Road sign recognition model construction and analysis based on deep residual network

Road sign recognition is an important basis of autonomous driving technology. The rapid development of automatic driving technology has higher requirements for road sign recognition, which has important theoretical and application value. The background of road sign recognition is briefly analyzed, and the network structure of convolutional neural network is introduced, as well as the deep residual network model（ResNet） which has achieved good recognition effect in recent years. An improved ResNet18 network model is proposed, which is trained and tested by using the German road sign data set, and compared with related algorithms. It is proved that the model has higher recognition accuracy and efficiency.