摘要
针对标准粒子滤波重采样阶段导致粒子多样性降低进而影响跟踪精度的问题,在原重采样阶段使用萤火虫算法优化,利用萤火虫的移动机制进化粒子,促使粒子向最优解区域移动。最后将改进的算法应用于使用HSV(Hue,Saturation,Value)颜色特征建模目标的视频目标跟踪中,实验结果表明:与标准粒子滤波相比,改进算法的跟踪精度有了一定提高。
For the problem of standard particle filter in process of resampling will reduces the diversity of particles and affects the tracking accuracy,the firefly algorithm is used to optimize the original resampling phase.The firefly's movement mechanism is used to evolve particles to promote the particles to move to the optimal solution region.Finally,the improved algorithm is applied to video target tracking using HSV(Hue,Saturation,Value)color feature modeling targets.The result shows that compared with stan⁃dard particle filtering,the tracking accuracy of the improved algorithm has been improved.
作者
杨永超
周昊
YANG Yong-chao;ZHOU Hao(School of Big Data and Artificial Intelligence,Chizhou University,Chizhou 247000,China)
出处
《电脑知识与技术》
2020年第19期1-2,5,共3页
Computer Knowledge and Technology
基金
安徽高校自然科学研究重点项目:人工细菌菌落算法研究及其在人群异常行为检测中的应用(KJ2019A0864)
安徽省池州学院自然科学研究项目(2016ZR011)
池州学院质量工程项目“线下精品课程——计算机网络”(2018XJPKC15)。
