计算机技术与发展2016，Vol.26Issue(7)：36-39,45,5.DOI:10.3969/j.issn.1673-629X.2016.07.008
一种改进的ViBe背景提取算法
An Improved ViBe Background Generation Method
摘要
Abstract
Background generation and foreground detection play a significant role in the field of intelligent video surveillance. By extrac-ting the background from video scenes,the algorithm can detect the movement of the foreground objects. An improved background gener-ation method based on ViBe background extraction is proposed. In terms of sample collection,samples are taken in the neighborhood of every pixel according to a Gaussian distribution. Regarding sample updating,an adaptive probability of random sub-sampling is presen-ted. When it comes to background generation,by constructing a color histogram of the sample pixels,a background generation method is put forward based on color histograms and Mean Shift iteration. And the foreground is detected with the pixel information in both time and space domain. It specifically describes an improved ViBe background generation method in this paper,which is compared with the o-riginal algorithm by means of experiments. The results show that the proposed algorithm is more effective in both background generation and foreground extraction.关键词
ViBe算法/背景生成/目标检测/颜色直方图/均值漂移Key words
ViBe algorithm/background generation/object detection/color histogram/Mean Shift分类
计算机与自动化引用本文复制引用
张夏清,茅耀斌..一种改进的ViBe背景提取算法[J].计算机技术与发展,2016,26(7):36-39,45,5.基金项目
国家自然科学基金资助项目(61472188) （61472188）
