With the development of optical communication, optical fiber sensing technology has come into people's view. Among them, the phase sensitive optical time-domain reflectometer φ-OTDR because of its high detection accuracy, good dynamic performance and has been widely studied. In this paper, a new signal processing algorithm is proposed to reduce the computational resources needed for the disturbance signal processing, so as to improve the detection speed of the system.
With the further construction and development of smart cities, the disadvantages of existing environmental information collection solutions in terms of cost, power consumption and signal coverage have become more and more obvious. In order to realize the basic functions of the environment information acquisition system of the Internet of Things (IoT), this paper develops the terminal node of the system based on embedded system hardware, and uses the method of combining Wireless Local Area Network (WLAN) technology and Low Power Wide Area (LPWA) technology to transmit data. IoT cloud platform adopts Ali Cloud City IoT platform, and makes Web interface for data visualization. The system has high reliability and stability, and can realize the basic function of long-term continuous environmental information collection. In addition, its cost-effective performance and strong scalability make it of certain promotion value, which can provide a new idea for the combination of environmental information collection and IoT.
The use of X-Ray rays to detect BGA solder joint bubbles and accurately segment the bubble area has always been a hot topic of research. In this paper, a dynamic enhancement algorithm is used to preprocess the background interference image to reduce the interference of complex background on the bubble segmentation result. The bubble is segmented by the threshold segmentation algorithm, and the segmentation accuracy of the enhanced data graph is about 23.6% higher than that of the original graph, and the area of the mis-segmented area is reduced by about 18.2%, and the segmentation accuracy is increased by 11%. It can be proved that the algorithm has better adaptability and segmentation accuracy in the context of interference.
A fingerprint image classification method based on improved Res Net was proposed through experimental research. We take the existing fingerprint data set as object, expand the number of sample images by combining data enhancement technology, finally four kinds of labels are classified according to the global characteristics of fingerprints. The new net structure was built based on convolutional neural network, and iterative optimization was carried out by SGD. Classification accuracy and training rounds were evaluated by comparing the experimental results of different networks. To verify the applicability of the proposed method, 11824 sample images were composed of four kinds of data for training and testing. The results show that after 100 times of iterative training, the accuracy of fingerprint image classification for four kinds of data is above 95.7%, and the highest is 97.4%. This method supports high precision classification of mixed fingerprint images and has good practical.
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.