Research of Traffic Sign Recognition Technology Based on Machine Vision

Abstract

In this paper, the technology of traffic sign recognition based on machine vision is studied, and the recognition results are analyzed. Firstly, aiming at the problem of image color distortion, the input image is grayed, and the OTSU method is applied to image segmentation to extract the marking target. Then, aiming at the imbalance of image data, the OCR function is applied to train and recognize the target. Finally, the test result shows that traffic signs can be accurately identified in different environments and have good engineering application.