Introduction - If you have any usage issues, please Google them yourself
This project is a traffic sign recognition system based on Matlab bp neural network in haze weather. There are two steps. One is image defogging, and the dark channel method is used to obtain light transmittance to remove haze. After getting clear pictures, use color method to locate traffic signs. As we all know, traffic signs are basically composed of red, blue and yellow. According to different combinations of RGB, different colors can be located. Because there are errors, we need to use morphological knowledge to remove the error interference area, so as to achieve accurate positioning. After positioning, the color icon is segmented on the basis of the original image. The BP neural network method is used to train and identify the color icon, and the result is obtained. This design is equipped with a GUI visual interface, which is easy to operate. It's a good topic.