Obstacle Detection in Range Image Sequences Using Radial Slope

Abstract The present paper proposes an approach to obstacle detection in range image sequences of traffic scenes using radial slope. Slope is a very useful feature for determining the traversability of an environment and can be computed efficiently in radial direction. Combined with 3D sensory information, a robust detection of obstacles in real-time is possible. Once an obstacle is detected, it can be tracked over time by means of matching along the time axis. Results of our approach are shown for real as well as simulated range image sequences.

[1]  Horst Bunke,et al.  SPATIOTEMPORAL SEGMENTATION OF RANGE IMAGE SEQUENCES INTO PLANAR SURFACES FOR COLLISION AVOIDANCE , 1997 .

[2]  Anil K. Jain,et al.  Range-intensity histogram for segmenting LADAR images , 1992, Pattern Recognit. Lett..

[3]  Heinrich Niemann,et al.  A robust cognitive approach to traffic scene analysis , 1994, Proceedings of 1994 IEEE Workshop on Applications of Computer Vision.

[4]  Takeo Kanade,et al.  The new generation system for the CMU Navlab , 1992 .

[5]  Markus Maurer,et al.  A compact vision system for road vehicle guidance , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[6]  H. Hügli,et al.  Range image segmentation based on function approximation , 1990, ISPRS International Conference on Computer Vision and Remote Sensing.

[7]  R. Highnam,et al.  Model-based image enhancement of far infra-red images , 1995, Proceedings of the Workshop on Physics-Based Modeling in Computer Vision.

[8]  Martial Hebert,et al.  Vision and navigation for the Carnegie-Mellon Navlab , 1988 .

[9]  Jean-Michel Jolion,et al.  Robust Clustering with Applications in Computer Vision , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Miguel Ángel García,et al.  Fast extraction of surface primitives from range images , 1996, Proceedings of 13th International Conference on Pattern Recognition.