Abstract: Moving shadows cause serious problem while segmenting and extracting foreground from video sequences, due to the misclassification of moving shadow as foreground. In order to detect the moving objects accurately, a method based on the similarity between little textured patches was proposed to remove the moving shadows. Firstly, the potential shadows were detected by analyzing the intensity and color properties. Secondly, the shadow detection approach was improved by evaluating the textural similarity between the current frame and the corresponding background model. Finally, the geometric heuristics were imposed to further improve the performance. Experimental results on both indoor and outdoor scenes exhibit that the proposed method succeeds in removing shadows robustly and achieves real-time performance.

Key words: background subtraction, moving object detection, shadow removal, Local Binary Pattern (LBP)

摘要： 视频序列运动目标检测过程中，运动目标往往会连同其投射阴影一起被检测为前景，这不利于对运动目标的进一步分类、识别等高层次视觉处理。为了提高运动目标检测的准确性，提出利用局部纹理不变性去除运动目标阴影。首先根据阴影的亮度色度属性分割出疑似阴影区域，然后在疑似阴影区域采用增强的局部纹理描述算子(ILT)提取纹理特征，利用背景在阴影覆盖前后的纹理相似性来去除阴影，最后结合阴影的空间几何属性优化运动目标检测结果。实验结果表明该算法可以有效去除阴影，并且具有较好的实时性。

关键词: 背景差分, 运动目标检测, 阴影去除, 局部二值模式