Particle swarm optimized particle filter and its application in visual tracking

Author

Zhao, Zeng-shun ; Wang, Ji-zhen ; Cheng, Xue-Zhen ; Qi, Yu-Juan

Author_Institution

Coll. of Inf. & Electr. Eng., Shandong Univ. of Sci. & Technol., Qingdao, China

Volume

5

fYear

2010

fDate

10-12 Aug. 2010

Firstpage

2673

Lastpage

2676

Abstract

Particle swarm optimization is proposed to optimize the particle filter in order to Travel out the well-known particle impoverishment and dependency problem. Through particle swarm optimization, particle samples are moved to neighbor higher likelihood areas. In this way, it can obtain more approximate to the true posterior probability density function. Meanwhile, the number of particle sample reducing significantly, make it the better choose to apply the real-time estimation and tracking problem.

Keywords

particle filtering (numerical methods); particle swarm optimisation; probability; dependency problem; higher likelihood areas; particle impoverishment; particle samples; particle swarm optimization; particle swarm optimized particle filter; real-time estimation; tracking problem; true posterior probability density function; visual tracking; Accuracy; Face; Particle filters; Particle swarm optimization; State estimation; Target tracking; particle filter; particle swarm optimization; tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation (ICNC), 2010 Sixth International Conference on

Conference_Location

Yantai, Shandong

Print_ISBN

978-1-4244-5958-2

Type

conf

DOI

10.1109/ICNC.2010.5583901

Filename

5583901