Traffic Sign Detection Based on Improved YOLOv3 in Foggy Environment

Abstract Aiming at the problem of poor detection accuracy and inaccurate positioning traffic signs under foggy conditions, this paper proposes an improved YOLOv3 algorithm. Firstly, a data set Chinese in environment is constructed; The dark channel priori algorithm based on guided filtering used to process image with fog, which overcomes quality degradation caused by fog. Mosaic enhancement performed annotated image, speeds up convergence speed network. Increased feature scale loss function network optimized, CIOU as loss, improved. At same time, method transfer learning overcome insufficient samples. enhanced yolov3 proposed has higher shorter time than standard SSD