摘要
对均值漂移算法的理论和应用作一全面的综述.首先根据密度函数的非参数估计推导出均值漂移公式的一般形式,说明了均值漂移迭代算法的步骤及收敛性;然后重点讨论核函数的选择以及带宽矩阵的计算等关键技术;最后归纳了均值漂移算法在模式检测、聚类、图像分割以及物体实时跟踪等方面的应用,并展望了均值漂移算法在理论和应用中的研究方向.
A comprehensive survey of the up-to-date research and application for mean shift are presented. A general mean shift procedure is defined by nonparametric estimation of density gradient, and the steps and convergency of the algorithm are proposed. The key techniques of the selection of kernel function and the computation of bandwidth matrix are introduced emphatically. Finally, the applications of mean shift to mode detection, clustering, image segmentation and tracking are surveyed in detail and the directions towards the future research are outlined.
出处
《控制与决策》
EI
CSCD
北大核心
2007年第8期841-847,共7页
Control and Decision
基金
国家自然科学基金项目(69975003)
关键词
均值漂移
非参数密度估计
核函数
带宽
Mean shift
Non-parameter density estimation
Kernel function
Bandwidth
