In recent years, railway accidents caused by railway perimeter disasters have occurred frequently. The development of deep learning target detection algorithm and the improvement of the picture quality of railway perimeter cameras make it possible to use video images to monitor railway perimeter disasters. Based on this, this paper proposes a railway perimeter disaster image monitoring method based on YOLOV4_Lite, aiming to replace the manual inspection with high risk.
In this paper, the YOLOV4_Lite model was trained by using the produced disaster data set, and the training results were compared with the YOLOV4 model. The results show that the number of parameters of YOLOV4_Lite algorithm is about 20% of that of YOLOV4 model, the processing speed of YOLOV4_Lite algorithm is 3.2 times of that of YOLOV4 model, and the detection accuracy of perimeter disasters is also slightly improved. In contrast, the YOLOV4_Lite model is more suitable for railway perimeter disaster monitoring scenarios.
Railway Perimeter Disaster Image Monitoring Method Based on YOLOV4_Lite
Lect. Notes Electrical Eng.
International Conference on Electrical and Information Technologies for Rail Transportation ; 2021 October 21, 2021 - October 23, 2021
Proceedings of the 5th International Conference on Electrical Engineering and Information Technologies for Rail Transportation (EITRT) 2021 ; Kapitel : 43 ; 380-387
23.02.2022
8 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Railway Perimeter Disaster Image Monitoring Method Based on YOLOV4_Lite
British Library Conference Proceedings | 2022
|PERIMETER MONITORING APPARATUS AND PERIMETER MONITORING METHOD
Europäisches Patentamt | 2015
|VEHICLE PERIMETER MONITORING DEVICE AND VEHICLE PERIMETER MONITORING METHOD
Europäisches Patentamt | 2021
|PERIMETER MONITORING DEVICE, PERIMETER MANAGEMENT METHOD, AND PERIMETER MANAGEMENT PROGRAM
Europäisches Patentamt | 2022
|