A Real-Time Recognition Algorithm for Speed Limit Signs Based on Convolutional Neural Networks
Wencai Sun, Wei Li, Shiwu Li · CICTP 2020 · 2020
The speed limit signs recognition algorithm system installed in mobile devices has real-time performance, and can avoid the unavailability of vehicle GPS without a network. Therefore, in this paper, speed limit signs are located by color segmentation, morphological filtering and Hough change, and then the clarity and tidiness of speed limit signs are improved by image enhancement and normalization. After that, the convolution neural network algorithm is described in detail, and the relevant parameters are set. Finally, the performance of the algorithm is verified by the real vehicle test. The results show that the accuracy rate of the real-time recognition and warning system for speed limit signs installed in mobile devices is more than 90% at different speeds and the maximum accuracy rate can reach 94%. Further, through the test process, found that it has strong real-time performance, system stability and recognition robustness. Its performance is more comprehensive than other algorithms.