The road, traffic isle and obstacle detection algorithm uses elevation maps to represent 3D dense stereo data. It provides the road surface parameters and an occupancy grid with three distinct cell types: road, traffic isles and obstacles. In this paper the authors extend the algorithm with temporal integration at multiple levels: the parameters of the road surface are tracked, the traffic isle cells are filtered based on their temporal persistence between successive frames, and obstacles are separated into static (probably infrastructure) and dynamic obstacles. An enhanced occupancy grid is built, containing road, traffic isle, static obstacle and dynamic obstacle cells. Finally, a global occupancy grid is built (integration of the occupancy grid along several successive frames), that can be used for alignment with external maps.
Temporal integration of occupancy grids detected from dense stereo using an elevation map representation
Zeitliche Berücksichtigung von Belegungsspuren erkannt aus einem dichten Stereobild mittels einer Höhenlagenkarte
2009
6 Seiten, 14 Bilder, 10 Quellen
Conference paper
Storage medium
English
Automotive engineering | 2009
|Robust occupancy detection from stereo images
IEEE | 2004
|