摘要
针对运动目标检测算法视觉背景提取(visual background extractor,ViBe)对光线突变适应缓慢和易产生目标丢失等问题,提出一种ViBe的改进算法。充分利用ViBe算法初始化特点,在ViBe中加入光线判定方法并设计生成ViBe新背景模型的方法,改变原始ViBe算法结构,使ViBe算法在光线突变时找到合适的帧重置背景模型,将光线突变时模型恢复问题变为模型重置问题,达到迅速适应光照突变的目的。实验结果表明,使用该方法可以使ViBe有两方面的改进:在光线突变时不会产生目标丢失;使ViBe算法更加快速地适应光照变化。
Moving target detection algorithm ViBe adapts to light slowly and is easy to produce mutant target loss?to solve the problem,an improved ViBe algorithm was proposed.Initialization characteristics of ViBe were used,light judging method and new ViBe background model generating method were added to change the original ViBe algorithm structure.Background model of the ViBe algorithm in light mutation was reset,the sudden change of illumination was adapted quickly.The problem of model restoration was changed into model reset problem,which was used to adapt to quick change of light.Experimental results show that the original ViBe algorithm is improved using this improved ViBe algorithm from two aspects.First,it will not cause the loss of target in the light of the mutation.Second,this method can make the ViBe algorithm more quickly adapt to the change of illumination.
作者
丁洁
况立群
韩燮
DING Jie;KUANG Li-qun;HAN Xie(School of Computer and Control Engineering, North University of China, Taiyuan 030051, China)
出处
《计算机工程与设计》
北大核心
2017年第2期374-378,共5页
Computer Engineering and Design
基金
国家科技支撑计划基金项目(2013BAH45F02)
关键词
视觉背景提取算法
光线突变
帧间差分
背景差分
运动目标检测
背景模型重置
ViBe algorithm
sudden illumination changes
frame difference
background subtraction
move object detection
background model reset
