基于三帧差分法和改进ViBe算法的前景检测方法  

Foreground Detection Method Based on Three-frame Difference Method and Improved Vibe Algorithm

作　　者：杨波 潘峥嵘[1] YANG Bo;PAN Zhengrong(College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050)

机构地区：[1]兰州理工大学电气工程与信息工程学院,兰州730050

出　　处：《计算机与数字工程》2021年第11期2242-2247,共6页Computer & Digital Engineering

摘　　要：在运动目标检测众多算法中,ViBe算法是一种快速并且高效的运动目标检测算法,在视频监控领域得到广泛的应用。但是这种方法存在对鬼影的消除速度缓慢,容易形成拖影,而且容易发生漏检,易受到噪音的干扰。当物体运动状态变化缓慢时,原位移区域会留下残影等现象。论文针对以上问题提出了一种结合三帧差分算法的改进的ViBe算法,改进背景模型的初始化来对鬼影进行消除,并且加快鬼影的消除速度。采用自适应阈值来提高检测效果,针对背景中存在的噪声和空洞点采用形态学处理来对其进行消除。实验结果表明,改进的算法可以在很短的时间内就可以消除鬼影,并且在消除噪声和空洞等方面有很好的效果。Among the moving target detection algorithms,ViBe algorithm is a fast and efficient moving target detection algo⁃rithm,which has been widely applied in video surveillance neighborhood.However,this method is slow to eliminate the ghost,easy to form drag shadow and easy to miss,easy to be disturbed by noise.When the motion state of the object changes slowly,the residu⁃al shadow will be left in the original displacement region.In this paper,an improved ViBe with the three-frame difference algorithm is proposed to solve the above problems,the initialization of the background model is improved to eliminate ghost,and the speed of ghost elimination is accelerated.Adaptive threshold is used to improve the detection effect,and morphological processing is used to eliminate noises and voids in the background.Experimental result shows that the improved algorithm can eliminate ghosts in a short time,and has a good effect in eliminating noise and holes.

关 键 词：ViBe 三帧差分法 鬼影消除 自适应阈值 形态学处理 

分 类 号：TN911.73[电子电信—通信与信息系统]