交通标志自动检测与识别算法研究-电路与系统专业论文.docxVIP

万方数据 万方数据 Abstract Road traffic problem has been a subject of concern. Traffic signs, as part of the road traffic, plays a key role in providing information, regulating the behavior of the driver, ensuring the smooth flow and security of traffic. This paper overview the research status of traffic sign detection and recognition, and design the basic framework of the system. The main work of the full text is as follow. For the noise and motion blur generated during image acquisition, this paper experiment with different ways of filtering, and choose better median filtering and Wiener filter for image pre-processing operations. This paper combine the image segmentation based on RGB and HSV color space with image morphological operations and binary image logic operations, and the experimental results show that the image detection on HSV color space has better effect. To the shape-based traffic sign detection, using the method of random Hough transform to detect circular sign, and using the contour approximation method based on Douglas-Peucker algorithm for rectangular, triangular and octagonal traffic sign detection. For the traffic sign recognition, this paper combine the SURF feature extraction and LSH algorithm with the RANSAC algorithm. At last, the experimental comparison show that the above method can achieve better recognition effect. Finally, the full text is summarized and prospected. Key words：traffic sign detection; traffic sign recognition; random Hough transform; SURF; LSH II 目录 摘 要.................................................................................................................................. I Abstract............................................................................................................................. II 第 1 章 绪论..................................................................................................................... 1 1.1 课题背景与意义..........................................................................