基于高斯混合模型和帧间梯度信息提出了一种新的运动目标分割算法。首先，在利用亮度信息对背景建立自适应高斯混合模型的基础上，进行前景的粗分割；其次，由于视频信号的亮度和色彩分量随光照突变有较大的改变，导致大片背景的高斯模型产生错误匹配，误判为前景，为了提高高斯模型分割算法的鲁棒性，结合结构梯度互相关函数对分割结果进一步校正，能适应剧烈的光照变化；最后，利用数学形态学进行后处理，消除影子和孤立的噪声点。通过不同场景的运动分割实验结果表明，该算法在复杂背景和剧烈光照变化条件下具有较强的鲁棒性和较高的分割精度。

In this paper, a novel segmentation algorithm is proposed which is based on MOGs and interframe gradient information. Firstly, a primary foreground segmentation is obtained, where an adaptive MOGs(Mixture of Gaussians) is established for each pixel’s luminance; Secondly, luminance and chroma of each pixel change largely due to the abrupt illuminance change, which causes the mismatch between a pixel’s luminance and its MOGs, and causes the misclassification of a vast of background pixels as the foreground as well. To adapt to the illuminance sudden variation, an improved method using the interframe gradient information is adopted to correct the initial segmentation. Finally, morphological methods are used to remove shadows and isolated noise pixels. Experimental results on various video sequences show that thismethod is robust and of high segmentation accuracy.