一种基于改进ViBe算法的运动目标检测方法  

A moving object detection method based on improved ViBe algorithm

作　　者：郭志涛 曹小青 厉成元 于洪泽 王卿粹 GUO Zhitao;CAO Xiaoqing;LI Chengyuan;YU Hongze;WANG Qingcui(Tianjin Electric Science Research Institute Co.Ltd.,Tianjin 300301,China;School of Electronic and Infoonatioo Enginering,Hebei University of Technology,Tianjin 300401,China)

机构地区：[1]天津电气科学研究院有限公司,天津300301 [2]河北工业大学电子信息工程学院,天津300401

出　　处：《电视技术》2018年第8期33-35,48,共4页Video Engineering

基　　金：河北省科技计划项目(17275404D)

摘　　要：当视频监控的背景较复杂或目标动态性强时易出现目标检测不完整的问题,通过定义两种误检情况下的评价函数动态调整阈值R,提出了一种阈值自适应更新的改进ViBe算法,并针对交通监控视频容易出现的阴影问题提出一种融合颜色特征和角点特征的阴影消除方法。实验结果表明,改进ViBe算法可以有效地解决传统ViBe算法在复杂情况下运动目标检测不完整的问题,且阴影消除效果良好,可以被广泛应用在视频监控领域。When the background of video surveillance is complex or the target is strongly dynamic,the problemof incomplete target detection is easy to appear.By defining the two evaluation functions in two misdiagnosis to dynamically adjust the threshold R!an improved ViBe algorithm with adaptive threshold updating is proposed.And a shadow elimination met!iod based on color featiure and corner feature is proposed for the shadow problemwhich is easy to appear in the traffic surveillance video.The experimental re-sults show that the improved ViBe algorithmcan effectively solve the problemof incomplete detection of moving targets in the com-plex situation of traditional ViBe algorithm,and the effect of shadow elimination is good,so it can be widely used in the field of video surveillance.

关 键 词：ViBe算法 运动目标检测 自适应阈值 阴影消除 视频监控 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]