The Improved Framework for Traffic Sign Recognition Using Guided Image Filtering

Abstract In the lighting conditions such as hazing, raining, and weak condition, accuracy of traffic sign recognition is not very high due to missed detection or incorrect positioning. this article, we propose a (TSR) algorithm based on Faster R-CNN YOLOv5. The road signs were detected from driver’s point view was assisted by satellite images. First, conduct image preprocessing using guided filtering for input remove noises. Second, processed into proposed networks model training testing. Three datasets are employed verify effectiveness method finally. outcomes promising.