视频序列中自适应背景的运动目标提取

Moving Object Detection Based on Self-Adaptive Background in Video Sequences

在视频序列的运动目标实时检测中常用的方法是背景差分法,但因背景图像未随监视场景光照变化及时更新而限制了本方法的适应性.对此,本文首先提出了一种自适应背景更新方法,得到差分二值图像;然后引入散布函数准则计算出差分图像的最佳分割阈值,实现运动目标的自适应阈值分割.实验结果表明:本文提出的背景更新方法可随着光照条件的变化实时、准确地更新背景图像,在此基础上提出的基于散布函数准则的自适应阈值分割方法可以实现运动目标的完整分割,这为运动目标的后续识别与理解奠定了基础.

For real-time detection of moving object,the general and simple method is based on background image difference.However, the background image don＇t update timely along with the surveillance illumination variance, which limits its applications. To overcome the above problem, a new self-adaptive background updating algorithm is first presented in this paper. Moreover, the differential binary image is obtained. And then a self-adaptive threshold segmentation method for moving object detection based on spread function criteria is developed and implemented.Experimental results demonstrate that the proposed new background updating method can update the background exactly and quickly along with the variance of illumination,and the self-adaptive threshold segmentation method based on spread function criteria can extract the moving object regions accurately and completely, which is the foundation for further objects recognition and understanding.