{"references": ["F. Moutarde, B. Stanciulescu, A. Breheret, \"Real time visual\ndetection of vehicles and pedestrians with new efficient adaboost\nfeatures,\" IEEE International Conference on Intelligent Robots\nSystems (IROS 2008), September 2008.", "H. Fleyeh, \"Color detection and Segmentation for road and Traffice\nsigns\", Proceedings of the 2004 IEEE Cybernetics and Intelligent\nSystems, vol. 2, pp. 809-814, December 2004", "R. Labayrade, D. Aubert, JP. Tarel, \"Real Time Obstacle detection in\nstereovision on Non flat road geometry through \u00d4\u00c7\u00ffV-disparity-\nRepresentation\", Proceedings of IEEE Intelligent Vehicle\nSymposium, vol. 2, pp. 646-651, 2002", "R. Miller, Z. Sun, G. Bebis, \"On Road vehicle Detection\", IEEE\nTransactions on Pattern Analysis and Machine Intelligence, vol. 28,\npp. 694-711, May 2004", "J. Rojas, J. Crisman, \"Vehicle Detection in Color Images\", in Proc.\nIEEE Conference Intelligent Transportation, pp. 403-408, November\n1998", "G. D. Sullivan, K. D. Baker, A. D. Worrall, \"Model based vehicle\ndetection and classification using orthographic approximations\",\nBritish Machine Vision Conference, vol. 15,Issue 8, pp. 649-654,\nAugust 1997", "P. Viola, M. jones, \"Rapid Object Detection using a Boosted Cascade\nof Simple Features\"In Proc. IEEE Conference on Computer Vision\nand Pattern Recognition, vol. 1, pp. 511-518, 2001"]}

This paper presents a robust vehicle detection approach using Haar-like feature. It is possible to get a strong edge feature from this Haar-like feature. Therefore it is very effective to remove the shadow of a vehicle on the road. And we can detect the boundary of vehicles accurately. In the paper, the vehicle detection algorithm can be divided into two main steps. One is hypothesis generation, and the other is hypothesis verification. In the first step, it determines vehicle candidates using features such as a shadow, intensity, and vertical edge. And in the second step, it determines whether the candidate is a vehicle or not by using the symmetry of vehicle edge features. In this research, we can get the detection rate over 15 frames per second on our embedded system.