A Rear Vehicle Location Algorithm for Lane Change Assist

A monocular vision based location algorithm is presented to detect and track rear vehicles for lane change assist. The algorithm uses the shadow underneath the vehicle to extract regions of interest (ROI), and then locates vehicles with the combination of symmetry, contour and the shadow underneath a vehicle. Three symmetry criterions are proposed to compute the vehicle symmetry axis: contour symmetry, gray symmetry and S component symmetry in HSV color space. Experimental results have shown better accuracy and robustness in different road and environmental conditions. 1 General Instructions In recent years, because of the increasing of vehicles traffic problems have become more and more serious worldwide. Intelligent driver assistance is an area of active research, and vehicle detection and tracking based on vision information are particularly interesting for their low cost [1,2,3]. The vision information is often obtained with monocular camera or stereo camera. In this paper, a nove...

• Type: Conference
• Year: 2007
• Where: MVA
• Authors: Wei Liu, Chunyan Song, Pengyu Fu, Nan Wang, Huai Yuan
• Keywords: Computer Vision, MVA 2007, Symmetry, Vehicle, Vehicle Symmetry