Translator Disclaimer
1 December 2009 Vision-based vehicle detection and tracking algorithm design
Author Affiliations +
The vision-based vehicle detection in front of an ego-vehicle is regarded as promising for driver assistance as well as for autonomous vehicle guidance. The feasibility of vehicle detection in a passenger car requires accurate and robust sensing performance. A multivehicle detection system based on stereo vision has been developed for better accuracy and robustness. This system utilizes morphological filter, feature detector, template matching, and epipolar constraint techniques in order to detect the corresponding pairs of vehicles. After the initial detection, the system executes the tracking algorithm for the vehicles. The proposed system can detect front vehicles such as the leading vehicle and side-lane vehicles. The position parameters of the vehicles located in front are obtained based on the detection information. The proposed vehicle detection system is implemented on a passenger car, and its performance is verified experimentally.
©(2009) Society of Photo-Optical Instrumentation Engineers (SPIE)
Junyeon Hwang, Kunsoo Huh, and Donghwi Lee "Vision-based vehicle detection and tracking algorithm design," Optical Engineering 48(12), 127201 (1 December 2009).
Published: 1 December 2009


Depth estimation from disparity of stereo images
Proceedings of SPIE (November 01 1990)
Improved error diffusion incorporating a visual model
Proceedings of SPIE (July 17 1998)
Color texture models for machine vision
Proceedings of SPIE (September 08 1993)
Stereo correspondence using geometric relational matching
Proceedings of SPIE (December 28 1998)

Back to Top