Road bridge sign detection and classification

In this research the problem of the detection and classification of road bridge sign has been faced, in particular, at butterfly bridge configuration. The first step concerns the robust identification of the rectangular sign, through the optical flow analysis. The second step concerns rectangular sign detection based on searching gray level discontinuity on the image and Hough transform. The classification is based on analysis of surface color on inner part, detection of shape and color of rectangular border around the sign, arrows direction analysis. Our approach can detect sign candidates in presence of complex background