- 期刊
YOLOv7-based Research on Foreign Object Intrusion Detection on Tracks
摘要
YOLOv7 has the highest accuracy of any known real-time target detector, improving the state of the art in target detection methods. In order to achieve real-time and accurate detection of track foreign objects and reduce cumbersome labour costs, this paper constructs a dataset of track foreign objects, applies the yolov7 network model based on the pytorch framework to train the dataset, and detects the track foreign object test images by the training model with an accuracy of 0.816 and a recall of 0.667, mAP@0.5 for 0.657, showing from the experimental results that the dataset achieved better results in terms of recognition accuracy.
參考文獻
Kuang Benfa,Yuan Jie,Liu Qiang. A robust RGB-D SLAM based on multiple geometric features and semantic segmentation in dynamic environments[J]. Measurement Science and Technology,2023,34(1).
Chollet F.Xception:Deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.2017:1251-1258.
Niu Hongxia X.,Hou Tao. Fast detection study of foreign object intrusion on railway track[J]. Archives of Transport,2018,47(3).
Sensor Research; Researchers from Changsha University of Science and Technology Describe Findings in Sensor Research (Foreign Object Intrusion Detection On Metro Track Using Commodity Wifi Devices With the Fast Phase Calibration Algorithm Dagger)[J]. Journal of Mathematics,2020.
Hussain Muhammad,AlAqrabi Hussain,Munawar Muhammad,Hill Richard,Alsboui Tariq. Domain Feature Mapping with YOLOv7 for Automated Edge-Based Pallet Racking Inspections[J]. Sensors,2022, 22(18).