Driver assistance system (DAS) is an essential component in providing the drivers with safety measures and precaution information in traffic sign recognition (TSR). In this system we propose a novel system for the automatic detection and recognition of traffic sign. As an application of intelligent transportation system (ITS) a method is used to detect and to recognize the information contained in the traffic panels in street level images. The detection and recognition of text and symbol is applied on those images where a traffic panel has been detected, in order to automatically read the text and symbol and through the speaker the data will be announced.in this system we have used a feature based method for traffic sign detection. The images of the traffic sign board are cropped and it is matched with the original image, it will identify the key points in both the images and match between those points to find the similarity. The SURF descriptor is used for key point and matching points.