A Detection Method of Electro-bicycle in Elevators Based on Improved YOLO v4

Currently, electro-bicycles are one of the most convenient and affordable traveling tools. However, when an electro-bicycle enters a crowded elevator, it will not only affect the service life of the elevator, but also it has great potential safety hazards when the battery charges at home. Thus, this study focuses on detecting various electro-bicycle intelligently in elevators in an online manner through improving the YOLO (You Only Look Once) v4 network. The YOLO v4 algorithm has high detection accuracy and speed due to its single stage mode and the priori frame mechanism. Thus, this study applies YOLO v4 as the fundamental network for electro-bicycle detections. The improved YOLO v4 algorithm, named as W_YOLO v4, reconstructs the YOLO v4 feature pyramid and the corresponding backbone feature extraction network. It also integrates the attention mechanism into the residual network of the backbone network to improve the detection accuracy. The experimental results show that the mAP (mean Average Precision) of W_YOLO v4 algorithm is 91.37 with 5000 data, which is 6.5 higher than the mAP of YOLO v4 algorithm, and the network parameters are reduced, such that the model is lighter than the YOLO v4 model.