Mugwort floss, valued in traditional Chinese medicine, varies in therapeutic properties and market price based on origin and production year. Traditional identification methods, due to their destructiveness and low accuracy, often confuse mugwort floss with A.stolonifera and cause a testing waste. Hyperspectral Imaging, a non-contact technique, offers potential for rapid identification of such medicinal materials. In this paper, we explore hyperspectral data to differentiate mugwort and A.stolonifera using deep learning and neural networks. Using a massive hyperspectral dataset from mugwort and wormwood from two regions across four years, we analyzed performance using metrics like Accuracy, Specificity, and F1 Score. The self-attention-based Backpropagation Neural Network model showed the most promising results for accurate classification. This approach has potential future applications in various fields using Hyperspectral data
Steel structure has been widely used in modern buildings because of its excellent building performance such as lightness, low-cost. However, the poor fire resistant of steel structure can lead to a lot of hidden troubles including life safety and economic losses, cause the construction of steel structure become a key research direction in fire investigationwork. Currently, the experts for fire investigation have to analyse the surface trace of the steel evidence after the fire throughempirical data by person because of the lack of intelligent methods and equipment, which have a great influence on the investigation efficiency. This article proposes an intelligent detection method for steel structure fire traces based spectral imaging technology. The surface spectrum and morphology information of heated steel structure can be collected through standard spectral imaging equipment in 400-1000nm, then establishing the connections between the highest temperature or time and its feature of spectral imaging data. More than 150 groups of steel plate samples were prepared in various temperature including 100 ℃ to 1200 °C by muffle furnace and stacking fire in the experiment. The spectral imaging data can be obtained and imported to the classification and recognition algorithm, 90%of the samples conditions can be identified recognized accurately. The results indicate that spectral imaging technology can effectively assist in the development of fire investigation work by quickly and intelligently identifying fire traces, and has good application prospects in the field of fire investigation.
Steel structure has been widely used in modern buildings due to their excellent building performance, but their poor fire resistance requires special coating which can resistant fire for protection. With the increasing demand for fire-resistant coatings for steel structure, fire-resistant coatings with different fire-retardant mechanisms have emerged one after another. The market complexity has increased, and there is a significant difference in fire resistance performance among products, requiring real-time and effective supervision by the relevant department. However, due to the wide variety of fire-resistant coatings and the complexity of standard detecting methods, it brings difficulties for fire supervision and on-site inspection of fire-resistant coatings. The research proposed an efficient method based spectral analysis to detect the fire-resistant coatings with different fire resistive mechanisms, including both intumescent coatings and non-intumescent coating. Principal component analysis is used to quickly identify the spectral consistency of different coatings. The experiment selected samples of fire-resistant coatings with excellent performance verified by standard detection methods and ordinary one for visible to shortwave infrared 400-2500nm spectral collection and spectral feature analysis. The experimental result indicates that samples of high-performance fireproof coatings have high consistency in spectral feature. Through intelligent recognition algorithms, coating samples with unsatisfactory performance can be quickly and accurately detected. The research has shown that the intelligent spectral imaging technology is expected to provide a reliable basis for rapid on-site identification of fire resistive coatings for steel structure.
Spectral imaging technology is a information acquisition method that can obtain three-dimensional information of the object, which is widely used in remote sensing, material classification and other fields. Spectral imaging micro-system is a compact, low-cost spectral imaging system based on structure of single-chip integration. It is suitable for utilizing on unmanned airborne and has a good prospect in the field of geological exploration and mineral classification. Coal gangue is a kind of gray rock associated with coal, which is also a kind of solid waste in mining industry. Due to demand of environmental monitoring and effective mining, it is urgent to effectively monitor the distribution of coal in minerals. This study carries out the experiment on classification of coal and coal gangue samples with the spectral imaging micro-system of visible-near-infrared spectrum. And experimental results show that 80% of the pixels in the sample can be accurately classified.
Spectral imaging technology is a non-contact detection technology that combines spectrum technology and imaging technology. It can be obtained more information than traditional RGB imaging. The algae outbreak is one of the more serious environmental problems in Yuqiao Reservoir and the algae seriously affects the daily water safety of people and destroys the ecosystem of the reservoir. In this paper, the UAV spectral imaging system based on FP cavity spectral imaging chip is used to monitor the reservoir area. Through the spectral data processing, the distribution of algae in the reservoir area was obtained.
Meat freshness degree is an important parameter while evaluating the quality and the price of the meat. Traditional method needs too long time to evaluate the meat on line in the meat slaughterhouse. To improve this situation, a system with potentiality of integrating miniaturized spectral camera is developed and tested in the practical situation. The results show that the system does not only improve the efficiency of meat grade evaluation procedure but also improve the accuracy compared with the method applied now.
Multispectral imaging technology is an advanced imaging method to acquire both spatial and spectral information in the image. Miniaturized spectral imaging system based on CMOS sensors integrated with Fabry-Perot interferometer makes it convenience for people to get the multispectral image outdoor because of the low-cost, high speed, especially compact size. However, the parasitic effects of the system makes the spectrum deviating from the ideal, applications for classification and recognition based on spectral information are limited. Hence, spectrum reconstruction method is applied to calibrate objects toward domain-specific with priori knowledge. And the experiment with test samples independent of the training data shows that, the MSE improves more than 20% compared with the raw data measured by the multispectral camera.
Spectral imaging is a technique which enables the ability of detecting the target by un-contact measurement with both imaging and spectral feature in every pixel inside the image. In this way, spectral imaging device is able to collect more detailed information than traditional RGB camera and hence classify the objects into a more precise category. Environment surveillance is a vital step in the environment protection in the terms of advance warning, pollution area measurement, pollution identification, emergency response and response effectiveness evaluation. In this case, a measurement with a large surveillance area and the capacity of recognizing the target object will be preferred in this application especially for the country with large land area such as China. In this paper, a solution based on the miniaturized spectral imaging system is proposed and the practical experiment has been performed. The result shows that the proposed system is able to be installed in a small UAV and work in an altitude up to 1.7 km.
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.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
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.