An improved particle filter based on cuckoo search for visual tracking
作者单位:School of Electrical & Electronic Engineering Shandong University of Technology College of Electronics and Information Engineering Sichuan University
会议名称:《第30届中国控制与决策会议》
会议日期:2018年
学科分类:080902[工学-电路与系统] 0809[工学-电子科学与技术(可授工学、理学学位)] 08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程]
基 金:supported by Nature Science Foundation of Shandong Province under Grant ZR2016FL14 China Postdoctoral Science Foundation under Grant 2017M612306 National Nature Science Foundation of China under Grant 61601266 Nature Science Foundation of Shandong Province under Grant ZR2016FP04
关 键 词:Particle filter Particle degradation Cuckoo search Target tracking
摘 要:Particle filter(PF) has been proven to be a powerful tool to solve visual tracking problem. However, the problem of sample impoverishment is a constraint of PF. To solve this problem, a cuckoo search-based particle filter is proposed. The particles in PF are optimized using cuckoo search. The meaningful particles are increased and can approximate the true state of the target more accurately. Experiments on visual tracking show that the proposed algorithm outperforms the standard particle filter in solve the visual tracking problems with various challenging conditions.