Detection of moving targets with a moving camera

Authors:
Dongxiang Zhou;Liangfen Wang;Xuanping Cai;Yunhui Liu
Affiliations:
School of Electronic Science and Engineering, National University of Defense Technology, Changsha, China;School of Electronic Science and Engineering, National University of Defense Technology, Changsha, China;School of Electronic Science and Engineering, National University of Defense Technology, Changsha, China;Dept. of Automation and Computer Aided Eng., The Chinese University of Hong Kong
Venue:
ROBIO'09 Proceedings of the 2009 international conference on Robotics and biomimetics
Year:
2009

Citing 6
Cited 0

Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography

Communications of the ACM
Motion Tracking with an Active Camera

IEEE Transactions on Pattern Analysis and Machine Intelligence
Moving Target Classification and Tracking from Real-time Video

WACV '98 Proceedings of the 4th IEEE Workshop on Applications of Computer Vision (WACV'98)
Distinctive Image Features from Scale-Invariant Keypoints

International Journal of Computer Vision
A Performance Evaluation of Local Descriptors

IEEE Transactions on Pattern Analysis and Machine Intelligence
Detection and Recognition of Moving Objects by Using Motion Invariants

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 01

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we propose a new method of detecting moving objects from a moving camera based on SIFT(The Scale Invariant Feature Transform) features matching and dynamic background modeling. Firstly, feature points are extracted by SIFT algorithm to compute the affine transformation parameters of camera motion, and guided by RANSAC to remove the outliers. We adopt background subtraction approach to detect moving objects, with shadow and ghost removing. The robustness of SIFT Features matching and the validity of picking out outliers by a RANSAC algorithm make the parameters of affine transform model to be computed accurately, and by the background subtraction approach with dynamically-updated background model, foreground objects can be detected perfectly. Experimental results demonstrate that our algorithm can detect moving objects accurately, and keep the integrity of foreground objects, comparing with optical flow method.