随机图像选取与自适应背景更新的运动物体检测  

Moving Object Detection Based on Random Image Selection and Adaptive Background Updating

作　　者：李新县 余烨[1] 

机构地区：[1]合肥工业大学计算机与信息学院VCC研究室,安徽合肥230009

出　　处：《图学学报》2014年第4期577-584,共8页Journal of Graphics

基　　金：国家自然科学基金资助项目(61070124);安徽省自然科学基金资助项目(11040606Q43);安徽省科技攻关资助项目(12010402108)

摘　　要：为克服运动目标检测中易出现的光照变化、遮挡、虚假目标等现象,提出了一种随机图像选取与自适应背景更新的运动物体检测方法。该方法从视频序列中随机选取一帧图像作为初始背景,根据变化标记矩阵对背景进行自适应迭代更新,以提取可靠的背景图像,实现运动物体的检测。实验结果表明,采用该算法提取的背景不存在混合现象,且在光照变化较大以及运动物体之间存在遮挡的情况下,能够构造出可靠的背景,检测出的目标物体清晰可见。In order to overcome the phenomena of illumination changes, occlusion and take target, which appear easily in the moving target detection process, a random image selection and adaptive background updating method for moving object detection is proposed in this paper. First, the background is initialized through selecting an image from the video sequence randomly, and then it is updated adaptively and iteratively based on change marker matrix to extract reliable background images. The experiment results demonstrate that the background extracted through this algorithm does not have the mixing phenomenon, so a reliable background can be constructed, and the target object can be detected clearly even in the case of large changes in illumination and occlusion between moving objects.

关 键 词：运动物体检测 随机图像选取 迭代更新 变化标记矩阵 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]