计算机科学 ›› 2014, Vol. 41 ›› Issue (7): 318-321.doi: 10.11896/j.issn.1002-137X.2014.07.066

• 图形图像与模式识别 • 上一篇    下一篇

基于改进高斯混合模型的运动目标检测方法

程全,马军勇   

  1. 郑州大学信息工程学院 郑州450001;周口师范学院物理与电子工程学院 周口466001;光电控制技术重点实验室 洛阳471009;中航工业洛阳电光设备研究所 洛阳471009
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受河南省科技厅重点科技攻关资助

Moving Target Detection Method Based on Gaussian Mixture Model

CHENG Quan and MA Jun-yong   

  • Online:2018-11-14 Published:2018-11-14

摘要: 基于高斯混合模型,提出了一种自适应的运动目标检测算法。首先,根据各像素点的像素值的集中程度,自适应地选择高斯分布的个数对背景模型进行学习与更新,再通过背景差分获取差分图像;其次,在对图像二值化的过程中,提出了一种改进的自动调整阈值的方法,用以对差分图像的像素进行分类后分别进行阈值化分割,这样就能得到前景目标;接着采用形态学重构的方法对阴影进行有效消除,从而使前景目标分割的效果得到有效的提高。实验证明,该方法具有较好的鲁棒性和检测效果,同时也具有较好的自适应性,特别是在检测目标本身灰度变化比较大等特殊情况下,更能体现出本算法的优越性。

关键词: 混合模型,图像,形态学 中图法分类号TP391文献标识码A

Abstract: Based on Gaussian mixture model,this paper put forward an adaptive moving target detection algorithm.First of all,according to the intensity of pixel values of each pixel,the number of Gaussian distribution is adaptively selected to learn and update background model,again get difference image by background subtraction.Second,in the process of image binarization,an improved method of automatic threshold is proposed to separately classifly difference image pixels after threshold segmentation,so future goal can be obtained.Then the method of morphological reconstruction is adopted to effectively eliminate the shadow,significantly improving prospect target segmentation effect.Experiment proves that the method has good robustness and detection effect,as well as good adaptability.Especially when the grayscale change of detection target itself is large,the superiority of the algorithm is more obvious.

Key words: Hybrid model,Image,Morphology

[1] Zhang R,Zhang S Z,Yu S Y.Moving objects detection method based on brightness distortion and chromaticity distortion[J].IEEE Transactions on Consumer Electronics,2007,53(3):1177-1185
[2] 杨莉,张荣国,胡静,等.基于C-V模型的运动目标水平集提取方法[J].太原科技大学学报,2012,33(4):254-258
[3] Papenberg N,Bruhn A,Brox T.Highly accurate optic flow computation with theoretically justified warping[J].International Journal of Computer Vision,2006,(2):141-158
[4] Stein F.Efficient computation of optical flow using the censustransform[C]∥DAGM 2004,LNCS 3175.Berlin:Springer,2004:79-86
[5] Elgammal A,Harwood D,Davis L.Nonparametric model forbackground subtraction[C]∥Proceedings of European Conference on Computer Vision.2000:751-767
[6] Mandellos N A,Keramitsoglou I,Kiranoudis C T.A background subtraction algorithm for detecting and tracking vehicles [J].Expert Systems with Applications,2011,38(3):1619-1631
[7] Stauffer C,Grimson W.Adaptive background mixture modelsfor real time tracking[C]∥Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition.1999:246-252
[8] Stauffer C,Grimson W E L.Learning patterns of activity using real-time tracking[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2000,22(8):747-757
[9] 白向峰,李艾华,李喜来,等.新型高斯混合模型[J].中国图象图形学报,2011,16(6):983-988
[10] 夏正友,陈临强.基于快速的高斯混合模型的运动目标检测算法[J].杭州电子科技大学学报,2011,31(2):58-61
[11] 陈明生,梁光明,孙即祥,等.利用时空背景模型的快速运动目标检测方法[J].中国图象图形学报,2011,6(6):1002-1007
[12] 何亮明,覃荣华,巩思亮,等.动态场景中的改进混合高斯背景模型[J].计算机工程,2012,8(38):10-15
[13] Zhao Qin-pei.Random swap EM algorithm for Gaussian mixture models[J].Pattern Recognition Letters,2012,16(33):2120-2126
[14] 张恒,胡文龙,丁赤飚.一种自适应学习的混合高斯模型视频目标检测算法[J].中国图象图形学报,2010,15(4):631-636
[15] 胡元文,王平,高颖慧.基于水平集的运动目标轮廓提取方法[J].重庆理工大学学报:自然科学版,2012,26(10):51-61

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!