In this paper, a low-cost lane detection algorithm is presented based on video sequences taken from monocular vision-system mounted on a vehicle. Real time lane detection and tracking system is an essential module of an object detection and recognition system for crash warning and avoidance. The basic idea is that the complete processing of each image can be avoided using the knowledge of the position of lane boundaries in the previous ones. The vehicle is supposed moving on a flat and straight road or with slow curvature. Hence, the lane boundaries are considered as straight lines. The proposed lane detection and tracking can be applied only in painted road. The lane boundaries were detected using Radon transform The proposed approach has been tested on off line on video data, and the experimental results have demonstrated a fast and robust system. A parallel programming will obviously reduce the computing time.
Lane boundaries detection and following for marking road
Erkennen der Fahrbahnbegrenzungen und Verfolgung der Straßenmarkierung
MechRob, Mechatronics and Robotics, 2004 ; 883-888
2004
6 Seiten, 6 Bilder, 13 Quellen
Conference paper
English
Stereo Augmented Detection of Lane Marking Boundaries
IEEE | 2015
|LANE MARKING DETECTION SYSTEM AND LANE MARKING DETECTION METHOD
European Patent Office | 2015
|European Patent Office | 2018
|