基于颜色信息和光学增益的阴影消除

Shadow Elimination Based on Information of Color and Photometric Gain

运动阴影常被误分为前景对象,会给目标的跟踪和识别带来很大困难,所以阴影消除在许多的监控系统中都是非常重要的。针对阴影检测算法受特定条件约束,不能自动适用于不同场景,为解决上述问题,提出了一种新颖鲁棒且无需复杂的参数调整的运动阴影消除方法。通过自适应高斯混合模型重建背景,采用背景差分法提取出包含阴影的运动区域,综合颜色信息与光学增益,将运动区域分类为运动对象区域和运动阴影区域。实验结果表明:所提方法在多种不同的场景下均能有效可靠的消除运动阴影。

Moving shadow elimination is critical for video surveillance system since shadow points may be misclassifted as object points, thus often resulting in problems for many applications such as object tracking/recognition and so on. In view of the problem that shadow elimination is not self - adaptive to different scenes owing to specific limi- tations at present, a novel and robust shadow detection algorithm without manual adjustment of scene parameters is presented. The background is modeled using adaptive Gaussian mixture models. Moving regions are extracted based on background subtraction. Moving shadow is detected by combining color information with photometric gain. The experimental results show that the proposed method can eliminate shadow efficiently and reliably in different scenes.