Title: Object tracking with improved firefly algorithm
Authors: Li Lv; Tanghuai Fan; Qi Li; Zhen Sun; Lizhong Xu
Addresses: College of Computer and Information, Hohai University, No. 8 Focheng West Road, JiangNing District, Nanjing, JiangSu Province 210098, China; Nanchang Institute of Technology, School of Information Engineering, No. 289 Tianxiang Road, Hi tech Development Zone, Nanchang, Jiangxi Province 330099, China ' Nanchang Institute of Technology, School of Information Engineering, No. 289 Tianxiang Road, Hi tech Development Zone, Nanchang, Jiangxi Province 330099, China ' College of Computer and Information, Hohai University, No. 8 Focheng West Road, JiangNing District, Nanjing, JiangSu Province 210098, China ' College of Computer and Information, Hohai University, No. 8 Focheng West Road, JiangNing District, Nanjing, JiangSu Province 210098, China ' College of Computer and Information, Hohai University, No. 8 Focheng West Road, JiangNing District, Nanjing, JiangSu Province 210098, China
Abstract: Resampling particle filter is easy to lead to particle impoverishment, and requires a large number of particles for state estimation. The standard firefly algorithm loses the tracking target easily when optimising particle filter. Therefore, this paper proposes target tracking with improved firefly algorithm. The proposed method utilises the movement trend of the target, combines the learning principles of the firefly and the movement characteristics of the tracking target, and designs a new position updating formula of the firefly. In the proposed method, the particles gradually move to the high likelihood region through iterative optimisation. Thus the estimated target state is closer to the true value, and the accuracy of targets is improved. The experiment results show that the tracking performance and the stability of the improved firefly algorithm are better than that of the standard firefly algorithm, and the particle number has certain influences on the experimental results.
Keywords: firefly algorithm; particle filter; object tracking.
DOI: 10.1504/IJCSM.2018.093158
International Journal of Computing Science and Mathematics, 2018 Vol.9 No.3, pp.219 - 231
Received: 03 Nov 2017
Accepted: 22 Jan 2018
Published online: 11 Jul 2018 *