Visual Detection of Moving Vehicles Ahead Based on the Characteristics

Abstract:

Machine vision perception technology is widely used in the vehicle’s active safety system. It provides more immediate and correct information of road and vehicles around, in which inspection of moving vehicle ahead is one of the important items. A method of inspection fused of detection of the shadow under the vehicle and symmetry of the vehicle’s tail is presented in this paper. At first, a region of interest is selected according to the lane lines. Then, the shadow can be detected with grayscale histogram in the region of interest and a suspected area of vehicle is obtained by expanding the shadow with empirical proportion. At last, the vehicle ahead is further affirmed by calculating the symmetry of such characteristic at its tail as grayscale value, taillight and the edges. Experimental results prove that this method can well solve the actual problems of vehicle detection.

• Periodical: Applied Mechanics and Materials (Volume 103)
• Pages: 165-169
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.103.165
• Online since: September 2011
• Authors: Hou Yun Yu, Wei Gong Zhang
• Keywords: Edge Detection, Machine Vision, Shadow Detection, Symmetry Measure, Vehicle Detection

References

[1] Collado J M, Hilario C, Escalera A, et al. Model based vehicle detection for intelligent vehicles[C]. IEEE Intelligent Vehicles Symposium. Parma, Italy, 2004. 572 -577.

DOI: 10.1109/ivs.2004.1336447

[2] Christian Hoffmann, Thao Dang, Christoph Stiller. Vehicle detection fusing 2D visual features[C]. IEEE Intelligent Vehicles Symposium. Parma, Italy, 2004. 280-285.

DOI: 10.1109/ivs.2004.1336395

[3] Gao Lei, Li Chao, Zhu Chengjun, et al. Video vehicle detection algorithm based on edge symmetry[J]. Journal of Beijing University of Aeronautics and Astronautics, 2008, 34(9): 1113-1116. (in Chinese).

[4] Guo Lei, Li Keqiang, Wang Jianqiang, et al. A Feature-based Vehicle Detection Method[J]. Automotive Engineering, 2006, 28(11): 1031-1035. (in Chinese).

[5] H Mori, N M Charkari. Shadow and rhythm as sign patterns of obstacle detection[C]. IEEE International Symposium on Industrial Electronics. Budapest, Hungary, 1993. 271 – 277.

DOI: 10.1109/isie.1993.268796

[6] Gu Boyuan. Study on the Forward Collision Warning System Based on Monocular Vision[D]: [PHD]. JiLin: Jilin University, 2006. (in Chinese).