Detection of lateral and longitudinal distances for driver assistant system

In this paper, a new approach to lane and obstacle detection is proposed, using reconstruction of 3D road surface. Based on it, a new method for detecting lateral and longitudinal distances of vehicle is developed. A fast reconstruction of 3D road surface, adaptive modification of parameters, from an input road scene image is presented, which can simplify the design of system without any lose of robustness and flexibility. Lane structure is extracted and obstacles are detected from a road image for determining lateral and longitudinal distances of vehicle, which is used for lateral and longitudinal control of driver assistant system.