摘要
ViBe算法是一种高效的像素级视频背景建模或前景检测算法。针对传统ViBe算法中存在的"死区"现象以及阴影干扰问题,提出了结合感知哈希算法和基于图像RGB色彩信息的高斯拉普拉斯差分算法的改进ViBe算法,实现了对"死区"的抑制,消除了运动目标阴影,完成了对视场范围内运动目标的检测。实验表明,相较于传统ViBe算法在视频的1315帧才能完成对"死区"的抑制,改进后的ViBe算法仅需15帧就完成了对"死区"的抑制,消除了运动阴影的干扰,并测试了算法的鲁棒性和实时性,并取得良好的效果。
ViBe algorithm is an efficient pixel-level video background modeling or foreground detection algorithm. Aiming at the phenomenon and shadow interference existing in the traditional ViBe algorithm, this paper proposed a method to combine perceptual hash algorithm with ViBe algorithm for the suppression of dead zone and used Gaussian Laplacian Difference Method based on image’s RGB color information to eliminate moving object shadow. Compared with the traditional ViBe algorithm which completed the suppression of dead zone in the 1315 th frame of video, the improved ViBe algorithm only completed the suppression of the dead zone in the 15 th frame of the video, and the interference of motion shadows was eliminated, and the robustness and real-time performance of the algorithm were tested, and good results were achieved.
作者
刘燕德
曾体伟
陈洞滨
周鑫
LIU Yan-de;ZENG Ti-wei;CHEN Dong-bin;ZHOU Xin(School of Mechatronic Engineering,East China Jiaotong University,Nanchang Jiangxi 330013,China)
出处
《计算机仿真》
北大核心
2019年第2期404-408,共5页
Computer Simulation
基金
江西省教育厅科学技术研究项目(GJJ160499)
江西省科技支持项目(20141BBE50025)
关键词
死区
运动目标阴影
哈希算法
Dead zone
Moving object shadow
Hash algorit