基于改进的背景差分的运动目标实时检测算法

J4 ›› 2014, Vol. 36 ›› Issue (07): 1352-1356.

基于改进的背景差分的运动目标实时检测算法

徐蔚鸿，严金果

(长沙理工大学计算机与通信工程学院，湖南长沙 410114)

收稿日期:2013-01-17 修回日期:2014-04-27 出版日期:2014-07-25 发布日期:2014-07-25
基金资助:
教育部重点科研资助项目(208098)；湖南省科技计划资助项目(2012FJ30052)

Moving objects realtime detection algorithm
based on the improved background subtraction

XU Weihong，YAN Jinguo

(College of Computer & Communication Engineering,
Changsha University of Science and Technology,Changsha 410114,China)

Received:2013-01-17 Revised:2014-04-27 Online:2014-07-25 Published:2014-07-25

摘要/Abstract

摘要：

针对传统混合高斯建模算法计算量过大与目标轮廓清晰度小的问题，提出了一种新的运动目标实时检测算法。该算法引入三帧差分的方法，提高了检测目标轮廓的清晰度；通过HSI混合高斯建模前进行分块处理有效减小了计算量，因此算法的实时性有了明显的改善；并利用逻辑运算融合三帧差分与HSI混合高斯模型进行高效的背景提取；最后运用数学形态学方法进一步优化检测结果。实验结果表明，相比混合高斯模型经典算法，该算法能更快速、更准确地检测出智能监控视频序列中的运动目标，并且目标轮廓清晰度也有明显的改善。

关键词: 运动目标实时检测, 分块处理, HSI混合高斯模型, 三帧差分, 目标轮廓清晰度

Abstract:

The traditional mixture Gaussian models require much computation and have little clarity of objects’ contours. Therefore, a new moving objects realtime detection algorithm is proposed. Firstly, the threeframedifferencing method is introduced in the algorithm in order to improve the contours’ clarity of detection objects. Secondly, the computation is reduced by block processing in the Gaussian mixture models on HSI, so the realtime performance of the algorithm is improved. Thirdly, the threeframedifferencing method and the adaptive Gaussian mixture model on HSI are merged by logic computation so as to extract background efficiently. Finally, the detection result is optimized further by the mathematical morphology. The experimental results show the new algorithm can detect the moving objects in the surveillance video sequences faster and more accurately than the classic mixture Gaussian models and improve the clarity of objects’ contours.

Key words: moving object real-time detection;block processing;Gaussian mixture model on HSI;three-frame-differencing;clarity of objects&rsquo, contours

徐蔚鸿，严金果. 基于改进的背景差分的运动目标实时检测算法[J]. J4, 2014, 36(07): 1352-1356.

XU Weihong，YAN Jinguo. Moving objects realtime detection algorithm
based on the improved background subtraction [J]. J4, 2014, 36(07): 1352-1356.

参考文献

［1］Lopez S, Horstrand P, Callico G M, et al. A lowcomputationalcomplexity algorithm for hyperspectral endmember extraction:Modified vertex component analysis［C］∥IEEE Geoscience and Remote Sensing Letter,2012,9(3):502506.
［2］Gan Minggang, Chen Jie, Liu Jin, et al. Moving object detection algorithm based on threeframedifferencing and edge information［J］. Journal of Electronics & Information Technology, 2010, 32(4):894897.(in Chinese)
［3］Kartika I, Mohamed S S. Frame differencing with postprocessing techniques for moving object detection in outdoor environment［C］∥Proc of IEEE 7th International Colloquium on Signal Processing and its Applications,2011:172176.
［4］Stauffer C, Grimson W E L. Learning patterns of activity using realtime tracking［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000,22(8):747757.
［5］Lee D S. Effective Gaussian mixture learning for video background subtraction［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence,2005,27(5):827832.
［6］Li Bin,Zhong Runtian,Wang Xianji,et al.A continuous optimization algorithm based on progressive estimation of Gaussian mixture model［J］. Chinese Journal of Computers, 2007, 30(6):979985.(in Chinese)
［7］Jin Guangzhi, Shi Linsuo, Bai Xiangfeng, et al. New objects detection system based on mixture Gaussian model［J］. Journal of Computer Applications, 2011, 31(12):33603362.(in Chinese)
［8］Tsai DuMing, Lai ShiaChih. Independent component analysisbased background subtraction for indoor surveillance［J］.IEEE Transactions on Image Processing, 2009,18(1):158167.
［9］Zhang Qiujia, Wang Hong, Liu Wei, et al. Moving object detection optical flow vectors compensation model ［J］. Laser & Infraed, 2011, 41(11):12631266. (in Chinese)
［10］Zhang Zhenghua,Xu Ye,Su Quan,et al.Moving object detection based on background subtraction and hybrid frame differencing［J］. Journal of Signal and Information Processing, 2012, 42(8):1417. (in Chinese)
［11］Lin Qing, Xu Zhu, Wang Shitong, et al. Moving objects detection of adaptive Gaussian mixture models on HSV［J］. Computer Science, 2010, 37(10):254256. (in Chinese)
［12］Li Yinghong, Xiong Changzhen, Yin Yixin, et al. Moving object detection based on edged mixture Gaussian models［J］. Journal of System Simulation,2009,21(suppl 1):7274. (in Chinese)
［13］Stauffer C, Grimson W E L. Adaptive background mixture models for realtime tracking［C］∥ Proc of IEEE International Conference on Computer Vision and Pattern Recognition,1999:246252.
［14］Bailo G, Bariani M, Ijas P,et al. Background estimation with Gaussian distribution for image segmentation, a fast approach［C］∥Proc of IEEE International Workshop on Measurement Systems for Homeland Security, 2005:25.
附中文参考文献：
［2］甘明刚，陈杰，刘劲，等. 一种基于三帧差分和边缘信息的运动目标检测方法［J］. 电子与信息学报, 2010, 32(4):894897.
［6］李斌，钟润添，王先基，等. 一种基于递增估计GMM的连续优化算法［J］. 计算机学报, 2007, 30(6):979985.
［7］金广智，石林锁，白向峰，等. 基于混合高斯模型的新型目标检测［J］. 计算机应用,2011, 31(12):33603362.
［9］张秋佳，王虹，刘威，等. 光流向量补偿模型的运动目标检测［J］. 激光与红外,2011, 41(11):12631266.
［10］张正华，许晔，苏权，等. 基于背景差分和混合帧差的运动目标检测［J］. 信号与信息处理，2012，42(8)：1417.
［11］林庆，徐柱，王士同，等. HSV自适应混合高斯模型的运动目标检测［J］. 计算机科学，2010，37(10)：254256.
［12］李颖宏，熊昌镇，尹怡欣，等. 一种基于边缘高斯混合模型的运动目标检测方法［J］. 系统仿真学报，2009，21(Suppl 1)：7274.