In recent years, high-speed railways have developed rapidly in China, but the detection of perimeter intrusion object is still difficult. Based on a brief introduction of the characteristics of small object detection, this article combines the development status of object detection algorithms based on convolutional neural networks, and relies on self-made high-speed railway datasets to compare and analyze some representative object detection algorithms. Experiments show that among the algorithms selected in this article, YOLOv5 has the best overall performance, reaching 68.1% mAP and 31 FPS. PP-YOLOv2 has the highest detection accuracy, YOLOv4 is the fastest one, and YOLOv5 has the lowest missing rate. The experimental results have good reference, and have a certain contribution to the detection of small object in the perimeter of high-speed railways.
Comparison on Detection Algorithms of Small Object Intrusion on High-Speed Railway
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 : 25 ; 226-234
2022-02-19
9 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Comparison on Detection Algorithms of Small Object Intrusion on High-Speed Railway
British Library Conference Proceedings | 2022
|Intrusion Detection for High-Speed Railway Perimeter Obstacle
British Library Conference Proceedings | 2018
|Application of Target Detection Algorithms in Railway Intrusion
Springer Verlag | 2020
|Application of Target Detection Algorithms in Railway Intrusion
British Library Conference Proceedings | 2020
|Application of Target Detection Algorithms in Railway Intrusion
TIBKAT | 2020
|