Road sign recognition is one of the core technologies in IntelligentTransport Systems. In the current study, a robust and real-time method ispresented to identify and detect the roads speed signs in road image indifferent situations. In our proposed method, first, the connected componentsare created in the main image using the edge detection and mathematicalmorphology and the location of the road signs extracted by the geometric andcolor data; then the letters are segmented and recognized by Multiclass SupportVector Machine (SVMs) classifiers. Regarding that the geometric and colorfeatures ate properly used in detection the location of the road signs, so itis not sensitive to the distance and noise and has higher speed and efficiency.In the result part, the proposed approach is applied on Iranian road speed signdatabase and the detection and recognition accuracy rate achieved 98.66% and100% respectively.
展开▼