This paper presents a road lane detection and interpretation algorithm for driver assistance systems (DAS). The algorithm uses an edge filter to extract lane borders to which a straight lane model is fitted. Next, the lane mark type (continuous, discontinuous or merge) is recognized using a Fourier analysis. The line type is essential for a robust DAS. Nevertheless, it has been seldom considered in previous works. The knowledge of the line types of the road helps to guide the search for other lines, to automatically detect the type of the road (one-way, two way or highway), and to tell the difference between allowed and forbidden maneuvers, such as crossing a continuous line. Furthermore, the system is able to auto calibrate, thus easing the process of installation in commercial vehicles.
Detection and classification of road lanes with a frequency analysis
2005-01-01
1166898 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Detection and Classification of Road Lanes with a Frequency Analysis
British Library Conference Proceedings | 2005
|Adaptative road lanes detection and classification
Tema Archiv | 2006
|