Pedestrian Detection and Recognition System Based on YOLOv5 Algorithm and ResNet50 Network
Pedestrian detection and recognition has an important application value in traffic management,intelligent monitoring and other fields.Aiming at the problems of low detection accuracy and recognition difficulty faced by the existing pedestrian detection and recognition,this study proposes a model of pedestrian detection and recognition system that integrates the YOLOv5 algorithm of attention mechanism and ResNet50 network.By introducing the SE attention mechanism into the YOLOv5 algorithm Backbone network,richer feature information is captured,thus im-proving the detection accuracy of the model.The pedestrian recognition system uses ResNet50 network to recognize and retrieve the detected cropped images.The experimental results show that the algorithm has high detection accu-racy and can recognize pedestrians in complex scenes,basically meeting the requirements of pedestrian detection and recognition in different scenes.
object detectionYOLOv5image recognitionResNetmechanism of attention
国家科技期刊平台
NSTL
万方数据
