Application of deep learning in traffic sign detection and recognition system

With the rapid development of economy and science and technology, motor vehicles quickly enter thousands of households, causing traffic congestion has become a common phenomenon in many cities, traffic safety problems are becoming more and more serious. With the national strategic layout and the development of 5G technology, driverless driving will play a huge role in solving this problem. In the unmanned driving system, the detection and recognition of traffic signs is an indispensable part. However, the existing traffic sign detection and recognition technology either has low accuracy or poor real-time performance, which is difficult to meet the real demand. Therefore, based on the above research significance and existing problems, this paper, on the basis of mainstream deep learning object detection algorithms, aims to improve the performance of the original algorithm, improve the accuracy and speed of traffic sign detection, and further study the design and implementation of traffic sign detection and recognition system on the embedded platform, aiming at the target characteristics of traffic signs and the shortcomings of current algorithms.