Traffic sign recognition by the use of speed up robust features

Variations in perspective, illumination, occlusion, motion blur, and weatherworn degeneration of signs could all be crucial in identifying road signs. The goal of this research is to evaluate the image processing technique’s performance in detecting and recognizing road signs, as well as determine the optimum threshold value range for doing so. The Speed Up Robust Features (SURF) detector was tested in the current work to detect and recognize road signs through Bagdad’s streets under various speeds and threshold values. The importance of the threshold’s value was highlighted here to occupy an accurate detection and hence recognize road signs at final. The optimum threshold value for best detection resulted usually in the range (20-25) for all speed signs except for the cross sign. The latter recorded its highest precision value at the five threshold values. On the other side, the highest precision value (i.e. 0.86, 0.76, and 0.059) resulted for speed sign 60 followed by 40 and 80-speed signs respectively.