基于改进的混合高斯背景模型的运动目标检测研究  

Research on Moving Object Detection Based on Improved Gaussian Mixture BackGround Model

作　　者：徐凯[1] 曾宪林[1] 颜广[1] 

机构地区：[1]贵州师范大学,贵州贵阳550014

出　　处：《科技广场》2012年第7期74-77,共4页Science Mosaic

基　　金：贵州省自然科学基金(黔科合J字LKS[2010]36号)

摘　　要：针对传统混合高斯背景建模(GMM)在一些复杂场景下未能有效地描述背景,目标容易出现错误的检测,本文提出一种改进算法。该算法先对视频帧进行分块处理,然后对单模区域和多模区域采用不同的更新速率进行更新,最后在空域上对检测结果进行数学形态学的处理,从而提取运动目标。实验结果表明,该算法能够提高背景建立和运动目标检测的速率,在多种场景下运动目标的检测都具有良好的鲁棒性。Traditional Gaussian mixture background modeling（GMM） in some complex scenes often fails to effectively describe the background,and moving target is detected incorrectly.So this paper presents an improved algorithm.The video frame is divided into blocks at first,and then the single-mode the regional and multimode area are updated by using different update rate.The test results are processed by mathematical morphology processing on airspace finally.So the moving target is extracted correctly.The experimental results show that the algorithm can improve the background to the establishment and moving target detection rate,and the detection of moving targets have good robustness in a variety of scenarios.

关 键 词：混合高斯模型 运动目标检测 背景建模 背景更新 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]