Traffic-signs recognition system based on multi-features

By now, most of the techniques on traffic signs recognition can only recognize those in particular groups, such as triangle signs for warning, circle signs for prohibition and etc but could not tell the exact meaning of every sign. In this paper, we proposed a framework on traffic system recognition system, which consists of two phrase that a segmentation method(FCM) is used to detect the traffic sign while the Content-Based Image Retrieval (CBIR) method is used to match the detect traffic signs to the traffic signs in the database in order to find out the exact meaning of every detected sign.