改进粒子群优化算法下的视频目标跟踪方法  

Tracking Video Target Using an Improved Particle Swarm Optimization Algorithm

作　　者：赖万钦[1] LAI Wanqin(Fujian Chuanzheng Communications College,Fuzhou,Fujian 350007)

机构地区：[1]福建船政交通职业学院,福建福州350007

出　　处：《武夷学院学报》2022年第9期53-56,共4页Journal of Wuyi University

基　　金：福建省中青年教师教育科研项目(JAT201058)。

摘　　要：提出一种基于改进的粒子群优化算法的视频目标跟踪方法,对视频目标跟踪问题模型进行分析设计,建立一种视频目标跟踪的四维参数模型,针对该模型使用粒子群优化算法寻求最优解。分析视频目标跟踪场景的特性,使用模式匹配相关系数作为粒子群算法的适应度估计函数,同时引入动态调整粒子群算法的惯性权重参数的改进方法,避免传统粒子群算法过早收敛,陷入局部最优的问题。实验测试表明该算法相较于传统算法,在视频目标的跟踪上具有更佳的检测精确度和跟踪稳定性。The main purpose is to design and implement a video target tracking method based on improved particle swarm optimization algorithms. Analysis of video target tracking models establishes a four-dimensional parameter model and uses particle swarm optimization algorithm to resolve it. According to the video target tracking scene, the pattern matching correlation coefficient is used as the appropriateness estimation function of the particle swarm optimization algorithm. At the same time, using a method of dynamically adjusted the inertia weight parameters to avoid the local optimal problems. Experimental tests show that the algorithm has better detection accuracy and track stability on the tracing of video objectives compared to traditional algorithms.

关 键 词：视频跟踪 粒子群算法 惯性权重 适应度估计 相关系数 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]