抄録
Recently, methods and techniques for Intelligent Transport Systems (ITS) are being researched actively. Many road environment detection systems that assist drivers' judgement and compensate misdetections of road conditions ahead are being proposed. In many of these systems, the recognition of road signs is achieved by resizing images and using template matching. However, resizing an image reduces its quality, and causes a system to fail in matching. In this paper, a recognition method based on vectors is proposed. After calculating vectors of each pixel, outline vector lists are generated by tracing the vectors. Then, signatures of the image are generated from previously detected outline vector lists. Recognition is processed by using string matching technique to match these signatures to template signatures.
