[1]刘光辉,任庆昌,孟月波,等.基于多尺度局部区域能量最小化算法的图像分割[J].西安建筑科技大学学报:自然科学版,2014,(04):588-592.[doi:10.15986/j.1006-7930.2014.04.024]
 LIU Guanghui,REN Qingchang,MENG Yuebo,et al.Algorithm of minimizing multi-scale local region energy for image segmentation[J].J.Xi’an Univ. of Arch. & Tech.:Natural Science Edition,2014,(04):588-592.[doi:10.15986/j.1006-7930.2014.04.024]
点击复制

基于多尺度局部区域能量最小化算法的图像分割()
分享到:

西安建筑科技大学学报:自然科学版[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2014年04期
页码:
588-592
栏目:
出版日期:
2014-08-31

文章信息/Info

Title:
Algorithm of minimizing multi-scale local region energy for image segmentation
文章编号:
1006-7930(2014)04-0588-05
作者:
刘光辉任庆昌孟月波徐胜军
(西安建筑科技大学信息与控制工程学院,陕西 西安 710055)
Author(s):
LIU Guanghui REN Qingchang MENG Yuebo XU Shengjun
(School of Information and Control Engineering Xi’an Univ. of Arch. and Tech., Xi’an, 710055, China)
关键词:
图像分割多尺度马尔可夫随机场置信度传播算法最大后验边缘准则
Keywords:
image segmentation Multi-scale Markov random fields belief propagation algorithm maximized the posteriori marginal(MPM)
分类号:
TP391.41
DOI:
10.15986/j.1006-7930.2014.04.024
文献标志码:
A
摘要:
常规多尺度MRF模型中固定的四叉树结构造成图像分割结果中常产生块现象和非连续边缘.为解决这一问题,提出了一种新的多尺度MRF 模型,并建立了基于区域消息传递的置信度传播(BP)算法,通过BP 算法在多尺度MRF 模型中对区域消息进行传递;在层间,从粗糙层向精细层进行消息传递时,利用提出的MRF模型父子区域之间的重叠,有效初始化了精细层消息的初值,避免了多尺度MRF 模型层间误分类的传递;最后基于MPM 准则对分割结果进行估计.实验结果表明提出的算法不仅得到了更准确的图像分割结果,而且具有较快的分割速度.
Abstract:
An efficient local region belief propagation (BP) algorithm based on multi-scale Markov random fields (MRF) model is proposed to solve the problem that the fixed quadtree structures of traditional hierarchical Markov random fields always results in blocky artifacts and discontinuous edges in image segmentation. The proposed algorithm builds different scales of local region messages, and the messages are propagated on the proposed MRF model through belief propagation (BP) algorithm. The proposed algorithm utilities the overlapping between parent and child regions efficiently to initialize the child region messages when passing messages from the coarser scale to the finer scale, thus avoiding the transfer of misclassification between scales in multi-scale MRF model. Finally segmentation results are estimated based on the maximized posteriori marginal (MPM) criterion. Experimental results on a wide variety of images have verified the effectiveness of the proposed algorithm.

参考文献/References:

[1] TERZOPOULOS D. Image analysis using multigrid relaxation methods [J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 1986, 8(2): 129-139.

[2] Bouman C, Shapiro M. A multiscale random field model for bayesian image segmentation[J]. IEEE Transactions on Image processing, 1994, 3(2): 162-177.

[3] WANG Xili, LIU Fang, JIAO Licheng. A hierarchical markov image model and its inference algorithm[J]. Journal of software, 2003, 14(9): 1558-1563.

[4] ROMBERY J K, CHOI H, BARANIUK R G. Bayesian tree-structured image modeling using wavelet-domain hidden Markov models[J]. IEEE Transactions on Image Proceeding, 2001, 10(7): 1056-1068.

[5] GRIMAND M. A new-measure of contrast-dynamics[C]. Proceedings of image A algebra and Morphological Image Processing Ⅲ. San Dicgo, USA: SPIE 1992, 292-305.

[6] Felzenszwalb P., Huttenlocher D. Efficient Belief Propagation for Early Vision [J]. International Journal of Computer Vision, 2006, 70(1): 41-54.

[7] YANG Qingxiong, WANG liang. Stereo matching with color-weighted correlation, hierarchical belief propagation, and occlusion handling [J]. IEEE Trans tractions on Pattern Analysis and Machine Intelligence. 2009, 31(3): 1-12.

[8] FAN Chen, Kazuyuki Tanaka, Tsuyoshi Horiguchi. Image segmentation based on bethe approximation for Gaussian mixture model [J]. Interdisciplinary Information Sciences, 2005, 11(1): 17-29.

[9] 杨勇,孙洪,何楚.基于区域确定的分层马尔可夫模型及其 MPM 算法[J].自动化学报,2007,33(7): 693-697.

YANG Yong, SUN Hong, HE Chu. A region-determined hierarchical Markov Model and its MPM algorithm[J]. ACTA Automatica Sinica, 2007, 33(7): 693-697.

[10] XIONG Lian, WANG Fei, ZHANG Changshui, Multilevel belief propagation for fast inference on Markov random fields[C]//2007 seventh IEEE International conference on Data Mining,Omaha,Nebraska,USA,371-380.

[11] ALEXANDER T. Ihler, SUDDERTH Erik B., FREEMAN William T. et al. Efficient Multiscale Sampling from Products of Gaussian Mixtures[C]//Advances in Neural Information Processing Systems 16, S. Thrun, L. Saul, and B. Scholkopf, Eds.Cambridge, MA: MIT Press, 2004.

[12] COMER Mary L. , DELP Edward J.. The EM/MPM algorithm for segmentation of textured images: analysis and further experimental results[J]. IEEE Transactions on Image Processing, 2000, 9(10): 1731-1744.

备注/Memo

备注/Memo:
收稿日期:2013-10-11 修改稿日期:2014-07-28
基金项目:陕西省自然科学基金项目(2012JM8026、2013JM8030);陕西省教育厅专项基金项目(2013JK1091);陕西省社会发展攻关项目(2013K13-04-08); 西安建筑科技大学基础研究基金项目(JC1415)
作者简介:刘光辉(1976-),男,博士生;主要从事智能建筑与图像处理方面的研究.E-mail:guanghuil@163.com
更新日期/Last Update: 2015-10-06