中文摘要：

为了能有效解决动态背景下运动目标的检测问题,给出了一种结合HSV颜色空间和图像分块思想的码书模型方法。这种方法可以分为三步：首先根据视频图像大小,合理确定其分块的大小;然后根据HSV颜色空间下的亮度和颜色扭曲度信息构造块码书模型;最后利用得到的块码书模型检测运动目标,并且同时更新此码书模型。实验结果证实：对存在动态因素的背景视频,提出的方法不仅能有效抑制伪目标的出现,而且能准确、实时地检测出真实的运动目标。

英文摘要：

In order to effectively solve the problem of moving object detection in dynamic background,it proposes a novel method using block-based codebook model in HSV color space.The method includes the following three steps.Firstly,according to the video image size,determine the size of the block.And then build block-based codebook model based on the degree of brightness and color distortion in HSV color space.Lastly the moving object can be extracted through the model and update the codebook model.The results confirmed that the proposed method can detect the moving target effectively in real-time for the dynamic background.