Vehicle Type Recognition Based on Harris Corner Detector

With the development of computer technology in recent years, video-based detection technology in the intelligent transportation system has been widely used, this article uses a identification method based on the Harris corner detection for rapid and accurate identification. First of all, extract moving goal vehicles of the video, and then extract its Harris corners in gray image, and select the Harris corner of cars, buses and lorries as standard samples. Then the Hausdorff distance are calculated of Harris corner between the standard types and the swatches which need to be recognized. The two whose Hausdorff distance is smallest can be judged as the same type. Experiment results show that the method is accurate, effective and better real-time.