Development of a Lane Mark Recognition System for a Lane Keeping Assist System

A lane mark recognition system that processes images taken by a camera for a Lane Keeping Assist System (LKAS) has been developed. This system uses a new algorithm that is a combination or Hough transformation and the process of extracting positive and negative edges which are features of lane marks. By employing this algorithm, we can remove noise and execute high speed image processing with a special chip. As a result, we can process images in real time even in such complex situation as a parting road or double broken line conditions. In experiments on real images, we were able to verify a high rate of recognition and accuracy of output data sufficient for steering control.