Vehicle localisation using a single camera

Lots of rear end collisions due to driver inattention have been identified as a major automotive safety issue. A short advance warning can reduce the number and severity of the rear end collisions. This paper describes a Forward Collision Warning (FCW) system based on monocular vision, and presents a new vehicle detection method: appearance-based hypothesis generation, template tracking-based hypothesis verification which can remove false positive detections and automatic image matting for detection refinement. The FCW system uses time to collision (TTC) to trigger the warning.In order to compute time to collision (TTC), firstly, haar and adaboost algorithm is utilized to detect the vehicle; Secondly, we use simplified Lucas-Kanade algorithm and virtual edge to remove false positive detection and use automatic image matting to do detection refinement; Thirdly, hierarchical tracking system is introduced for vehicle tracking; Camera calibration is utilized to get the headway distance and TTC at last. The use of a single low cost camera results in an affordable system which is simple to install. The FCW system has been tested in outdoor environment, showing robust and accurate performance.

[1]  O. Mano,et al.  Forward collision warning with a single camera , 2004, IEEE Intelligent Vehicles Symposium, 2004.

[2]  Christopher Rasmussen,et al.  Grouping dominant orientations for ill-structured road following , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[3]  Larry S. Davis,et al.  Multiple vehicle detection and tracking in hard real-time , 1996, Proceedings of Conference on Intelligent Vehicles.

[4]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[5]  Dani Lischinski,et al.  A Closed-Form Solution to Natural Image Matting , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  J Hong,et al.  GRAY LEVEL-GRADIENT COOCCURRENCE MATRIX TEXTURE ANALYSIS METHOD , 1984 .

[7]  Michael F. Cohen,et al.  Image and Video Matting: A Survey , 2007, Found. Trends Comput. Graph. Vis..

[8]  G. Bebis,et al.  On-road vehicle detection using optical sensors: a review , 2004, Proceedings. The 7th International IEEE Conference on Intelligent Transportation Systems (IEEE Cat. No.04TH8749).

[9]  Frans C. A. Groen,et al.  Vehicle detection with a mobile camera: spotting midrange, distant, and passing cars , 2005, IEEE Robotics & Automation Magazine.

[10]  Takahiro Ishikawa,et al.  The template update problem , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Ryuzo Okada,et al.  Obstacle detection using projective invariant and vanishing lines , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[12]  Luis Salgado,et al.  Real-time vehicle detection and tracking based on perspective and non-perspective space cooperation , 2009, Electronic Imaging.

[13]  Simon Baker,et al.  Lucas-Kanade 20 Years On: A Unifying Framework , 2004, International Journal of Computer Vision.

[14]  Tom Duff,et al.  Compositing digital images , 1984, SIGGRAPH.

[15]  Zehang Sun,et al.  A real-time precrash vehicle detection system , 2002, Sixth IEEE Workshop on Applications of Computer Vision, 2002. (WACV 2002). Proceedings..

[16]  Chris J. Harris,et al.  VEHICLE DETECTION AND RECOGNITION IN GREYSCALE IMAGERY , 1995 .

[17]  G. F. McLean,et al.  Vanishing Point Detection by Line Clustering , 1995, IEEE Trans. Pattern Anal. Mach. Intell..