Improved Vision Based Driving Space Analysis by Condensation of Image Data

Abstract:

Sophisticated ADAS (Advanced Driver Assistance Systems) use vision based methods for detection and keeping track of ahead driving cars. With thus acquired data it is possible to implement e.g. following up functions to automatically keep equal distance to ahead driving vehicles or avoid collisions with obstacles ahead. Known vision based methods for detection and tracking of vehicles use its underneath shadow on the road. The main drawbacks of those methods are the detection and identification of a shadow belonging to a vehicle is neither reliable nor robust, and the thereto required processing of the camera images is very expensive concerning processing time. To improve reliability and detectability we propose here to use an approach which is different from the known methods a nonparametric one; to improve processing speed we propose to apply diversity-sampling to condense the image data before processing it.

• Periodical: Advanced Materials Research (Volumes 605-607)
• Pages: 2260-2264
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.605-607.2260
• Online since: December 2012
• Authors: Yan Fen Mao, Hans Wiedmann, Ming Chen
• Keywords: Diversity-Sampling, Driver Assistance, Nonparametric Estimation, Vehicle Detection

References

[1] M. Lu, K. Wevers, R.V.D. Heijden, Technical Feasibility of Advanced Driver Assistance Systems for Road Traffic Safety, Transportation Planning and Technology, 28 (2005) 167-187.

DOI: 10.1080/03081060500120282

[2] C. Maag, D. Mühlbacher, C. Mark, H. Krüger, Studying Effects of Advanced Driver Assistance Systems on Individual and Group Level using Multi-driver simulation. 2011 IEEE Intelligent Vehicles Symposium, Baden-Baden, Germany (2011) 589-594.

DOI: 10.1109/ivs.2011.5940419

[3] Z. Sun, G. Bebis, and R. Miller, On-Road Vehicle Detection: A Review. IEEE Trans. Pattern Analysis and Machine Intelligence, 28 (2006) 694-711.

DOI: 10.1109/tpami.2006.104

[4] H. Mori and N.M. Charkari, Shadow and Rhythm as Sign Patterns of Obstacle Detection. In: Int'l Symp. Industrial Electronics, (1993) 271-277.

DOI: 10.1109/isie.1993.268796

[5] C. Tzomakas and W. Seelen, Vehicle Detection in Traffic Scenes Using Shadows, Technical Report 98-06, Institut für Neuroinformatic, Ruhr-Universität, Bochum, Germany (1998)1-8.

[6] M. Rosenblatt, Remarks on Some Nonparametric Estimates of a Density Function, Annals of Mathematical Statistics, 27 (1956) 832-837.

DOI: 10.1214/aoms/1177728190

[7] A.M. Elgammal, Efficient Nonparametric Kernel Density Estimation for Real Time Computer Vision, Ph.D. Dissertation, University of Maryland (2002).

[8] Y. Mao, P. Shi, Diversity sampling based kernel density estimation for background modeling, Journal of Shanghai University (English Edition), 9 (2006) 506-509.

DOI: 10.1007/s11741-005-0008-z