A fast method for detecting moving vehicles using plane constraint of geometric invariance

Authors:
Dong-Joong Kang;Jong-Eun Ha;Tae-Jung Lho
Affiliations:
Dept. of Mechatronics Eng., Tongmyong University, Busan, Korea;Dept. of Automotive Eng., Seoul National University of Technology, Seoul, Korea;Dept. of Mechatronics Eng., Tongmyong University, Busan, Korea
Venue:
ICCSA'06 Proceedings of the 2006 international conference on Computational Science and Its Applications - Volume Part III
Year:
2006

Citing 8
Cited 0

Symmetry-based recognition of vehicle rears

Pattern Recognition Letters
Geometric invariance in computer vision

Geometric invariance in computer vision
Quantitative planar region detection

International Journal of Computer Vision
SUSAN—A New Approach to Low Level Image Processing

International Journal of Computer Vision
Vision-Based Vehicle Guidance

Vision-Based Vehicle Guidance
Epipolar Geometry in Stereo, Motion, and Object Recognition: A Unified Approach

Epipolar Geometry in Stereo, Motion, and Object Recognition: A Unified Approach
ASSET-2: real-time motion segmentation and shape tracking

ICCV '95 Proceedings of the Fifth International Conference on Computer Vision
Road lane segmentation using dynamic programming for active safety vehicles

Pattern Recognition Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a new method of detecting on-road highway vehicles for active safety vehicle system. We combine a projective invariant technique with motion information to detect overtaking road vehicles. The vehicles are assumed into a set of planes and the invariant technique extracts the plane from the theory that a geometric invariant value defined by five points on a plane is preserved under a projective transform. Harris corners as a salient image point are used to give motion information with the normalized cross correlation centered at these points. A probabilistic criterion without demand of a heuristic factor is defined to test the similarity of invariant values between sequential frames. Because the method is very fast, real-time processing is possible for vehicle detection. Experimental results using images of real road scenes are presented.