北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2016, Vol. 39 ›› Issue (5): 72-77.doi: 10.13190/j.jbupt.2016.05.015

• 研究报告 • 上一篇    下一篇

一种基于蝙蝠算法的视觉跟踪方法

高明亮1, 尹丽菊1, 蒋俊2, 申晋1   

  1. 1. 山东理工大学 电气与电子工程学院, 山东 淄博 255000;
    2. 西南石油大学 计算机科学学院, 成都 610500
  • 收稿日期:2016-03-31 出版日期:2016-10-28 发布日期:2016-12-02
  • 作者简介:高明亮(1985-),男,讲师;蒋俊(1976-),男,讲师,E-mail:iocomp@163.com.
  • 基金资助:
    国家自然科学基金项目(61601266);山东理工大学青年教师支持计划项目;山东省自然科学基金联合专项项目(ZR2015FL034)

Visual Tracking Using Bat-Inspired Algorithm

GAO Ming-liang1, YIN Li-ju1, JIANG Jun2, SHEN Jin1   

  1. 1. College of Electrical and Electronic Engineering, Shandong University of Technology, Shandong Zibo 255000, China;
    2. School of Computer Science, Southwest Petroleum University, Chengdu 610500, China
  • Received:2016-03-31 Online:2016-10-28 Published:2016-12-02

摘要: 提出了一种基于蝙蝠算法的目标跟踪新方法.首先描述了基于蝙蝠算法的目标跟踪流程,在此基础上研究了蝙蝠算法的迭代终止条件、蝙蝠种群数、脉冲频率因子和脉冲音强衰减因子对跟踪性能的影响,并通过实验确定了以上参数,最后与粒子滤波、均值漂移和粒子群优化3种跟踪算法进行了比较.实验结果表明,基于蝙蝠算法的视频目标跟踪方法的跟踪效果优于其他3种算法.

关键词: 视觉跟踪, 蝙蝠算法, 迭代终止条件, 参数确定

Abstract: A new visual tracking method based on bat algorithm (BA) was presented, in which, the BA-based tracking architecture is proposed. Then, the impact of the iterative termination condition, population size, pulse frequency increasing coefficient and pulse amplitude attenuation coefficient on the tracking performance are studied and these parameters are determined. To demonstrate the tracking ability of the proposed tracker, comparative studies of tracking accuracy and speed of the BA-based tracker with three representative trackers, namely, particle filter, meanshift, particle swarm optimization are given. Comparative results show that the proposed algorithm outperforms the other three trackers.

Key words: visual tracking, bat algorithm, iterative termination condition, parameters determination

中图分类号: