A detection and recognition method for prohibition traffic signs

The automatic detection and recognition of traffic signs have practical importance for intelligent traffic system. A method for some prohibition traffic signs designed for drivers is proposed in this paper. The color information in HSI color space and the symmetry property of circles are used to detect signs, and the Histograms of Oriented Gradients feature and the nearest distance method are used to recognize them. The proposed algorithms are tested on datasets we collected in the real environment, and the experimental results demonstrate their satisfying performance.