一种基于蝙蝠算法的视觉跟踪方法
发布时间：2019-01-09 11:33
【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.
[Abstract]:A new target tracking method based on bat algorithm is proposed. Firstly, the object tracking process based on bat algorithm is described. Based on this, the iterative termination conditions of bat algorithm, the effects of bat population number, pulse frequency factor and pulse tone attenuation factor on tracking performance are studied. The above parameters are determined by experiments, and compared with three tracking algorithms: particle filter, mean shift and particle swarm optimization. The experimental results show that the video target tracking method based on bat algorithm is better than the other three algorithms.
【作者单位】： 山东理工大学电气与电子工程学院;西南石油大学计算机科学学院;
【基金】：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)
【分类号】：TP391.41
，
本文编号：2405567
[Abstract]:A new target tracking method based on bat algorithm is proposed. Firstly, the object tracking process based on bat algorithm is described. Based on this, the iterative termination conditions of bat algorithm, the effects of bat population number, pulse frequency factor and pulse tone attenuation factor on tracking performance are studied. The above parameters are determined by experiments, and compared with three tracking algorithms: particle filter, mean shift and particle swarm optimization. The experimental results show that the video target tracking method based on bat algorithm is better than the other three algorithms.
【作者单位】： 山东理工大学电气与电子工程学院;西南石油大学计算机科学学院;
【基金】：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)
【分类号】：TP391.41
，
本文编号：2405567
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2405567.html
