Abstract: A new statistical recognition method of the road signs was proposed. In view of the basic characteristics of the road signs,and through a comprehensive analysis to turn road signs, we found that the image global features are more likely to be detected and not liable to be disturbed by noise and local distortion. An approach to the image feature selection and extraction based on the global feature was presented,in which the mean gray-scale of the image was taken as the major feature value. In what follows, the statistical recognition method was developed to accomplish these steps:the image was divided into several blocks; projective method with directional feature was adopted; the invariant of the feature marks was established and corresponding invariant values of nine turn road signs was computed. The experimental results show that the statistical approach has a good anti-interference ability by which the road sign in an image with moderate noise can be accurately and quickly recognized.

Key words: Vchiclc engineering, Outdoor mobile robots, Statistical recognition, Road sign extraction, Directional projective feature