Title :
A Novel Particle Filter Based Background Subtraction Method
Author :
Zhang, Enwei ; Chen, Feng ; Zhang, Weidong
Author_Institution :
Dept. of Autom., Tsinghua Univ., Beijing
Abstract :
Background subtraction is a typical method to segment foreground from background in motion analysis. Most of background subtraction methods base on static backgrounds. We propose a particle filter based method to estimate dynamic backgrounds, such as water waves, foliage, smoke, etc. We consider the dynamic scenes vary under some rules and use particle filter to track background. Experiments show our results are promising, compared with other background subtraction methods
Keywords :
image motion analysis; image sampling; image segmentation; image texture; iterative methods; particle filtering (numerical methods); background subtraction; dynamic background; dynamic scenes; foliage; image segmentation; motion analysis; particle filter; smoke; water waves; Automation; Autoregressive processes; Gaussian noise; Layout; Motion analysis; Nonlinear dynamical systems; Object segmentation; Particle filters; Particle tracking; Underwater tracking;
Conference_Titel :
Computational Intelligence and Security, 2006 International Conference on
Conference_Location :
Guangzhou
Print_ISBN :
1-4244-0605-6
Electronic_ISBN :
1-4244-0605-6
DOI :
10.1109/ICCIAS.2006.295382
