基于局部层次化混合高斯模型的视频序列运动目标检测

The Moving Object Detection based on Local Hierarchical GMM

在视频分析的过程中,背景建模和运动目标提取是一个非常重要的问题。混合高斯模型是进行背景建模常用的模型之一。但是单纯运用混合高斯模型进行运动目标提取的效果并不是非常理想。本文提出了一种自上而下的局部层次化混合高斯模型,该算法首先确定更新区域,然后在区域中运用分块的混合高斯模型和点像素混合高斯模型进行背景建模和目标提取。实验表明该方法具有较好的处理效果,同时也提高了处理的时间效率。

In the process of the video sequence analysis,the method of the background modeling and moving target detection is an important problem. The Gaussian Mixture Model（GMM） is one of the model which used frequently. But the effect is not so ideal if we use the GMM only to detect the moving targets. A top-to-bottom Local Hierarchical GMM （LHGMM） is proposed in this paper, firstly the updating area is found, then in the area we use the block-based GMM and the pixel-based GMM to update the background and to detect the moving target. The experiment results show that the dispose effects are fair,and the dispose efficiency is improved at the same time.