DocumentCode :
1840113
Title :
An Improved Particle Filter Tracking Algorithm Based on Motion and Appearance Features
Author :
LiChun Wang ; Lanxiao Li ; Dehui Kong
Author_Institution :
Beijing Key Lab. of Multimedia & Intell. Software Technol., Beijing Univ. of Technol., Beijing, China
fYear :
2013
fDate :
21-23 June 2013
Firstpage :
110
Lastpage :
113
Abstract :
Particle filter (PF) has proven successfully for nonlinear and non-Gaussian estimate problems, but its degeneracy will influence the results of tracking. Therefore in the paper, the optical flow algorithm is utilized to generate the proposal distribution of particle filter. With the velocity message which is estimated by optical flow algorithm, the particles could be generated in a right direction. So that, more particles are given a relatively greater weight, which improves the particle degradation. This article proposes the integration of color histogram and wavelet moment into tracking algorithm for improving the veracity of object tracking. An improved particle filter tracking algorithm based on optical flow, color histogram and wavelet moment is proposed in this paper.
Keywords :
feature extraction; image colour analysis; image motion analysis; image sequences; object tracking; particle filtering (numerical methods); statistical analysis; wavelet transforms; appearance feature; color histogram; improved particle filter tracking algorithm; motion feature; nonlinear nonGaussian estimation problems; object tracking veracity; optical flow algorithm; particle degradation; particle filter distribution; velocity message; wavelet moment; Computer vision; Image color analysis; Image motion analysis; Optical filters; Optical imaging; Particle filters; Target tracking; Human motion tracking; Optical flow; Particle filter; Wavelet moment invariant;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational and Information Sciences (ICCIS), 2013 Fifth International Conference on
Conference_Location :
Shiyang
Type :
conf
DOI :
10.1109/ICCIS.2013.37
Filename :
6642952
Link To Document :
بازگشت