Computer Engineering Paper Massimo Bertozzi, Alberto Broggi, Denis Colla, and Alessandra Fascioli, Sensing of Automotive Environments Using Stereo Vision, In Proceedings ISATA - International Symposium on Automotive Technology and Automation, Florence, Italy, 1997. ISATA,
Download the paper in compressed PostScript format


This work presents a vision system aimed to the sensing of the surrounding environment to be used on moving vehicles for the improvement of road safety. Thanks to the processing of pairs of stereo images acquired by a set of cameras installed on the vehicle, it allows to detect both the lane position with respect to the vehicle and generic obstacles on the path (without constraints on symmetry or shape). The process devoted to lane detection is based on a pattern matching technique which relies on the presence of road markings and is performed by the analysis of the left image only. Conversely, both stereo images are used for the obstacle localization task. In order to speed up the processing the system does not perform any 3D world reconstruction, thus reducing obstacle detection to the determination of the free-space in front of the vehicle. The two functionalities share the same underlying approach: the assumption of a flat road in front of the vehicle that allows to remove the perspective effect from the acquired image thus producing a remapped image that represents a bird's~eye view of the road surface. A geometric transform is applied that remaps both stereo images into a new domain (the road domain), in which the processing is extremely simplified. Since both functionalities are based on the processing of images remapped into the same domain, the fusion of the results of the two cooperating processes is straightforward. This integrated solution leads to both an improved quality of the results and to a faster and more efficient processing. The system has been tested both in laboratory and on board of MOB-LAB experimental vehicle and demonstrated to be robust with respect to shadows and vehicle movement.

For more information send email to broggi@CE.UniPR.IT