摘要： 论述了一种静态摄像机下从彩色图像中检测运动目标的背景减除算法，该算法在彩色图像亮度偏差和色度偏差概念的基础上被提出的，计算量小但可以很好地克服亮度变化、反光和阴影的影响。对于一个序列，首先用像素点的中间值作为背景的像素值，生成背景帧；然后依据当前帧和背景帧的色度偏差值将图像二值化；最后，使用数学形态学算子和连通分量分析对检测结果进行后处理。实验结果表明，该算法有很好的处理效果。

关键词: 运动目标检测；背景减除；色度偏差；连通分量

Abstract: This paper presents a background subtraction algorithm for detecting moving objects from a static background scene using color images. The algorithm is proposed on the basis of brightness distortion and chromaticity distortion in three-dimensional RGB color space, which is able to cope with varying illumination, reflection and shadow problems with low computational complexity. For an image sequence, background frame is first generated using the medium pixel values as background values; then, color images are converted into binary ones according to chromaticity distortion of current-frame pixels from background-frame pixels; at last, morphologic operation and connected component measurement are introduced to solve background disturb problem. Evaluation shows that the proposed algorithm is highly effective.

Key words: Moving objects detection; Background subtraction; Chromaticity distortion; Connected component