Urban Road Traffic Sign Recognition Algorithm Based on Monocular Vision

Intelligent Transportation System (ITS) is a new industry and development greatly in recent years. The advanced electro-communication and computer technology combines detection, control, active warning and other information means. It provides a great guarantee for improving traffic safety and high efficiency. This paper puts forward the traffic rate-limiting sign and Lane Mark recognition method. In this study, four steps are applied to recognize a traffic and Lane Mark sign in an image. The first step, we transform the color space of image from RGB into HIS. Then the image blended HIS and RGB information is segmented. The third step, the text restrain and eliminate the isolated noise point based on eight-neighborhood. Finally, the traffic speed-limit sign and Lane Mark is recognized with template match. Our experiment results, in the out door, show that the method mentioned in the paper is useful and effective.