Implementation of embedded Linux platform based vehicle feature extractor for ITS

This paper propose the vision based vehicle feature extractor which can be operated on the low performance embedded linux platform. In addition, we implement an intelligent transportation system applying the OGC SWE framework for integrating with conventional analog CCTV based traffic surveillance system. The proposed method is comprised of the moving vehicle detection, license plate recognition and vehicle color detection stage. The fast operation is possible by limiting the feature extraction area in the each step. The experimental result feature extractor can be effectually integrated with the conventional system. It provides the exact vehicle identification information.