{"title":"A Novel Fire Detection Method Based on Deep Neural Network and Image Processing","authors":"Yuning Wang","doi":"10.1109/INSAI56792.2022.00022","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00022","url":null,"abstract":"The development of computer vision and deep neural networks has enabled the accurate classification of objects like fire. Fire can pose a serious threat to humans, so the prevention of fire becomes a major concern for society. In this paper, a trained ResNet-50 model is combined with an RGB-based analysis of fire to achieve higher accuracy. In short, the ResNet-50 model categorizes images quickly, while the RGB model can be adjusted based on the actual environment. They complement each other. In images, the RGB model correctly detects the fire, whereas the ResNet-50 model achieves 96.2% accuracy.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132796233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Energy Efficiency Optimization for DAS Based on Neural Network","authors":"Yifan Liu, Hai‐Ping Wang, Ni Ma","doi":"10.1109/INSAI56792.2022.00054","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00054","url":null,"abstract":"Aiming at the huge energy consumption problem in the communication industry, this paper proposes an optimization algorithm of geometric topology of base station based on neural network to improve the system energy efficiency. The relative position between the base station and the user affects the path loss of signal propagation and the size of interference signal, thus affecting the spectral efficiency and energy efficiency of the system. In this paper, communication simulation experiments are conducted to obtain some location coordinates of randomly distributed BTS and their corresponding system energy efficiency values, which are put into the neural network for training. Finally, the network model of base station location and system energy efficiency is obtained, and the maximum value of system energy efficiency is solved. The experimental results show that the algorithm can improve the system energy efficiency by 10 times, which has achieved the desired goal.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125438151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Night Pedestrian Detection Method Based on Image Fusion","authors":"Yiming Jiang, S. Chai, Bai-wen Zhang","doi":"10.1109/INSAI56792.2022.00025","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00025","url":null,"abstract":"Improving pedestrian detection accuracy under conditions of insufficient light at night is a priority for automatic driving. The clarity of conventional visible images at night cannot be as good as that during the daytime. In contrast, the images captured by infrared cameras are almost unaffected by ambient light changes, but there are also defects of not apparent texture features. Therefore, the pedestrian characteristics can be enhanced by combining the thermal and the visible images, which are not affected by the ambient heat. There have been many types of research on pixel-level image fusion in medical diagnosis, military enhancement and navigation, and it has an excellent application value. In this paper, pixel-level fusion method is applied to the fusion of thermal and visible images, and tests on the pedestrian Dataset CVC-14 acquired in a real-world environment. The YOLOV5 method, which has significant advantages in recognition speed and application flexibility, was used for detection. The experimental results showed that the average recognition accuracy and iteration speed was superior to that of the source image. The performance of maximum fusion, principal component analysis (PCA) fusion based on thermal imaging, and weighted average fusion was outstanding. The method proposed in this paper is not complicated and widely applied, which can be improved by other researchers based on this direction.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126236969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Sponsors: INSAI 2022","authors":"","doi":"10.1109/insai56792.2022.00009","DOIUrl":"https://doi.org/10.1109/insai56792.2022.00009","url":null,"abstract":"","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127620606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research on Mask Wearing Detection of Natural Population Based on Improved YOLOv4","authors":"Qian Zhang, Bingdian Yang, Zhichao Liu","doi":"10.1109/INSAI56792.2022.00012","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00012","url":null,"abstract":"Recently, the domestic COVID-19 epidemic situation has been serious. At present, the most effective epidemic prevention and control measure is still to wear masks. Therefore, setting up automatic detection devices for wearing masks in public places can better help relevant departments to carry out epidemic prevention and control work. Aiming at the problems of the existing mask detection algorithms, such as low accuracy, poor robustness, and inability to meet the real-time requirements of the proposed method, this paper proposes a new mask wearing detection method based on improved YOLOv4. Specifically, first of all, we add the Coordinate Attention Module to the backbone to coordinate feature fusion and representation. Secondly, we conduct a huge number of network structural improvements to enhance the model's performance and robustness. Thirdly, we deploy the K-means clustering algorithm to make the nine anchor boxes more suitable for our NPMD dataset. The extensive experimental results show that the improved YOLOv4 performs better, exceeding the baseline by 4.06% AP with a comparable speed of 64.37 FPS. It can complete a comprehensive and accurate mask wearing detection task in natural scenes.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124675892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Design of License Plate Recognition System Based on OpenCV","authors":"Chenxu Duan, Shiqiang Luo","doi":"10.1109/INSAI56792.2022.00013","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00013","url":null,"abstract":"With the continuous improvement of people's living standards, the number and types of motor vehicles are increasing, and it is more and more difficult to manage these vehicles. The most important part of vehicle management information is the license plate of each motor vehicle. They are the identity cards of motor vehicles, so a license plate recognition system that can accurately identify is indispensable. Manual recognition has been difficult to cope with the growing demand for license plate recognition. The traditional license plate recognition system has been overwhelmed. They can not cope with a variety of cars and their license plates, especially the emergence of new energy license plates with green as the background color, resulting in a significant decrease in the accuracy of the traditional license plate recognition system. Therefore, it is necessary to introduce more accurate and efficient technologies in the license plate recognition system, such as widely used machine learning. In this paper, Python as the building language, based on OpenCV library to build a license plate recognition system. The system mainly uses the display function in CV2 and the Gaussian filter grayscale processing function to complete the image display and denoising grayscale processing. Locating the license plate is based on the results of the previous step to extract the corresponding mathematical and color features to complete, and then segment the license plate area of continuous strings, the use of neural networks based on kears framework to identify a single character. Through the test set test of the license plate recognition system, the accuracy of the system to identify the license plate is 93.33 %.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132602202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optimization Strategy and Case Analysis of 5G NSA Wireless Network","authors":"Mingyan Li","doi":"10.1109/INSAI56792.2022.00016","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00016","url":null,"abstract":"In June 2019, the Ministry of Industry and Information Technology issued 5G commercial licenses to China Telecom, China Mobile, China Unicom and China Radio and Television, which marked the official entry of China into the first year of 5G commercial use and the penetration of 5G into every corner of social life, redefining traditional industries, improving efficiency and reducing costs, which will better change society. In this paper, first of all, the principle of 5G NSA wireless network optimization, the causes of coverage optimization and optimization methods were discussed in detail based on the architecture of 5G NSA wireless network system. Secondly, in view of the common coverage problems in a city, the main problems existing in the network area were judged through the statistics and analysis of the main indicators of the network, and the network optimization was effectively implemented by means of adjusting the downward inclination angle and the azimuth angle of the antenna, and adjusting the power of the reference signal, thereby improving the utilization rate of network resources and the performance of the network, and effectively improving the performance of the network and the user experience. Based on the actual work, the different problems in the optimization of 5G NSA wireless network were analyzed, and the practical solutions were given, which are practical.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121291572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Camera-LiDAR Fusion Based Three-Stages Data Association Framework for 3D Multi-Object Tracking","authors":"Zeguo Fu, Huiliang Shang, Liang Song, Zengwen Li, Changxue Chen","doi":"10.1109/INSAI56792.2022.00037","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00037","url":null,"abstract":"3D multi-object tracking (MOT) ensures safe and efficient motion planning and vehicle navigation and plays an important role in perception systems in autonomous driving. Currently MOT is divided into tracking by detection and end-to-end, but most of them are tracking by detection using only single depth sensor such as LiDAR to detect and track objects. However, LiDAR has the limitation of not being able to obtain information about the appearance of the object due to the lack of pixel information, which can lead to obtaining inaccurate detection results thus leading to erratic tracking results. Therefore, in this paper, we propose a novel 3D MOT framework that combines the unique detection advantages of cameras and LiDAR. To avoid the IDs generated by the early death of the detection of the same object that produced low scores in successive frames, we design a 3D MOT framework with three-stages data association. And we also design a data association metric based on 3D IoU and Mahalanobis distance. The camera-LiDAR fusion-based 3D MOT framework we propose proves its superiority and flexibility by quantitative experiments and ablation experiments.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121346958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Classification of Emotional Stress and Physical Stress Using Electro-Optical Imaging Technology","authors":"Kan Hong","doi":"10.1109/INSAI56792.2022.00027","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00027","url":null,"abstract":"Emotional stress status is normally to be intertwined with physical stress information. It is meaningful to classify stress for real application such as homeland security and health. In this study, the classification algorithms are proposed based on electro-optical imaging system, such as thermal imaging (TI) system and multispectral imaging (MSI) system. Through the proposed model, the classification signals of ES and PS are successfully obtained. Experiments show that the classification result is encouraging, and the accuracy of the proposed algorithm is over 90%. This study can lead to a useful system for the stress classification and real applications. This research can lay a foundation for the application of stress recognition.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121351549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Steel Strip Defect Classification Model Based on Convolutional Neural Network Composites","authors":"Fei Ye, Linjie Bian","doi":"10.1109/INSAI56792.2022.00024","DOIUrl":"https://doi.org/10.1109/INSAI56792.2022.00024","url":null,"abstract":"In actual industrial scenes, the complex and diverse textures of surface flaw of steel strips lead to poor performance and generalization of the classification and recognition tasks of surface flaw of steel strips. In this paper, an end-to-end classification model of surface flaw of steel strips based on convolution neural network-EDESPNet is proposed. Improve the accuracy of steel belt defect classification. The model inputs defect samples into different branch networks for simultaneous feature extraction, and adds a weight distribution network to enhance category-related feature information, promotes model efficiency and feature spread, and rise the model's portrayal ability. The results indicate that the EDESPNet classification model is surpass VGG19, DenseNet, ResNet50, Xception and other models. The classification precision in the NEU-CLS data set is 94.17%, and the classification accuracy rate in the BS4-CLS data set is 72.52%. The EDESPNet classification model proposed in the article has higher experimental evaluation standards on the BS4-CLS data set than other classification models, and it has the highest correct rate in the NEU-CLS public datasets. The results indicate that the EDESPNet classification model has better recognition task effect and good robustness.","PeriodicalId":318264,"journal":{"name":"2022 2nd International Conference on Networking Systems of AI (INSAI)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116002157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}