复杂背景下的运动前景分割是计算机视觉领域研究的一个重点研究问题。为了对复杂背景下的运动前景进行有效分割，提出了一种复杂背景下自适应前景分割算法。该算法的背景模型是由一系列聚类和聚类的权重构成。每个聚类表示背景的一个历史状态，并能够根据背景的变化，自适应创建、更新或删除聚类，使得背景模型能够准确反映出场景的变化。每个聚类权重是根据聚类的大小和更新时间自动确定的。为了自动确定该方法的重要阈值，还提出一种基于非参数密度估计的阈值估计方法，并在不同的场景下与多个背景建模方法进行了比较， 实验结果表明，该算法是有效的。

A robust background subtraction technique is proposed based on adaptive clustering of temporal color/intensity. An un-supervised clustering method is proposed to model a background with a group of weighted clusters. The clusters and their weights can be updated with a background change. In addition, the unimodal or multimodal distributions of background are detected adaptively. We also present a novel statistical threshold estimation scheme to determine the thresholds using in our method. Experimental results on different types of videos demonstrate the utility and performance of the proposed approach.