the dynamic response of the structure, three seismic waves with peak ground acceleration of 0.07g, 0.2g, and 0.4g were used to comprehensively evaluate the seismic performance of the antiseismic support hangers. The test results show that during the vibration table seismic simulation test, the natural frequency of the structure does not change significantly with the increase of seismic intensity, and there is no obvious damage to the specimen and the support hangers. The antiseismic support hangers have a good inhibitory effect on the displacement response of the pipeline system, with a maximum relative displacement of only 17.9mm for the water pipe. However, the acceleration amplification coefficients in the X and Y directions are higher than the specified values in the Chinese code, indicating that the antiseismic support hangers cannot reduce the acceleration response. Strain along the antiseismic support hangers does not show stress concentration, and under seismic action, the braces can effectively protect the vertical suspension rods. Therefore, the installed antiseismic support hangers meet the requirements for use.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Identification methods based on modal parameters for monitoring the health condition of concrete dams and assessing their performance during the operational period have gained significant research interest due to their economic efficiency and simplicity. This study summarizes current researches in signal acquisition techniques and parameter identification methods employed in the field of dam health monitoring, including ways of data collecting, identification algorithms, and the developmental trajectory of parameter identification for dams. Additionally, future research directions for utilizing modal parameters in dam performance identification are suggested. This study provides valuable insights for the advancement of health monitoring technologies in hydraulic structures.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Distributed deep learning has become an essential technique for accelerating deep learning, but its performance is often influenced by the heterogeneous computing nodes and heterogeneous communication networks within the distributed deep learning platform. Due to the high costs of practical deployment and running of distributed deep learning, it is almost impossible for researchers to optimize the training strategy of distributed deep learning tasks in real-world environments. In this paper, we propose HeterSim, a simulator specifically designed for heterogeneous distributed deep learning platforms. HeterSim enables flexible configuration of computing node performance and network connections, and supports to define and simulate distributed deep learning workloads using graph-based representations. HeterSim also allows for the modification of communication strategies in the distributed deep learning process, thereby assisting researchers in validating their designs for distributed deep learning. We verify the feasibility and flexibility of HeterSim, by generating a simulation platform at the scale of millions of nodes, and successfully simulate the distributed deep learning process of Resnet50. We aim to provide HeterSim as a flexible and user-friendly simulator for researchers, targeting heterogeneous distributed deep learning platforms, and helping researchers evaluate and optimize the strategy of distributed deep learning tasks at a lower cost.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In today's network environment, the application and dissemination of multimedia are becoming more and more extensive, and people are increasingly inseparable from the internet and multimedia in their daily lives, with a wide range of video applications emerging. But with this comes the problem of video security protection. Many images and videos on social networking platforms are stolen and re-published by other users for secondary creation. Therefore, it is not only necessary to protect the copyright of original videos, but also to ensure the information security during the storage and application of these video data. This paper proposes a video hashing algorithm based on the combination of global hashing based on invariant moments and partial hashing based on singular value decomposition, and the algorithm is used to detect the similarity of large-scale vehicle videos. By comparing with other video hashing algorithms, this paper's algorithm shows better results in terms of storage space, running time and recognition accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The prolonged duration of magnetic resonance imaging (MRI) presents a formidable challenge, necessitating the emergence of undersampling implementation as the primary strategy for expediting the imaging process. The optimization of the undersampling mask holds the potential to enhance imaging quality under equivalent acceleration. Diffusion model has showcased exceptional performance in image generation, offering heightened flexibility and an unsupervised nature. Consequently, it serves as a robust deep generation method for effectively addressing the inverse problem in MR reconstruction. Denoising diffusion probabilistic model (DDPM), distinguished by its enhanced flexibility in controlling the noise distribution, demonstrates superior adaptability to various undersampling modes, establishing itself as a promising deep learning method. In this study, we employ a novel approach to directly learn undersampling masks from data points, applying it to a reconstruction method for DDPM defined in K-space. Experimental evaluations conducted on publicly available fast MRI datasets reveal the method's commendable performance, surpassing conventional random bar mask-based and U-Net-based reconstruction methods and achieving superior reconstruction quality.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An improved YOLOv5 network with SPP and CBAM added to the YOLOv5 network is proposed to identify airport service lane damages accurately, quickly, and completely on the basis of ground-penetrating radar technology. First, the structure and performance of YOLOv5 are described in detail, while the common hidden defects are marked in combination with the actual GPR image data acquired at the airport. The improved network achieved a 6.2% improvement in network detection accuracy, a 5.7% improvement in recall rate, and a 5.1% improvement in mAP compared to the original one. These findings demonstrated the high accuracy and feasibility of the proposed optimized network for the detection of hidden defects in airport service lanes.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Focusing on a plunger type constant current output device designed for high sealing and long-life with complex multi-component coupling composition, high assembly precision, many influencing factors and other manufacturing problems, based on the in-depth analysis of the assembly precision of the plunger type constant current output device and the mechanism of the influencing factors of the operational performance, the surface quality of the structure, the axial runout of the system and the structural dimensional accuracy are accurately controlled by the precision error analysis and the corresponding design; The analysis decouples the rotational precision of the screw to propose a directional assembly method based on the principle of error compensation. Analyzing the leakage rate of sealing sub-factors in the constant-flow device, this study proposes solutions such as analyzing and controlling surface quality, dynamic sealing sub-filtering and matching, as well as regulating sealing sub-factor parameters to achieve shaft system runout accuracy of over 0. 007mm; The system's overall lifespan has been verified through an equivalent simulation test, achieving over 10500 hours of safe and stable operation and performance. This verification was conducted through an equivalent simulation test, resulting in a lifespan increase of 60%. This provides strong assurance for the space transportation system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To solve the problems of serious backlog of work-in-process (WIP) and high production cost in the manufacturing workshop of core components of an aerospace engine, this paper proposes a two-stage simulation optimization method based on Allocated Clearing Function-Genetic Algorithm (ACF-GA), aiming to optimize the production cost in the processing and manufacturing process. Firstly, a discrete event simulation model is established according to the existing data of the factory, and the model is warmed up based on the historical input data. In this process, the input-output loads of each work center are collected in a unit period, and the Clearing Function (CF) curve is fitted by piecewise linear fitting. Then, the ACF model is established to solve the initial production plan to minimize the production cost. In the second stage, it is iteratively optimized by combining the GA algorithm with simulation. The final result reduces the cost by about 40.45% compared with the initial production plan and has significant advantages over other solutions in reducing the on-line time of parts and improving the production stability of the production line, which verifies the effectiveness of the optimization method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Electricity, as a basic industry supporting economic development, will certainly be affected by the behavior of "government-enterprise", and electricity input is closely related to regional capital stock, economic growth, industrial structure and other factors. Therefore, based on the 2015-2022 panel data of 38 districts and counties in Chongqing Municipality, this paper empirically examines the causal relationship between various factors by using the panel vector autoregression (PVAR) model. The results of the study show that economic growth and capital stock positively promote electricity input in Chongqing Municipality, and industrial structure and labor force variables inhibit regional electricity consumption. In the future, the economic development of Chongqing should adhere to the goal of "stabilizing growth and adjusting structure", and through policy guidance and support for the development of the tertiary industry, the optimization of Chongqing's industrial structure, economic development, and energy saving and emission reduction can be achieved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The hydrogeological conditions of mining areas in western China are complicated, and roof flooding accidents often occur. However, Conventional inorganic water chemical parameters can not accurately distinguish the water source with similar composition in some well fields. Therefore, organic indicators including TOC, UV254, and dissolved organic matter (DOM) were added based on inorganic indicators including K++Na+, Ca2+, Mg2+, Cl-, SO42-, HCO3- and TDS, and the water source discrimination model was established in combination with random forest algorithm(RF). The main components and fluorescence intensity of DOM were obtained by fluorescence map and PARAFAC. After principal component analysis (PCA) of the inorganic data set and inorganic-organic data set, the discriminant model of inorganic indicators and the integrated discriminant model of inorganic-organic indicators were constructed by RF. The results show that the performance of the integrated discriminant model is significantly better than that of the inorganic discriminant model. To further improve the accuracy of the model, the artificial fish swarm algorithm(AFSA) is used to optimize the number of trees and the depth of trees in RF. To avoid local optimization, the adaptive speed, adaptive step, and weight attenuation mechanism are introduced into AFSA, and the water source identification model based on PCA-AFSA-RF was established by using the inorganic-organic data set. The results show that the precision, accuracy, recall, and f1_score of PCA-AFSA-RF model reach 93.67%percnt;, 91.93%, 95.19% and 94.13%, respectively, which are 8.48%, 6.75%, 9.18% and 10.48% higher than RF. And the model also accurately discriminates 13 unknown types of water samples. Therefore, it can be considered that the inorganic-organic comprehensive indicators can significantly improve the identification accuracy of coal seam roof inrush water sources, and the RF algorithm improved by AFSA has better global searching ability and convergence.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In order to investigate the flow characteristics of liquid-solid two-phase flow in an annular jet pump, computational fluid dynamics (CFD) was applied to simulate the internal flow field of a jet pump with sand flow. The three-dimensional flow field of solid particles transported by an annular jet pump was numerically simulated using RNAS and Euler two-phase flow models. The effects of sand particle concentration, solid particle diameter, and flow rate ratio parameters on the flow field and characteristics of the jet pump were analyzed. The findings indicate that during solid particle transportation, the jet pump exhibits asymmetry in the flow field due to gravity. Moreover, an increase in solid particle concentration or diameter exacerbates this asymmetry. Additionally, increasing the particle diameter suggests that a lower flow rate ratio can enhance the jet pump's efficiency and reduce the flow field's non-uniformity.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In recent years, with the increasing use of educational technology and online learning platforms, there has been a growing interest in developing intelligent systems that can automatically predict the knowledge points associated with educational questions. This paper presents a novel approach for knowledge point prediction in middle school mathematics questions. The dataset used in this study consists of a large collection of 591,379 middle school mathematics questions. To leverage the power of natural language processing techniques, the questions are preprocessed using a tokenizer and encoded into word embeddings using BERT (Bidirectional Encoder Representations from Transformers). Matrix dot production is then employed to calculate the similarity between each test question and the training set, and the top-n most similar vectors are selected for each test question. A voting mechanism is introduced to eliminate duplicate knowledge points and rank them based on their total scores in descending order, resulting in k candidate solutions. The performance of the proposed approach is evaluated using top-k accuracy as the evaluation metric and a lowest common ancestor (LCA) algorithm is used to calculate accuracy at different levels of the knowledge point tree. The results demonstrate the effectiveness of the proposed approach in predicting knowledge points for middle school mathematics questions and its potential for application in intelligent tutoring systems and educational assessment tools.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Correcting students' multiple-choice answers is a repetitive and mechanical task that can be considered an image multi-classification task. Assuming possible options are 'abcd' and the correct option is one of the four, some students may write incorrect symbols or options that do not exist. In this paper, five classifications were set up - four for possible correct options and one for other incorrect writing. This approach takes into account the possibility of non-standard writing options.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A vehicle with an articulated steering system is widely used in engineering due to its mobility advantage. The mechanism of the articulated steering system is different from the Ackman steering system equipped in regular vehicles. For the vehicle dynamic control process, the reference model is a quite important part. Thus, it is necessary to determine the proper reference model establishment method to improve the dynamic characteristics of the articulated steering vehicle (ASV). The dual body model, equivalent skid and slip model, and a novel model establish method, recombination structure model are built and compared. To analyze the calculated yaw rate of different models, varies mass center on the front and rear vehicle parts are set. The yaw rate during the steady-steering condition is gained and compared to the result gained from the Simulink/Recurdyn platform. According to the comparison, the recombination model has the best model accuracy in all conditions, when the distance between the mass center is lower than the axle distance, the skid and slip model has good accuracy. The dual-parts model has a complex structure and the worst accuracy in all conditions. The conclusion summarized based on the comparison could provide a reference to the work relevant to the dynamic control of the ASV
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This work is based on the prior paper, “Data-driven prediction of battery cycle life before capacity degradation”, by Kristen A. Severson et al. Using the same dataset, the effect of the Variance group in the prior paper is regenerated first, and then three types of input data and three models based on self-attention modules are proposed. One of the models is selected to conduct research comparisons on different input types of data. The advantages and disadvantages of different data types based on the self-attention module are analyzed, and finally the advantages and disadvantages of the self-attention mechanism compared with the Elastic Net used in the original research and the possible future improvement directions of battery life prediction based on the self-attention mechanism are discussed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Prediction of the distribution of mesoscale eddies plays a crucial role in sea navigation, guiding ships and submarines, and in route planning. In the past, the prediction of the trajectory of mesoscale eddies has primarily focused on their paths, often overlooking their distribution. However, given that the diameter of mesoscale eddies can span several hundred kilometers, their spatial distribution holds significant importance for the navigation and route planning of ships and submarines. To address this critical gap, this study concentrates on the South China Sea, a region renowned for its intricate sea conditions. The study introduces a novel BiST-LSTM model tailored to predict the distribution of mesoscale eddies. Leveraging the capabilities of the long short-term memory (LSTM) network, the model incorporates bidirectional spatial-temporal streaming and an attention mechanism. In this investigation, both the CMEMS dataset and a dataset constructed using the PYEDDYTRACKER method are employed to validate the efficacy of the proposed model and to compare its performance against other existing models. Experimental findings consistently demonstrate that the BiST-LSTM model outperforms alternative image prediction models in terms of evaluation metrics. Recognizing the impact of offshore typhoons on mesoscale eddy distribution, wind speed data from the ERA5 dataset are integrated to enhance the model's predictive capabilities. This augmentation significantly bolsters the model's accuracy in forecasting the distribution of mesoscale eddies, thereby reinforcing its utility in sea navigation, route planning, and ensuring safe passage during typhoon events.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In fact, three-dimensional complex cracks and other defects are often contained by the components of marine engineering structures. In the early stage of fracture mechanics, for the sake of simplicity and convenience in mathematical processing, the actual cracks are often simplified as two-dimensional models, and ideal two-dimensional cracks almost do not exist. In addition, the hull structure not only has low cycle fatigue failure, but also has obvious accumulative plastic failure under the severe sea conditions accompanied by cyclic ultimate external loads. Therefore, it is particularly important to study the accumulative plasticity of hull structures with surface cracks under low-cycle fatigue loads. In this paper, the hull structure of AH32 steel with surface cracks is taken as the research object, combined with three-dimensional elastic-plastic fracture mechanics, and then the large nonlinear finite element software Abaqus is used to conduct numerical simulation, and the constitutive model of surface crack plate is established. The stress-strain relationship at the crack tip was analyzed by discussing different ellipse shape factors, crack depth and crack plate thickness, and based on this, the variation law of the accumulative plasticity at the crack tip is further discussed. It lays a sturdy foundation for the follow-up study of crack propagation fracture and life evaluation of hull structures with surface cracks.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
National fundamental geo-entity construction has been vigorously promoted by the Ministry of Natural Resources of the People's Republic of China. As one of the core products, it is a digital abstract expression of the real world in the formof2 dimensions, which is difficult to fully meet the actual needs of analysis, calculation and use in the 3-dimensional digital space. This paper designs the process and key technologies for collecting elevation and height information of fundamental geo-entities to upgrade them from 2D to 3D form. This paper elaborates on the implementation process of key technologies such as elevation information acquisition based on DEM, stereo measurement of elevation information, height information acquisition. A practical and feasible process for elevation and height acquisition method of fundamental geo-entity has been formed, which effectively utilizes the advantages of geo-entities in information carrying, expression, sharing, and association. The experimental results show that the production process described in this paper can quickly and effectively obtain elevation and height information of geo-entities and result in 3Dfundamental geo-entities, providing technical support and important references for upgrading fundamental geo-entities from 2D to 3D form.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In light of the growth in the numbers of maritime disasters during recent years, there is a growing interest in the evacuation of passengers and crew at the sea, as documented in recent discussion at the Marine Safety Committee of the IMO. This study proposes an improved path selection algorithm to derive solutions for evacuation process. As preparations, we provide a precise mathematical model to describe evacuation process. With respect to the minimum evacuation time algorithm we develop a new path selection algorithm called network occupation force algorithm. Then the proposed algorithm is applied to Ro-Ro passenger ship to simulate the evacuation process. With the computational results compared with IMO simplified evacuation analysis, there are some discussions about the evacuation process analysis.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Intelligent ships are the new trend of maritime transportation tools, in which the secure, reliable, flexible, and resilient communication network is the core technology of intelligent ships and the critical challenge faced. To address this, the paper systematically analyzed the network access terminal characteristics and requirements of intelligent ships, focusing on the study of the resilient network framework and topology structure of intelligent ships. It designed a high-resilience communication network technology architecture for intelligent ships based on Software-defined Networking (SDN). Building on a systematic analysis of resource utilization in intelligent ship resilient communication network, the paper proposed an intelligent ship resilient communication network fragility segmented dynamic monitoring strategy (IS-SDMS), which was deployed in the control system of the intelligent ship resilient communication network. This achieved reliable and flexible resource management and security monitoring of the communication network, effectively ensuring the network security of intelligent ships.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As an important source of peak load, air conditioning plays an important role in alleviating the operating pressure of the power grid by reasonably controlling it through demand response. First, a baseline load forecasting method for residential user demand response under time-of-use electricity prices is proposed. The impact of electricity price fluctuations on CBL is considered from the peak and valley attributes and price differences of the period to be predicted, and historical load and temperature data are used as input features, using various regression methods such as Multiple Linear Regression (MLR) to predict CBL. Secondly, combined with the thermostatically controlled loads (TCLs) control method based on temperature set value adjustment, the impact of user comfort is considered, the adjustment characteristics are analyzed, and the response potential of TCLs is evaluated.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the improvement of the electrification level of transportation energy consumption, the coverage rate of port shore power gradually increases. Based on considering the uncertainty of distributed power generation output, this paper proposes a distributed power generation planning method for shore power distribution networks that takes into account the impact of uncertainty. Firstly, from the perspective of shore power distribution grid operators, an objective function is established that includes investment costs and operating costs. Secondly, a distributed robust optimization method is used to handle the uncertainty of distributed power output, and the confidence set of uncertainty probability distribution is constrained by both 1-norm and ∞ norm simultaneously. Finally, a column and constraint generation algorithm is used to solve the model. Through corresponding examples, the influence of model parameters such as confidence and historical data on the planning results of distributed power generation in coastal power distribution networks was analyzed. The superiority of this model was verified by comparison with deterministic and robust models.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To surmount the disorganization and scarcity of shared bicycle parking, it is imperative to devise a judicious allocation and blueprint of urban shared bike storage locations. The present study has constructed a mathematical model aimed at minimizing the total cost of shared bicycle enterprises, and designed a genetic algorithm to discern the optimal deployment scheme for shared bike parking points and the respective quantity allocation for each such site. To validate the viability of the model, this paper has selected the Tiexi Square in Shenyang as a case study. A total of 25 alternative locations were selected within the case region, following the established selection criteria. The optimal parking location was determined through the genetic algorithm designed for the problem, utilizing data from the 15 demand points received, yielding the most favorable position for parking.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Conduit rack platform is an important facility for offshore oil development. This paper is based on the environmental load theory and ANSYS theory, firstly, the size of the wind and wave, the role of the orientation, the role of the direction, etc., and then use the mechanical analysis module in ANSYS to establish a three-layer conduit rack platform finite element model and loading analysis, and derive the three-layer conduit rack platform finite element model in a variety of combinations of the role of the load of the stress, strain, the overall deformation, the maximum bending moment, etc., and then the conduit rack platform strength check, which is a guide to engineering practice. The strength of the platform is calibrated, which has certain guiding significance for engineering practice.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This study makes use of the daily vessel traffic flow in the waters near the Three Gorges Dam as its object of study and uses the bat algorithm to optimize the regularization parameter C and the kernel parameter σ of the least-squares support vector machine (LSSVM). Subsequently, a hybrid BA-LSSVM model proposed in this work is utilized to forecast the flow of vessel traffic. The algorithm was validated by real data, selecting 70% of the data as training set and 30% of the data as test set, and compared with the least squares support vector machine (LSSVM) prediction method. The results show that the model has good RMSE index, correct method and high prediction accuracy. Meanwhile, the engineering value and practical significance of the model provide a feasible method for traffic flow prediction.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the increase of air traffic volume in recent years, the airport operation environment has been in complex conditions, which may lead to the occurrence of scratches or collisions between aircrafts in the active area of the airport. To solve this problem, the airport is regarded as a discrete system, the flight activity area is divided with the idea of modularization, and the topological relation model is constructed by the divided modules. Then the topological relation model is simulated and designed quickly using simulation software. Also, the taxiing path of the aircraft is analyzed, and the potential conflicts between the aircrafts are found and controlled. Based on this practical requirement, this paper aims to construct a model of the activity area of the airport by using a colored stochastic Petri net. Through the simulation function and control library of CPN Tools, the aircraft can run on the constructed traffic network model and support conflict control. The simulation results show that the airport activity area aligns with the actual operating environment. Compared with the activity area model without a control base, the control base plays a role in conflict control of aircraft on the traffic network, and the aircraft can take off and land smoothly in the airport.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper, the performance of OFDM is simulated and evaluated from various dimensions by comparing with similar modulation methods. OFDM is compared with F-OFDM (filtered OFDM), FBMC (Filter Bank multi-carrier modulation) and UFMC(Universal Filtered multi-carrier modulation).Firstly, the concept and background of each modulation are introduced, and then the corresponding parameters of OFDM modulation are simulated and compared . Then the model of the channel-free receiver is built and simulated. Finally, the UFMC has the characteristics of high spectral efficiency, large bandwidth and sidelobe suppression .
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To mitigate the parameter variations within the turbine volute during operation, which resulting in turbine component damage and increased hydraulic losses, this paper proposes an optimization and retrofitting strategy for the turbine volute. The volute's structural conFigure.uration is optimized using the adjoint solver approach, and the changes in the internal flow field are numerically evaluated before and after the turbine optimization. Computational results demonstrate that the proposed optimization scheme reduces the gradient of the overall pressure distribution within the volute's internal flow field, leading to a smoother and controlled alteration of the overall flow velocity of water. Moreover, the optimization effectively improves the distribution pattern of the volute's flow field. A comprehensive comparative analysis of the inlet and outlet pressure drop and energy loss, before and after optimization, reveals an increase in pressure drop and a decrease in energy losses. These findings highlight the potential of employing the adjoint solver-based approach for volute optimization.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Automotive Grade Chips need to take into account the effects of unintended failures during operation, such as open circuits caused by chip aging or flipped memory devices caused by particle bombardment. Therefore, in the design process of automotive-grade chips, it is necessary to evaluate their safety and reliability. In this paper, we analyze the failure sources of automotive-grade chips and classify them according to the failure modes, comprehensively consider the impacts of different safety mechanisms, qualitatively and quantitatively analyze the functional safety of automotive-grade chips under transient failures, and quantify the level of protection against failures of automotive-grade chips under the transient failure conditions by means of mathematical models and evaluation indexes. The experimental results prove that it is feasible to evaluate the safety of the vehicle-grade chip through the functional safety analysis method at the design stage, and verify the protective effect of the safety mechanism of the chip on the vehicle-grade chip, which can guide the work of the vehicle-grade chip designers.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Since its introduction by the National Cryptography Administration in 2016, the SM9 algorithm has encountered various opportunities and challenges. This study introduces a refined bottom-up modular framework, resulting in a more systematic and comprehensible algorithm structure with standardized naming conventions and clear module component descriptions. At the same time, we have proposed a comprehensive optimization scheme for bilinear pairing, optimizing the internal calculation process and improving the usability and efficiency of the SM9 algorithm. The experimental results at both the software and hardware levels have confirmed the effectiveness of our proposed method. Our work goes beyond previous research that mainly focuses on individual modules or sub-modules, and has made significant progress in promoting the widespread application of the SM9 algorithm in practical scenarios.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Distributed Electric Propulsion aircrafts are a novel concept in electric aviation, characterized by the wing-propeller interaction. The interaction depends on several factors, such as the distance between the wing and propeller, the wing chord-to-propeller diameter ratio, the angle between the wing and the propeller, and so on. Firstly, the MRF model was utilized to simulate the APC9060 propeller and to analyze the wing-propeller interaction. This study focuses on the angle between the wing and the propeller, examining its impact on the interaction. A multiplicative factor was introduced to quantify the effects of propeller slipstream velocity and develop a surrogate model for this factor. An independent, low-cost DEP slipstream experimental platform was designed for model parameterization. The numerical simulation revealed that the slipstream increased the flow velocity on the wing surface and enlarged the low-pressure area on the upper surface of the wing. The results from experiments demonstrated the model accurately predicted lift changes as the angle of attack varied from 11 to 15 degrees.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Generally, large-scale or intelligent software development has the programming characteristics of complex algorithm, such as uncertain requirements, complex workflows, and frequent modifications. Under the constraint of limited perspective, studying progressive logic refinement and adaptive architecture have great significance for complex algorithm development. Starting from the analysis of the task characteristics, the big decision tree theory was proposed to solve complex algorithm development problems, and the methods for constructing, expanding, and running big decision trees were introduced. Focusing on the dynamic addition or modification of logic conditions, adaptive running mechanism, and running of big decision tree systems, this paper proposed the adaptive architecture and its implementation techniques. Taking the technique implementation of state monitoring and robust debugging as examples, the implementation methods of adaptive architecture were introduced. Finally, the future directions were summarized.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The development of autonomous driving and augmented reality has led to increased demands for accuracy and speed in visual SLAM. Traditional feature-based visual SLAM performs real-time image analysis with high complexity and computation requirements, necessitating devices with exceptionally high arithmetic power. This poses challenges for deployment on low-power devices. Additionally, relying solely on photometric errors makes the system too sensitive to the environment. To tackle these issues, our algorithm integrates the feature-based method and the sparse direct method. We employ the sparse joint method for pose estimation in the tracking thread, considering both accuracy and speed. Incase of tracking failure, we extract feature point lines for tracking. Furthermore, to enhance system robustness in low-texture environments and fuse point-line features, we employ adaptive weighted extraction of point-line features, considering computational resources. Our experimental results, based on public datasets, demonstrate that this algorithm achieves faster operation speeds in simple environments and higher accuracy compared to other SLAM algorithms in complex environments and fast camera movement scenes.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As an important tool for material stacking, handling and transportation in enterprises, wooden pallets have many problems such as large loss, low recovery rate and high procurement cost. Therefore, improving the recycled rate and recycling utilization rate of wooden pallet, and reducing its failure rate, is an effective means to reduce the operation cost of enterprises. Starting from the actual situation, this paper first establishes the closed-loop supply chain model of Company J, then concretizes and generalizes the entire supply chain model by constructing the closed-loop supply chain mathematical model. Finally, through analyzing and comparing the data related to wooden pallets of the company in two years, it proves that wooden pallets adopt the closed-loop supply chain recycling mechanism can greatly reduce the operating costs of enterprises, and also has important reference significance for realizing resource protection and green production.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As clean energy generation grows, power systems require greater flexibility to ensure safe operation. As a critical emerging entity that assists the power grid in balancing peak and valley load fluctuations, virtual power plants can provide flexible real-time power through demand response and coordinate reasonable dispatch responses among all participants to alleviate the contradiction between supply and demand during peak electricity consumption periods. Large industrial loads have the characteristics of a large load base. Currently, there is a lack of research on accurately analyzing the adjustable potential of power users in large industrial industries and then integrating them into virtual power plant operation dispatching. This paper introduces a multi-time-scale demand response assessment method, analyzes the response load characteristics of the steel and electrolytic aluminium industries, and proposes an optimal dispatch method based on assessing the adjustable potential of large industrial loads to optimize the load aggregator's cost benefits. Taking maximization as the goal, taking the data of Xinjiang Province as an example, the planning declaration curve of a large industrial load is solved, and the economic dispatch of a large industrial load is realized.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the rapid development of the power market and the large-scale integration of renewable energy, the load adjustment capacity of the power system has become increasingly important. Large industrial users, as the main load component of the power system, have significant load adjustment potential. This paper proposes a load adjustable potential assessment method for large industrial users based on the decadal time scale. By analyzing various adjustable potential indicators within the decade, the TOPSIS method is used to comprehensively evaluate the load adjustment potential of large industrial users. First, collect historical load data and related environmental data of large industrial users; then, quantify the indicators on the decadal time scale; finally, propose corresponding load adjustment strategy suggestions based on the indicator analysis results. The actual example shows that the evaluation method and indicator system proposed in this paper are reasonable and effective.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To address the dynamic autonomous collision avoidance problem of ships in coastal waters, this paper proposes an autonomous collision avoidance model by combining the trajectory prediction with the improved velocity obstacle model. Firstly, the relative motion model of the two ships is established and the motion parameters of collision avoidance are calculated in real time. Based on the three-degree-of-freedom Maneuvering Modelling Group (MMG in short) model and dead reckoning algorithm, the motion state of the own ship (OS in short) and the target ships (TS in short) at the next moment will be predicted, respectively. Then dynamic autonomous collision avoidance decision-making model is designed with the combination of "International Regulations for Preventing Collisions at Sea” (COLREGs in short), good seamanship, trajectory prediction model, improved velocity obstacle algorithm, and course control system. Finally, the feasibility and validity of the dynamic autonomous collision avoidance decision-making model are verified through simulation experiments. The simulation results show that the model can realize the autonomous collision avoidance in different typical encounter situations, allowing the safe navigation of ships under dynamic collision avoidance conditions.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
According to the relevant test standards and the characteristics of special electric vehicles, the first-level evaluation indexes are set up as power, trafficability, mass geometric parameters and electric drive parameters, and the second-level indexes are determined based on typical driving conditions. The usability of trafficability index is analyzed, the weights of primary index and secondary index are determined by AHP and CRITIC respectively, and the comprehensive weights are determined by the combination of qualitative analysis and quantitative analysis. The correlation degree of vehicle index is calculated by grey correlation theory, and the vehicle evaluation result is determined by comprehensive weight. A comprehensive evaluation method for off-road capability of special electric vehicles is established by combining subjective and objective methods.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Visual object detection and ranging is a technology that uses computer vision technology to detect and distance target objects in images or videos. This technology plays an important role in the field of industrial automation. In this paper, based on the improved yolov5 algorithm, the adaptive receptive field convolution module is introduced in target detection, and the triangulation principle in the traditional method is used to achieve target detection and ranging, so as to achieve more accurate target detection and improve ranging accuracy. The accuracy of the system can be verified by many experiments. The experimental results show that the ranging error of the system is less than 5% in the range of 0.2~1m.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Aiming at the problems of difficult data collection, poor quality and low utilization rate of highway in high temperature and rainy mountains areas, an intelligent maintenance management system is designed to realize the combination of inspection and maintenance, and integrate the design, construction and maintenance information of highway facilities. Based on GIS map service, the system integrates regression prediction algorithm and maintenance decision algorithm to realize six functional modules, such as road performance evaluation and maintenance decision-making, to provide users with scientific decision support, improving management efficiency, saving management costs and reduce maintenance costs, and achieving the goal of intelligent management and maintenance.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper takes 57 national surface meteorological observation stations as the research object, based on the fault and maintenance records in the remarks of the meteorological data operation system (MDOS), carries out statistical analysis on the fault conditions of national meteorological observation stations, analyzes the fault causes according to the different characteristics of various faults, classifies them, and puts forward the fault detection and maintenance steps.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To solve the problem of low efficiency of the current drill pipe counting method, a new drill pipe counting algorithm based on human posture recognition is proposed in this paper. Through the back-end algorithm, real-time analysis the video of drilling withdrawal which is in drilling face, record the key point coordinates of the human body, and determine the actual number of drill pipe by detecting whether the worker captures the drill pipe and whether there is carrying action. The method detects the continuous action of the workers taking down the drill pipe through human posture recognition, and automatically calculates the number of the workers taking out the drill pipe, so as to improve the accuracy of the intelligent video analysis of the drill pipe count, and has good engineering application value.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To enhance the resilience of the power grid’s key material supply chain and ensure the safety, stability, and robust development of the power industry, the paper employs the Wuli-Shili-Renli (WSR) Methodology to construct a resilience evaluation system comprising both quantitative and qualitative indicators. The paper utilizes the cloud model to facilitate the conversion between qualitative concepts and quantitative parameters, employing comprehensive method to determine the weights of indicators. Moreover, the study evaluates the supply chain resilience of critical and representative enterprises through the example of the cable supply chain. The findings indicate that the cable supply chain achieves the highest score in the 'Renli' primary indicator, with other two primary indicators and overall resilience performing well. However, shortcomings are identified in the secondary indicators related to 'supply chain network structure' and 'human resource management.' Based on these evaluation results, the paper proposes corresponding strategies for enhancing resilience, offering guidance for the State Grid Corporation of China.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In response to the lack of systematic planning in the welding workshop layout of a certain bogie frame, resulting in unreasonable workshop layout and unsmooth logistics, this paper proposes a simulation optimization method for workshop layout based on simulated annealing algorithm. Firstly, the workshop production simulation model is established on the basis of the detailed investigation of the workshop business, so as to realize the accurate simulation of the real workshop production.Then, aiming at the shortest transport distance, an improved simulated annealing algorithm is designed, and the field layout experience rules are summarized and added into the algorithm as the initialization method and heuristic operator. Finally, the algorithm is combined with the simulation model, the simulation model is used as a decoding tool, and the optimization framework based on intelligent algorithm is realized through algorithm iteration, and the optimal solution of the layout scheme is obtained.The optimized framework welding workshop layout scheme reduces the handling distance by about 16% compared to the current layout of the workshop under the same production plan, making the material transportation path more reasonable and bringing logistics benefits to the factory.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
To address the inefficiency of target detection algorithms for UAV in maritime search and rescue, an improved UAV target detection algorithm is proposed. First, the MobileNetV3 network is fused in the backbone network of YOLO-v5s to reduce the computational parameters of the algorithm model and make the algorithm easier to be deployed on the UAV embedded platform. Then, to address the problem of more interference factors in the detection image, CBAM (Convolutional Block Attention Module) attention mechanism is introduced to make the algorithm model pay more attention to the feature information of the trapped person, so as to improve the detection efficiency of the algorithm model. The experimental results show that the precision rate P of the improved algorithm reaches 0.966, the recall rate R reaches 0.937, and the mean average precision mAP_0.5 reaches 0.955, which is of high practical value and is expected to play an active role in the maritime search and rescue work.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Orthogonal frequency division multiplexing (OFDM) capitalizes on the orthogonal properties inherent in various subcarriers to optimize spectral efficiency within the system. This technique stands as the foundational technology for forthcoming wireless communication transmission and multiple access methodologies. Although conventional OFDM systems effectively mitigate multipath interference through the addition of cyclic prefixes (CPs), CP-OFDM systems are not without their limitations: 1) OFDM systems inherently exhibit high sidelobes in the sinc function, rendering them susceptible to frequency offsets; and 2) while this characteristic aids in interference elimination, it concurrently diminishes the system's spectrum utilization.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Hydrological forecasting is of great significance for flood control and disaster reduction, water resource management, hydropower energy production, and environmental protection. This article proposes a novel approach that combines an artificial intelligence algorithm and a decomposition algorithm. The proposed method first uses a Seasonal Decomposition (SD) algorithm to decompose the runoff series into subcomponents including trend, seasonal, and residual components. Then, the Simple Recurrent Unit (SRU) algorithm is used to make predictions for each subcomponent, which are then integrated into the final prediction result. The proposed model was tested on the Dongjiang dataset along with several other common machine learning prediction models including Gated Unit Recurrent (GUR), Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), Backpropagation (BP), and Linear Regression (LR). The results showed that the proposed model outperformed the other models, with a coefficient of determination R2 of 88.94 and a Nash-Sutcliffe efficiency NSE of 0.936. The proposed model is particularly effective at predicting flood season runoff sequences.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper takes the concrete structure berth wharf of Tianjin Port as the research object, conducts on-site investigation on the status of freeze-thaw of the port's concrete structure, determines the distribution law and the most unfavorable position of the freeze-thaw damage of the port's hydraulic concrete building components, and based on the osmotic pressure theory analyzes the concrete freeze-thaw damage mechanism and calculation of average pore pressure during concrete freeze - thaw process. Considering the coupling effect of temperature field, seepage field and stress field, the failure process of concrete under freeze-thaw cycles was simulated by Comsol Multiphysics, and the distribution characteristics and variation laws of temperature field, seepage field and stress field were obtained. The research shows that the freeze-thaw damage of concrete components in Tianjin Port Wharf shows obvious regularity, and the freezing of internal pore water is the direct cause of freeze-thaw damage of concrete structures. The average p process of concrete is closely related to the freeze-thaw temperature and the number of freeze-thaw cycles. During the freezing and thawing process of the concrete specimen, the temperature of the specimen gradually decreased from the surface to the center, and the seepage velocity inside the specimen showed the law of Z direction>Y direction>X direction. and the distribution characteristics of stress and strain were consistent, and both showed big peripheral, the center is small and relatively symmetrical. The rese arch results can provide a theoretical reference for the safe operation and maintenance of port hydraulic concrete structures.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
China is actively promoting the development of renewable energy sources such as wind power and photovoltaic to achieve the goal of "carbon neutrality and carbon peaking". However, renewable energy sources such as wind and solar power have the characteristics of randomness and volatility, leading to a continuous decline in the safe and stable operation level of China's power system. There is an urgent need to equip a large number of reliable and flexible regulatory resources. Among the existing flexible regulation resources, pumped storage power stations are currently the most mature, reliable, and construction-effective large-scale energy storage power sources. They can provide peak shaving, frequency regulation, and other services, as well as undertake emergency backup for power grid accidents, improving the safe and stable operation level of the power grid. Especially, pumped storage power plants have a second level response speed, which can provide a large amount of flexible and reliable regulation capacity for wind power plants, storing excess energy generated by wind power instantaneously, and effectively avoiding waste of wind power resources. Blockchain is a chain composed of blocks one after another. Each block stores a certain amount of information, which is connected into a chain in the chronological order of their respective generation. Therefore, tampering with information in the blockchain is extremely difficult. Compared to traditional networks, blockchain has two core characteristics: first, data is difficult to tamper with, and second, decentralization. Based on these two characteristics, the information recorded by blockchain is more authentic and reliable, which can help solve the problem of people's mutual distrust.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
There is a general upward wind on the windward slope of mountains. Under the action of this vertical upward airflow, the vertical and lateral wind deflections of the conductor and ground wire will increase significantly. The current specifications in China do not consider the vertical wind speed component in the upward wind on the windward slope, which may lead to significant calculation deviations in the wind deflection displacement of conductor and the ground wire. Taking the example of the interphase discharge of the conductor and ground wire in a microtopography area in Xinjiang, with the help of fluid dynamics simulation, the upward wind speed component at the windward slope is obtained, and the wind deflection response analysis of the ground wire considering the influence of the vertical wind speed component is carried out. The analysis shows that as the wind speed increases, the vertical and lateral distances of the ground wire are continuously decreasing, and the decrease in vertical distance is more significant. Eventually, when the wind speed reaches a certain magnitude, the interphase distance of the conductor and ground wire is less than the discharge gap limit, which induces the interphase discharge of the ground wire.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This article aims to construct and validate a student information literacy evaluation model based on the BP neural network algorithm. In the current information age, the information literacy of students is crucial for their academic success and future development. However, traditional evaluation methods often lack accuracy and comprehensiveness. To address this issue, we have designed an evaluation model based on BP neural network, which can comprehensively analyze student questionnaire survey data and learning behavior logs, providing a more accurate and comprehensive assessment of information literacy. We demonstrated the training and optimization process of the model through detailed experimental design, and demonstrated its effectiveness through performance evaluation. The experimental results show that the model performs well in performance indicators such as accuracy, recall, and F1 score, and is superior to traditional machine learning models. Finally, we discussed the limitations of the model and its potential applications in the field of education. This study provides a new perspective and method for utilizing advanced machine learning techniques for educational evaluation and decision-making.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This article delves into the design and implementation of innovation and entrepreneurship systems based on cloud model data mining algorithms in university environments. Firstly, through in-depth communication with stakeholders such as university management, teachers, and students, a comprehensive analysis and planning of the system's requirements were conducted. Next, the article provides a detailed description of the system architecture design, including key components such as data collection and management, data preprocessing, cloud model data mining, user interface, reporting and visualization, security and privacy protection, as well as user feedback and support. During the system development process, emphasis was placed on the design principles and construction of cloud model data mining algorithms, as well as the optimization measures taken to improve system efficiency and accuracy. The system implementation process follows a series of strategies, including preliminary planning, system development, data preparation and migration, system integration, user training, as well as system deployment and continuous maintenance. Each stage is carefully designed and executed to ensure the effectiveness and reliability of the system. After implementation, a comprehensive performance evaluation of the system was conducted, including accuracy, efficiency, and user satisfaction. The results showed that the system performed well in multiple aspects, but also pointed out some areas that needed improvement. Finally, the article summarizes the main advantages and limitations of the system, and puts forward suggestions for future work directions, including further optimizing user experience, algorithm performance, and expanding application scope. This study provides an efficient and reliable data support and decision-making assistance tool for innovation and entrepreneurship activities in universities, which is of great significance for promoting the development of the innovation and entrepreneurship environment in universities.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This study uses association rule mining algorithms to conduct in-depth analysis of sales data of retail enterprises, aiming to reveal the purchasing patterns and correlations between products. By analyzing the transaction records of supermarkets, this study identified multiple frequently purchased product combinations and proposed targeted product layout strategies based on these findings. These strategies include neighboring placement of interrelated products, bundled promotions, and inventory management optimization. The preliminary results after implementation indicate that these data-driven layouts and promotional strategies have effectively improved sales and customer satisfaction. This study not only demonstrates the practical application of association rule mining in the retail industry, but also provides a new perspective for retail enterprises to utilize sales data and optimize business operations. Finally, this article also discusses the limitations of the research and proposes possible directions for future research.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the "14th Five-Year Plan" tourism development plan, the distribution of B&B has gradually become essential optimizing tourism spatial layout and promoting regional coordinated development. Based on the POI data and the GIS spatial analysis method, this paper analyzes the spatial distribution and influencing factors of B&Bs in Liangshan Prefecture. The results show that the spatial distribution trend of B&Bs in Liangshan is in the direction of "northeast-southwest", and the distribution of B&Bs has a high degree of aggregation, and the distribution is unbalanced. The spatial distribution of B&Bs showed a significant aggregation pattern, and formed a spatial distribution pattern with Xichang city as the core, Yanyuan County as the secondary core, Huili, Dechang and other places gathering in a small range. At the same time, the paper uses the geographical detector and the geographical weighted regression model to comprehensively analyze the factors affecting the spatial distribution of B&Bs in Liangshan Prefecture from the global scale and local scale. The results show that the spatial distribution of B&Bs is highly positively correlated with the density of scenic spots.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The proliferation of recommendation systems has revolutionized information retrieval by helping users efficiently navigate through large data sets. However, these systems often suffer from information bias, especially in multimodal recommendation setups. This paper addresses the issue of bias mitigation in multimodal recommendation systems using expert systems. Through a comprehensive literature review, various techniques such as knowledge graph integration, multimodal fusion, and deep learning architectures are explored. Furthermore, a novel approach using dynamic expert meeting algorithms for bias mitigation is proposed. Theoretical frameworks of expert systems are discussed, highlighting their adaptive capabilities and applicability to diverse domains. Then, the methodology for addressing information bias in multimodal recommendation systems is presented, including experimental analysis and relevance tagging. The results demonstrate the effectiveness of the proposed approach in reducing bias and improving recommendation accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In the wide application of computer technology and multimedia technology, corpus is widely used in linguistic theory research, military field and so on. Especially in the era of big data, the branch of language research with the corpus method as the core has been attached great importance by scholars all over the world, and a large number of corpora have emerged one after another. After scientific selection and effective annotation, corpora with appropriate scale can fully demonstrate and record the use of language, and people can grasp language facts by observing corpora. This paper analyzes the operation rules of language system and provides convenient conditions for language research. On the basis of understanding the current situation of corpus design and application in the new era, this paper mainly explores the speech emotion feature extraction method based on deep neural network. The final experimental results show that the classification of emotion features by extracting statistical features has a good recognition effect, and the corpus acquisition method based on neural network has a large development space.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In the development of social economy and scientific and technological progress, the ecological and environmental protection awareness of social residents is becoming stronger and stronger. Due to the extensive industrial development period experienced by China for a long time, more environmental pollution problems have been left, so in the face of the new market development situation, emission enterprises in various fields should focus on their own carbon asset management capabilities. Especially under the development trend of economic globalization, China's key emission enterprises have begun to actively participate in carbon asset management, and have built a new management system by using big data technology. Therefore, in order to understand the current status of carbon asset management in key emitters under the background of China's carbon market, this paper comprehensively explores the auxiliary decision-making model and value evaluation method of enterprise carbon asset management, and then puts forward corresponding management measures from the perspective of new era development, so as to provide reference for the carbon asset management of key emitters
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As a very popular teaching resource in recent years, "micro video" has the characteristics of highly focused learning themes, highly focused learning content and repeatable learning. When applied to life science teaching, it can refine essence content for knowledge teaching, provide students with repeatable learning, and improve the memory strength and mastery of new knowledge according to the reality of students' short attention time. Artificial intelligence technology can effectively help online course teachers solve this problem. The use of facial recognition, emotion detection, and behavior analysis technologies in artificial intelligence technology can conveniently monitor and analyze the learning behavior of online course students, achieve multidimensional evaluation of the teaching process, and more efficiently achieve precise teaching.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Computer assisted language learning has developed rapidly and has received widespread attention from more and more scholars. This discipline mainly promotes simple foreign language teaching activities through the assistance of computers and information technology. This article designs and implements an English speaking learning system based on speech recognition technology, which provides users with more time and space convenience advantages as well as diverse interactive characteristics when learning English speaking. The focus of this paper is on the computer-aided evaluation method of pronunciation quality in English speaking follow-up reading. Participants in the oral test can randomly select speech samples from the speech training database and follow them according to their standard pronunciation. During the operation of the system, the feature information of the two is extracted for comparison, and the pronunciation quality is evaluated by calculating the Euclidean distance between the feature parameters of the standard template and the training template. Finally, through experiments and testing, it was confirmed that the system can effectively improve the system's scoring performance by introducing average pronunciation level, and has a certain effect on improving users' oral pronunciation level.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the further promotion of information technology in universities, the continuous enrichment of online teaching resources, and the continuous standardization of online teaching management, online learning as a main learning method has been increasingly accepted by more and more people. This article selects fuzzy neural networks for modeling and constructs a network learning evaluation system based on the characteristics of network learning evaluation. The English teaching classroom based on artificial intelligence and virtual reality proposed in this article can provide scarce educational resources while improving students' learning outcomes. The experimental results show that the method proposed in this article has a higher level of acceptance among students compared to traditional teaching methods, with a 12.7% higher learning efficiency among students. It is feasible in theory and reliable in practice, and can be used as an intelligent computing model for developing network learning evaluation systems. It has good promotional value and provides a new evaluation method for network learning evaluation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Convolutional Neural Network models can be used to train a large amount of data for subsequent data analysis. This paper proposes a Convolutional Neural Network framework model based on Bootstrapping for Deep Learning and analysis of massive face data containing noise. According to the analysis of experimental results, it is concluded that the proposed Bootstrapping based Convolutional Neural Network framework model can achieve Deep Learning and subsequent prediction of large-scale noisy labeled facial data. It has the advantage of low computational cost and has achieved stateof- the-art results in various facial benchmark tests.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
With the rapid development of information technology, especially the rise of deep learning technology, traditional financial management methods are facing enormous challenges and transformation opportunities. This study aims to explore how to use deep learning technology to build an efficient and intelligent shared financial management system, providing more accurate, fast, and forward-looking decision support for enterprises. On the basis of in-depth analysis of the current financial management status and existing problems, this article first introduces the key role of deep learning in financial data processing, such as data preprocessing, anomaly detection, trend prediction, etc. By combining the characteristics of financial data, we propose a shared financial management system architecture based on deep learning, which includes core modules such as data collection, data preprocessing, model training, and financial decision support. In addition, this study also conducted in-depth discussions on how to optimize the model structure, select appropriate algorithms, and ensure data security and privacy. Our experimental results indicate that compared to traditional methods, the shared financial management system based on deep learning not only significantly improves the accuracy and efficiency of data processing, but also provides enterprises with more in-depth and comprehensive data analysis, helping them better respond to market changes and achieve financial goals. Finally, considering that the application of deep learning technology in financial management is still in its early stages, this article also conducts a forward-looking analysis of potential challenges and countermeasures in the future. I hope this study can provide useful insights and references for enterprises and researchers in the combination of deep learning and financial management.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In the context of the new era, because the potential information of the data itself is difficult to accurately query, so in the face of faster and faster development requirements, industries have begun to fully penetrate big data technology, big data analysis has become the mainstream of industry development, but also an important means for enterprises to break through the development bottleneck. Therefore, after understanding the main content and development status of big data mining technology, this paper mainly studies the data mining technology with hybrid neural network as the core, clarifies the application direction of big data mining technology in various industries in the new era, and provides technical support for the innovation and development of The Times.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Power engineering as a basic component of economic construction and development, the construction and management of intelligent substation affects national energy security and economic development. Especially after entering the era of big data, China's power engineering has begun to develop steadily in the direction of intelligence, digitalization and information technology, the construction management of intelligent substation is undergoing profound changes, and the project cost management has new ideas and means. Therefore, after understanding the practical significance of power engineering cost control in the era of big data, this paper mainly studies the cost management countermeasures of smart substation supported by big data technology according to the architecture of smart substation big data control system, so as to provide an effective basis for the engineering construction management of power enterprises in the new era.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The Generative Adversarial Network learns discriminative facial feature representations by training a generator network and a discriminator network to compete and cooperate with each other. This paper proposes a Generative Adversarial Network framework model based on mean feature matching. By mean feature matching losses, the Generator Network can learn feature representations that are closer to real faces, thereby improving the performance of face recognition. Through the analysis of experimental results, it is concluded that the Generative Adversarial Network framework model based on mean feature matching proposed in this paper has higher accuracy and stronger robust, and has achieved the most advanced results in various facial benchmark tests.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Graph Convolutional Networks (GCN) have emerged as effective models in graph data analysis and Deep Learning tasks, achieving significant advancements in recent years. In this paper, we propose GCN based on clustering algorithm for face recognition. By transforming the clustering problem into two sub-problems, two graph convolutional networks are employed, one graph convolutional network estimates the confidence of vertices in the graph, and another graph convolutional network estimates the connectivity of edges in the graph. This approach effectively captures both local and global information among face images. Experimental results demonstrate that the clustering-based GCN model achieves state-of-the-art performance on various public face datasets, highlighting the effectiveness and potential of clusteringbased GCN in face recognition tasks. These findings provide valuable insights for further improving and applying GCN models and serve as a beneficial reference in this field.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Digital platform has become one of the key factors for enterprises to succeed in today's competitive business environment. However, in order to maintain competitiveness and continuously meet the needs of users, the digital platform must be continuously upgraded and intelligent. This paper discusses the intelligent upgrade strategy based on machine learning (ML), aiming at improving the performance, user experience and business value of the digital platform. Firstly, the paper introduces the architecture of digital resource service platform, including data analysis, personalized recommendation, automatic process optimization and so on. The application of ML technology enables the digital platform to better understand user behavior, optimize resource allocation and provide more attractive functions. Next, this paper introduces a series of intelligent upgrade strategies, including key steps such as data collection and analysis, model development and training, deployment and monitoring. These strategies aim to ensure the reliability, accuracy and expansibility of ML model, thus improving the intelligence level of digital platform. The intelligent upgrade of digital platform is necessary to meet the changing market demand and user expectations. By adopting ML technology and following effective upgrading strategies, enterprises can make better use of data assets, improve operational efficiency, increase competitiveness and achieve sustainable success.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Information is the end of the big data resource system construction, providing a source of fresh water for data resources and an important foothold for data demand. This paper analyzes the current situation and problems of smart campus construction under the background of informatization. The practical needs and development goals of smart campus construction are analyzed, and the four core construction directions of quasi smart campus construction business architecture, data architecture, application architecture, and technical architecture are explored. In response to the shortcomings in information data construction and application capabilities, data ownership, data nodes, data talents, and data should be the key factors, forming a data rights and responsibilities relationship that fits the characteristics, supporting infrastructure, and professional talent teams, and promoting normalized data drills for information.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.