基于混合模型的运动目标检测算法  

Hybrid model based on the target motion detection algorithm

作　　者：赵陈[1] 刘鹏飞[1] 闫亮[1] 

机构地区：[1]兰州交通大学电子与信息工程学院,甘肃兰州700030

出　　处：《电子测试》2011年第1期27-30,77,共5页Electronic Test

摘　　要：自然界中复杂的环境对视频监控系统有一定的干扰,对运动物体的检测也将产生显著的影响,通过分析光照变化、背景干扰等现象对运动目标的检测的影响,本文提出了一种能适应环境变化的混合模型运动目标检测算法,该算法结合帧间差分法和混合高斯模型算法的优点,将帧差法作为进行高斯背景减法的指导,既能得到较高的灵敏度又能进一步提高检测效果,提高了算法时效性和鲁棒性,通过仿真实验,达到了预期的实验目的,证明了算法的可靠性与实时性。In the nature of the complex environment has certain interference to video monitoring system, The detection of moving objects which will significantly impact. Through the analysis of illumination change, background interference phenomenon of moving targets detectio, This paper puts forward a kind of adaptive hybrid model target motion detection algorithm. The algorithm combines frame difference method and the advantages of gaussian mixture model algorithm, The frame difference method as the Gaussian background subtraction guidance, both to obtain high sensitivity and can further improve the detection results,To improve the timeliness and robustness of the algorithm. To achieve the desired purpose of the experiment. Through the simulation experiment proved that the algorithm of reliability and performance.

关 键 词：运动目标检测 光照变化 背景影响 帧间差分 混合高斯模型 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]