一种基于改进帧差法的运动目标检测  

Moving Object Detection Based on Improved Frame Difference Algorithm

作　　者：严晓明[1] 

机构地区：[1]福建师范大学数学与计算机科学学院,福建福州350108

出　　处：《莆田学院学报》2011年第5期69-72,共4页Journal of putian University

基　　金：福建省自然科学基金资助项目(2009J01273)

摘　　要：针对视频中的运动目标离摄像头较近时由于形状的变化而导致目标边界测定不准确的问题,在帧间差分法的基础上,提出了一个求精运动目标边界的算法。先用得到边界值计算目标中心点,再对视频中的运动目标进行运动距离和速度的检测,并在Matlab中进行仿真。实验结果表明,该算法对帧差法的结果进行求精后,得到的运动目标的状态值更接近于实际情况,改进的算法更适合于视频中运动目标形状变化较大的情况下进行运动目标状态的检测。An improved algorithm for getting more precise boundary of moving object base on frame difference algorithm is listed in this paper.The situation of imprecise value of boundary because of moving object shape variation on video was solved by the improved algorithm.It performs simulation of detecting distance and velocity of moving object with object central point calculating by the precise value in matlab.Experiment results show that the proposed method has closer to realistic value than frame difference algorithm.It is suitable for moving object detection when shape of moving object varies greatly in video.

关 键 词：帧差法 目标检测 速度 改进算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TN911.72[自动化与计算机技术—计算机科学与技术]