基于帧间差分的自适应运动目标检测方法  

Detection algorithm of adaptive moving objects based on frame difference method

作　　者：薛丽霞[1] 罗艳丽[1] 王佐成[2] 

机构地区：[1]重庆邮电大学计算机科学与技术学院,重庆400065 [2]重庆邮电大学软件学院,重庆400065

出　　处：《计算机应用研究》2011年第4期1551-1552,1559,共3页Application Research of Computers

基　　金：重庆邮电大学博士基金资助项目(A2008-14)

摘　　要：提出了一种基于帧间差分的自适应运动目标检测算法。算法利用直方图统计各像素点处最大概率灰度的方法提取出连续视频的背景图像;相邻帧利用帧差法得到运动区域图像;利用运动区域图像与背景图像差分的方法提取出运动目标。实验结果表明,该算法能在多个不确定性因素的序列视频中较好地提取背景图像,能及时响应实际场景变化,提高运动目标检测的质量。This paper proposed a detection algorithm of adaptive moving objects based on frame difference method.Reconstructed the background image of continuous video frequency by calculating the maximum probability grayscale using grey histogram;Gained moving regions by frame difference,took out moving objects by subtracting moving regions and background image.Experimental results show that the algorithm can pick up preferably background image from the video sequence including many uncertainties factors,can response timely to the actual scene changes and improve the quality of moving objects detection.

关 键 词：帧间差分 直方图 运动区域 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]