一种基于自适应混合模型的背景减除法

计算机工程与科学

一种基于自适应混合模型的背景减除法

李伟生，李辉飞

（重庆邮电大学计算智能重庆市重点实验室，重庆 400065）

收稿日期:2015-09-06 修回日期:2015-11-05 出版日期:2016-10-25 发布日期:2016-10-25
基金资助:
国家自然科学基金（61272195,61472055）；重庆市基础与前沿研究(cstc2014jcyjjq40001)

A background subtraction method based

on adaptive hybrid model

LI Wei-sheng，LI Hui-fei

（Chongqing Key Laboratory of Computational Intelligence,
Chongqing University of Posts and Telecommunications,Chongqing 400065,China）

Received:2015-09-06 Revised:2015-11-05 Online:2016-10-25 Published:2016-10-25

摘要/Abstract

摘要：

针对局部二进制相似度（LBSP）背景建模方法易受外界环境变化如动态背景、光照改变、相机抖动等干扰的问题，在融合像素纹理与亮度信息的基础上，建立一种自适应混合背景模型进行运动目标检测。首先，利用每个像素的多通道自适应局部二进制相似度（LBSP）信息和亮度信息建立混合背景模型。然后，根据当前像素与混合背景模型的比较结果对其进行分类，并采用随机更新机制更新背景模型。实验结果表明，本方法不仅在正常外界环境下取得了较好的检测结果，而且还可以有效地减少动态背景、光照变化等复杂外界环境条件造成的干扰，提高检测结果的准确性。

关键词: 运动目标检测, 纹理背景模型, 自适应LBSP, 时间背景模型, 随机更新机制

Abstract:

In view that the existing background modeling method based on local binary similarity pattern (LBSP) is very vulnerable to external environment changes,such as dynamic background,illumination changes,camera shaking and so on,based on fusing pixel texture information with intensity information,we propose an adaptive hybrid model for moving object detecting.First,we use the texture descriptor of each pixel, named multi-channel adaptive local binary similarity pattern (LBSP) information,combined with intensity information,to build a hybrid background model.We then classify the current pixels according to the comparison results between the current pixel and the corresponding hybrid background model,and update the background model with the random updating mechanism.Experimental results show that the proposed method cannot only achieve good results in an ideal outside environment,but also effectively reduce the interference caused by complicated external environment conditions such as dynamic background,illumination changing and camera shaking,thus achieving better detection results.

Key words: moving object detection, texture background model, adaptive LBSP, temporal background model, random update mechanism

李伟生，李辉飞. 一种基于自适应混合模型的背景减除法[J]. 计算机工程与科学.

LI Wei-sheng，LI Hui-fei.

A background subtraction method based

on adaptive hybrid model

[J]. Computer Engineering & Science.

参考文献

［1］Wren C R, Azarbayejano A, Darrell T, et al. Pfinder:Real-time tracking of the humanbody［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(7): 780-785.

［2］Stauffer C,Grimson W E L.Adaptive background mixture models for real-time tracking［C］∥Proc of IEEE Computer Society Conference on Computer Vision and Pattern Recognition,1999: 246-252.

［3］KaewTraKulPong P,Bowden R.An improved adaptive background mixture model for real-time tracking with shadow detection［C］∥Proc of European Workshop Advanced Video Based Surveillance Systems,2002:135-144.

［4］Lee D.Effective Gaussian mixture learning for video background subtraction［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(5): 827-832.

［5］Zivkovic Z.Improved adaptive Gaussian mixture model for background subtraction［C］∥Proc of the 17th International Conference on Pattern Recognition,2004: 28-31.

［6］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.

［7］Elgammal A,Harwood D,Davis L.Non-parametric model for background subtraction［C］∥Proc of European Conference on Computer Vision,2000: 751-767.

［8］Wang H,Suter D.Background subtraction based on a robust consensus method［C］∥Proc of IEEE 18th International Conference on Pattern Recognition,2006: 223-226.

［9］Barnich O,Droogenbroeck M Van.ViBe: A universal background subtraction algorithm for video sequences［J］. IEEE Transactions on Image Processing,2011,20(6):1709-1724.

［10］Hofmann M,Tiefenbacher P,Rigoll G.Background segmentation with feedback: The pixel-based adaptive segmenter［C］∥Proc of 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW),2012: 38-43.

［11］Heikkil M,Pietikinen M.A texture-based method for modeling the background and detecting moving objects［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(4): 657-662.

［12］Liao S,Zhao G,Kellokumpu V,et al.Modeling pixel process with scale invariant local patterns for background subtraction in complex scenes［C］∥Proc of 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2010: 1301-1306.

［13］Ji Z,Wang W.Detect foreground objects via adaptive fusing model in a hybrid feature space［J］.Pattern Recognition,2014,47(9): 2952-2961.

［14］Bilodeau G A,Jodoin J P,Saunier N.Change detection in feature space using local binary similarity patterns［C］∥Proc of 2013 International Conference on Computer and Robot Vision (CRV),2013: 106-112.

［15］St-Charles P L,Bilodeau G A.Improving background subtraction using local binary similarity patterns［C］∥Proc of 2014 IEEE Winter Conference on Applications of Computer Vision (WACV),2014: 509-515.

［16］Li Wei-sheng,Wang Li-dou,Zhou Li-fang.Research on the local adaptive threshold ternary pattern for face recognition［J］.Journal of Chinese Computer Systems,2014,35(9):2099-2103.(in Chinese)

［17］Wang Y,Jodoin P M,Porikli F,et al.CDnet 2014: An expanded change detection benchmark dataset［C］∥Proc of 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW),2014: 393-400.

附中文参考文献：

［16］李伟生,王立逗,周丽芳.一种基于LTP自适应阈值的人脸识别方法［J］.小型微型计算机系统，2014,35(9): 2099-2103.

[1]	张寅, 蔡旭阳, 许倩倩, 闫钧华, 苏恺, 张琨 . 融合时空上下文的复杂背景下多运动目标检测[J]. 计算机工程与科学, 2020, 42(12): 2186-2192.
[2]	马琴，张兴忠，李海芳，邓红霞. 基于谱残差和聚类法的运动目标检测研究[J]. 计算机工程与科学, 2018, 40(10): 1867-1873.
[3]	金静1,党建武1,王阳萍1,2,翟凤文1. 一种基于鲁棒局部纹理特征的背景差分方法[J]. 计算机工程与科学, 2017, 39(08): 1513-1519.
[4]	杨国亮，丰义琴，唐俊，谢乃俊. 基于低秩表示动态更新投影的在线运动目标检测[J]. 计算机工程与科学, 2016, 38(11): 2289-2296.
[5]	李博川1，丁轲2. 结合阴影抑制的混合高斯模型改进算法[J]. J4, 2016, 38(03): 556-561.
[6]	刘军，梁久祯，柴志雷. 基于DM642的运动目标检测[J]. J4, 2013, 35(1): 107-112.
[7]	吴亮周东翔范才智蔡宣平. 基于直方图匹配的鬼影检测算法[J]. J4, 2008, 30(7): 57-60.
[8]	倪崇嘉[1] 刘文奇[2] 张爱英[2]. 基于数学形态学的视频图像序列中的运动目标检测[J]. J4, 2006, 28(6): 69-70.