[1]王淑青,周博文.基于改进ViBe算法的运动目标识别方法[J].湖北工业大学学报,2019,34(4):58-62.
　WANG Shuqing,ZHOU Bowen,et al.Research on Moving Target Recognition Method Based on Improved ViBe Algorithms[J].,2019,34(4):58-62.
点击复制

基于改进ViBe算法的运动目标识别方法()
分享到：

《湖北工业大学学报》[ISSN:1003-4684/CN:42-1752/Z]

卷:
34卷
期数:
2019年第4期
页码:
58-62
栏目:
湖北工业大学学报
出版日期:
2019-08-30

文章信息/Info

Title:
Research on Moving Target Recognition Method Based on Improved ViBe Algorithms
文章编号:
1003-4684(2019)04-0058-05
作者:
王淑青1; 2;  周博文1; 2
1 湖北工业大学湖北省电网智能控制与装备工程技术研究中心, 湖北 武汉 430068; 2 湖北工业大学电气与电子工程学院, 湖北 武汉 430068
Author(s):
WANG Shuqing1;  2;  ZHOU Bowen1;  2
1 Hubei Provincial Research Center for Power Grid Intelligent Control and Equipment Engineering Technology,Hubei Univ. of Tech., Wuhan 430068, China; 2 School of Electrical and Electronic Engineering, Hubei Univ. of Tech., Wuhan 430068, China
关键词:
ViBe算法;  自适应阈值;  颜色畸变值;  伪影消除;  空洞填充
Keywords:
ViBe algorithm;  adaptive threshold;  color distortion value;  artifact elimination;  hole filling
分类号:
TP391.4
文献标志码:
A
摘要:
针对ViBe算法存在的伪影及易受复杂环境干扰的缺点,提出一种基于改进ViBe算法的运动目标识别方法。该方法使用20邻域建模来创建背景模型; 通过依据模型复杂度的自适应阈值和颜色畸变阈值双重作用来识别背景点; 采用基于像素生命周期的更新策略来完善背景模型; 对识别后的图像采用噪声斑点丢弃和空洞填充策略来提高识别目标的完整性。实验结果表明,改进后的算法能够在较少帧数内消除目标的伪影,当视频中存在光照变化、阴影以及各种遮挡物等复杂干扰时,该方法对于运动目标识别具有较高的精确度和稳定性。
Abstract:
Aiming at the artifacts of ViBe algorithm and its vulnerability to complex environment interference, this paper proposes a moving target recognition method based on the improved ViBe algorithm. The method uses 20-neighborhood modeling to create a background model, and identifies background points according to the dual effects of adaptive threshold of model complexity and color distortion threshold. It improves background model by updating strategy based on pixel life cycle, and improves the integrity of recognition target by using noise speckle discarding and hole filling strategy for the recognized image. The experimental results show that the improved algorithm can eliminate artifacts within a small number of frames. When there are complex interferences such as illumination changes, shadows and various occlusions in the video, the method has high accuracy and stability for moving object recognition.

参考文献/References:

[1] Zivkovic Z. Improved adaptive gaussian mixture model for background subtraction[C]// null. IEEE Computer Society, 2004.
[2] Kim K, Chalidabhongse T H, Harwood D, et al. Real-time foreground-background segmentation using codebook model[J]. Real-Time Imaging,2005, 11(3):172-185.
[3] Barnich O,Droogenbroeck M V.ViBE:A powerful random technique to estimate the background in video sequences[C]// IEEE International Conference on Acoustics, 2009.
[4] Barnich O, Droogenbroeck M V. ViBe: A universal background subtraction algorithm for video sequences[J]. IEEE Transactions on Image Processing, 2011, 20(6):1709-1724.
[5] Pan C, Zhu Z, Jiang L, et al. Adaptive ViBe background model for vehicle detection[C]// Advanced Information Technology, Electronic & Automation Control Conference, IEEE, 2017.
[6] 闵卫东,郭晓光,韩清.改进的ViBe算法及其在交通视频处理中的应用[J].光学精密工程,2017,25(3):806-811.
[7] Han G, Wang J, Cai X. Improved visual background extractor using an adaptive distance threshold[J].Journal of Electronic Imaging, 2014, 23(6):063005.
[8] Yang S, Hao K, Ding Y, et al. Improved visual background extractor with adaptive range change[J]. Memetic Computing, 2017(2006):1-9.
[9] Van Droogenbroeck M, Paquot O. Background subtraction: Experiments and improvements for ViBe[P]. Computer Vision and Pattern Recognition Workshops(CVPRW), 2012 IEEE Computer Society Conference on,2012.
[10] Wang Y,Jodoin P M,Porikli F,et al.CDnet 2014: An expanded change detection benchmark dataset[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops(CVPRW). IEEE Computer Society, 2014.

备注/Memo

备注/Memo:
[收稿日期] 2019-04-16
[基金项目] 国家自然科学基金青年项目(61601176)
[第一作者] 王淑青(1969-), 女,河北衡水人,工学博士,湖北工业大学教授,研究方向为智能检测与控制,系统分析与集成
更新日期/Last Update: 2019-08-25