Vehicle headlights detection using markov random fields

Authors:
Wei Zhang;Q. M. Jonathan Wu;Guanghui Wang
Affiliations:
Computer Vision and Sensing Systems Laboratory (CVSSL), Department of electrical and computer engineering, University of Windsor, Windsor, Ontario, Canada;Computer Vision and Sensing Systems Laboratory (CVSSL), Department of electrical and computer engineering, University of Windsor, Windsor, Ontario, Canada;Computer Vision and Sensing Systems Laboratory (CVSSL), Department of electrical and computer engineering, University of Windsor, Windsor, Ontario, Canada
Venue:
ACCV'09 Proceedings of the 9th Asian conference on Computer Vision - Volume Part I
Year:
2009

Citing 6
Cited 0

Nighttime Vehicle Detection for Driver Assistance and Autonomous Vehicles

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 01
Vehicle Detection and Counting by Using Headlight Information in the Dark Environment

IIH-MSP '07 Proceedings of the Third International Conference on International Information Hiding and Multimedia Signal Processing (IIH-MSP 2007) - Volume 02
An HMM/MRF-based stochastic framework for robust vehicle tracking

IEEE Transactions on Intelligent Transportation Systems
Multilevel Framework to Detect and Handle Vehicle Occlusion

IEEE Transactions on Intelligent Transportation Systems
Learning, Modeling, and Classification of Vehicle Track Patterns from Live Video

IEEE Transactions on Intelligent Transportation Systems
3-D model-based vehicle tracking

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Vision-based traffic surveillance is an important topic in computer vision. In the night environment, the moving vehicles are commonly detected by their headlights. However, robust headlights detection is obstructed by the strong reflections on the road surface. In this paper, we propose a novel approach for vehicle headlights detection. Firstly, we introduce a Reflection Intensity Map based on the analysis of light attenuation model in neighboring region. Secondly, a Reflection Suppressed Map is obtained by using Laplacian of Gaussian filter. Thirdly, the headlights are detected by incorporating the gray-scale intensity, Reflection Intensity Map, and Reflection Suppressed Map into a Markov random fields framework, which is optimized using Iterated Conditional Modes algorithm. Experimental results on typical scenes show that the proposed method can detect the headlights correctly in the presence of strong reflections. Quantitative evaluations demonstrate that the proposed method outperforms the existing methods.