指挥信息系统与技术
指揮信息繫統與技術
지휘신식계통여기술
COMMAND INFORMATION SYSTEM AND TECHNOLOGY
2011年
6期
51-54
,共4页
视频目标跟踪%均值漂移算法%粒子滤波算法
視頻目標跟蹤%均值漂移算法%粒子濾波算法
시빈목표근종%균치표이산법%입자려파산법
video target tracking%Mean Shift algorithm%particle filter algorithm
为解决有遮挡视频图像目标跟踪问题,确保其性能的鲁棒性、准确性和实时性,提出了一种视频图像目标抗遮挡跟踪算法。该算法将均值漂移(MeanShift)算法嵌入到粒子滤波算法,同时对比介绍了MeanShift和粒子滤波算法的优缺点。试验证明,该算法具有运算量少、跟踪速度快、鲁棒性高和抗遮挡性强等优点。
為解決有遮擋視頻圖像目標跟蹤問題,確保其性能的魯棒性、準確性和實時性,提齣瞭一種視頻圖像目標抗遮擋跟蹤算法。該算法將均值漂移(MeanShift)算法嵌入到粒子濾波算法,同時對比介紹瞭MeanShift和粒子濾波算法的優缺點。試驗證明,該算法具有運算量少、跟蹤速度快、魯棒性高和抗遮擋性彊等優點。
위해결유차당시빈도상목표근종문제,학보기성능적로봉성、준학성화실시성,제출료일충시빈도상목표항차당근종산법。해산법장균치표이(MeanShift)산법감입도입자려파산법,동시대비개소료MeanShift화입자려파산법적우결점。시험증명,해산법구유운산량소、근종속도쾌、로봉성고화항차당성강등우점。
To solve the problem of occlusion existence in video target tracking and to ensure the robust ,accurate and real-time video target tracking, an anti-occlusion video target tracking algorithm is proposed. Attributes of the Mean Shift algorithm and the particle filter algorithm are analyzed. In this algorithm, the Mean Shift algorithm is embedded into the particle filter algorithm. Experimental results show that the algorithm has characterisctics of computation quantity, rapid tracking speed, good robustness, and stronger anti-occlusion.