A lane mark recognition system that processes images taken by a camera for a Lane Keeping Assist System (LKAS) has been developed. The following conclusions were reached concerning the development of a new lane mark recognition algorithm that combines Hough transformation with extraction of positive and negative edges, which are used as characteristic of lane marks: (1) Eliminating extraneous linear components (noise) and using customized computer chips to repeat simple calculations speeds up processing and made possible the development of a lane mark recognition system that can recognize, in real time, double broken line lane marks and lane marks in turnouts and other complex situations on actual expressways. (2) Testing of the system on images of an actual expressway yielded a high rate of recognition and showed position accuracy to also be of a level sufficient for steering control.
Development of a lane mark recognition system for a lane keeping assist system
Entwicklung eines Fahrbahnmarkierungs-Erkennungssystems für ein Spurhaltungs-Assistenzsystem
2003
9 Seiten, 18 Bilder, 1 Tabelle, 5 Quellen
Conference paper
English
Development of a Lane Mark Recognition System for a Lane Keeping Assist System
SAE Technical Papers | 2003
|Development of a Lane Mark Recognition System for a Lane Keeping Assist System
British Library Conference Proceedings | 2003
|Development of a lane mark recognition system for a lane keeping assist system
Automotive engineering | 2003
|LANE KEEPING ASSIST DEVICE AND LANE KEEPING ASSIST METHOD
European Patent Office | 2020
|Lane keeping assist device and lane keeping assist method
European Patent Office | 2017
|