Traffic Sign Detection Algorithm Based On Improved YOLOv4

Abstract This paper proposes a traffic sign detection algorithm using improved lightweight YOLOv4 in order to solve the problems of large parameters quantity, poor real-time performance, and low accuracy for unmanned driving system. Firstly, we use MobileNetV2 network replace original feature extraction improve speed by reducing number model parameters. Secondly, add an attention mechanism backbone effectively extract information. Trim layer prediction part residual structure, so that reuse enhancement can avoid gradient disappearing when is deeper. The size reduced 79.5%. experimental results show average 2.5% lower than YOLOV4 network, 58.8% higher network.