Traffic Sign Recognition Based on Joint Convolutional Neural Network Model

This paper takes the automatic driving technology as the research background, and studies the algorithm and model of traffic sign recognition. Traffic sign recognition is the basis of automatic driving. This paper takes common traffic signs as the research object, and uses the current international standard traffic sign image database GTSRB as the data set of this paper. According to the development status of deep learning and image recognition technology in recent years, this paper analyzes and compares several different image recognition models in ImageNet competition. Based on these experimental results, a new joint network model is proposed, which overcomes some The shortcomings of the existing model, using the model to test on the GTSRB data set, can be found that the model has faster training convergence speed and better recognition accuracy for the GTSRB data set.