Research on Traffic Sign Detection and Recognition Algorithm Based on Convolutional Neural Network

Abstract

∗With the continuous progress of science and technology in China, unmanned driving technology has been continuously developed. The use of deep learning for traffic sign recognition has a strong capability of feature representation [1], which is the most popular method at present. In this paper, the convolutional neural network algorithm is used to detect and classify traffic signs based on the German traffic sign data set, and the algorithm is verified experimentally. The algorithm preprocesses the data set by means of color image to grayscale image, histogram equalization and other methods, and then continuously optimizes the neural network model. The experimental results show that the accuracy rate can reach 96.39%.