基于边缘约束局部区域MRF的图像分割方法

doi:10.19678/j.issn.1000-3428.0056414

摘要/Abstract

摘要： 针对常规马尔科夫随机场（MRF）模型对复杂自然图像分割时，存在对噪声敏感且边缘模糊的问题，构建一种基于边缘约束局部区域MRF（ECLRMRF）的图像分割模型。利用欧氏距离度量局部区域内邻接像素的相似度，依据其相似度构建局部空间来约束高斯混合模型，有效描述丰富的局部区域统计特征，并建立MRF模型的局部区域一致性约束项。利用Canny边缘检测算子提取图像的边缘特征，并在分割过程中建立图像分割区域的边缘约束，通过在MRF模型框架下将局部区域统计特征和图像边缘特征相融合，解决局部区域MRF模型对图像分割边缘模糊的问题，再采用Gibbs采样算法实现对复杂自然图像的准确分割。实验结果表明，该模型能够更好地保留图像边缘信息，并且具有更好的分割效果。

关键词: 图像分割, 马尔科夫随机场, 局部区域一致性, 边缘约束, 高斯混合模型

Abstract: The conventional Markov Random Field(MRF) model is sensitive to noise and produces fuzzy edges when segmenting complex natural images.To address the problem, this paper proposes an Edge Constrained Local Region MRF(ECLRMRF) segmentation model.Euclidean distance is used to measure the similarity of adjacent pixels in the local region, and the local space is constructed according to the similarity to constrain the Gaussian Mixture Model(GMM), which can effectively describe the rich statistical features of the local region and establish the local region consistency constraints of the MRF model.Canny edge detection operator is used to extract the edge features of the image, and the edge constraints of the image segmentation region are established in the process of segmentation.By fusing the local region statistical features and image edge features in the framework of MRF model, the problem of blurring the edge of image segmentation in the local region MRF model is solved, and then Gibbs sampling algorithm is used to achieve accurate segmentation of complex natural images.Experimental results show that the model can better retain the edge information of the image and has better segmentation effect.

Key words: image segmentation, Markov Random Field (MRF), local region consistency, edge constrain, Gaussian Mixture Model(GMM)

中图分类号:

TP391

胡高珍, 徐胜军, 孟月波, 刘光辉, 冯峰, 段中兴. 基于边缘约束局部区域MRF的图像分割方法[J]. 计算机工程, 2021, 47(6): 253-261,270.

HU Gaozhen, XU Shengjun, MENG Yuebo, LIU Guanghui, FENG Feng, DUAN Zhongxing. Image Segmentation Method Based on MRF with Edge Constrained Local Region[J]. Computer Engineering, 2021, 47(6): 253-261,270.

https://www.ecice06.com/CN/Y2021/V47/I6/253

图/表 5

20210618184719

20210618184723

20210618184727

20210618184730

20210618184733

参考文献

[1] HU Can,FAN Wentao,DU Jixiang,et al.Model-based segmentation of image data using spatially constrained mixture models[J].Neurocomputing,2018,283:214-227.
[2] ZHAO Quanhua,SHI Xue,WANG Yu,et al.Remote sensing image segmentation based on spatially constrained Gaussian mixture model with unknown class number[J].Journal on Communications,2017,38(2):34-43.(in Chinese)赵泉华,石雪,王玉,等.可变类空间约束高斯混合模型遥感图像分割[J].通信学报,2017,38(2):34-43.
[3] XU Shengjun,HAN Jiuqiang,LIU Guanghui,et al.Image segmentation based on local spatial adaptive Markov random field model[J].Control and Decision,2013,28(6):889-893.(in Chinese)徐胜军,韩九强,刘光辉,等.基于局部空间自适应MRF模型的图像分割[J].控制与决策,2013,28(6):889-893.
[4] ZHANG Hui,WEN Tian,ZHENG Yuhui,et al.Two fast and robust modified Gaussian mixture models incorporating local spatial information for image segmentation[J].Signal Processing Systems,2015,81:45-58.
[5] SONG Yantao,JI Zexuan,SUN Quansen.Brain MR image segmentation algorithm based on Markov random field with image patch[J].Acta Automatica Sinica,2014(8):1754-1763.(in Chinese)宋艳涛,纪则轩,孙权森.基于图像片马尔科夫随机场的脑MR图像分割算法[J].自动化学报,2014(8):1754-1763.
[6] LÜ Miaomiao,SUN Jianming.Moving image target detection based on modified Gaussian mixture model[J].Semiconductor Optoelectronics,2019,40(6):874-878.(in Chinese)吕苗苗,孙建明.基于改进高斯混合模型的运动图像目标检测算法[J].半导体光电,2019,40(6):874-878.
[7] JI Zexuan,HUANG Yubo,SUN Quansen,et al.A spatially constrained generative asymmetric Gaussian mixture model for image segmentation[J].Journal of Visual Communication and Image Representation,2016,40:611-626.
[8] WANG Quan.HMRF-EM-image:implementation of the Hidden Markov random field model and its expectation-maximization algorithm[EB/OL].[2019-09-10].https://arxiv.org/abs/1207.3510.
[9] LIN Wenjie,LI Yu,ZHAO Quanhua.High-resolution remote sensing image segmentation using minimum spanning tree tessellation and RHMRF-FCM algorithm[J].Acta Geodaeticaet Cartographica Sinica,2019,48(1):64-74.(in Chinese)林文杰,李玉,赵泉华.结合MST划分和RHMRF-FCM算法的高分辨率遥感图像分割[J].测绘学报,2019,48(1):64-74.
[10] FENG Bao,CHEN Yehang,LIU Zhuangsheng,et al.Segmentation of breast cancer on DCE-MRI images with MRF energy and fuzzy speed function[J].Acta Automatica Sinica,2020,46(6):1188-1199.(in Chinese)冯宝,陈业航,刘壮盛,等.结合MRF能量和模糊速度的乳腺癌图像分割方法[J].自动化学报,2020,46(6):1188-1199.
[11] MENG Yuebo,LIU Guanghui,XU Shengjun,et al.Image segmantation method using multi-resolution Markov random field model with edge-preserving[J].Journal of Xi'an Jiaotong University,2019,53(3):1-9.(in Chinese)孟月波,刘光辉,徐胜军,等.一种具有边缘保持的多尺度马尔科夫随机场模型图像分割方法[J].西安交通大学学报,2019,53(3):1-9.
[12] SHI Xue,LI Yu,LI Xiaoli,et al.Gaussian mixture model with neighbor relationship for image segmentation and simplified solving method[J].Journal of Image and Graphics,2017,22(12):1758-1768.(in Chinese)石雪,李玉,李晓丽,等.融入邻域作用的高斯混合分割模型及简化求解[J].中国图象图形学报,2017,22(12):1758-1768.
[13] XIONG Taisong,ZHANG Lei,ZHANG Yi.Double Gaussian mixture model for image segmentation with spatial relationships[J].Journal of Visual Communication and Image Representation.2016,34:135-145.
[14] JI Zexuan,LIU Jinyao,YUAN Hengdong,et al.A rough set bounded spatially constrained asymmetric Gaussian mixture model for image segmentation[J].PLoS One,2017,12(1):697-708.
[15] LI S Z.Markov random field modeling in computer vision[EB/OL].[2019-09-10].https://xueshu.baidu.com/usercenter/paper/show?paperid=52f6ec28127160e610a2294729aeefec&site=xueshu_se.
[16] WANG Qingping,ZHAO Hongyu,WU Weiwei,et al.An adaptive Bayesian segmentation method fused of local and non-local information[J].Journal of Electronics & Information Technology,2014,36(4):1003-1007.(in Chinese)王青平,赵宏宇,吴微微,等.融合局部和非局部信息的自适应贝叶斯分割方法[J].电子与信息学报,2014,36(4):1003-1007.
[17] YU Miao,HU Zhanyi.Higher-order Markov random fields and their applications in scene understanding[J].Acta Automatica Sinica,2015,41(7):1213-1234.(in Chinese)余淼,胡占义.高阶马尔科夫随机场及其在场景理解中的应用[J].自动化学报,2015,41(7):1213-1234.
[18] ZHU Feng,LUO Limin,SONG Yuqing,et al.Adaptive spatially neighborhood information Gaussian mixture model for image segmentation[J].Journal of Computer Research and Development,2011,48(11):2000-2007.(in Chinese)朱峰,罗立民,宋余庆,等.基于自适应空间邻域信息高斯混合模型的图像分割[J].计算机研究与发展,2011,48(11):2000-2007.
[19] NGUYEN T M,WU Q M J.Fast and robust spatially constrained Gaussian mixture model for image segmentation[J].IEEE Transactions on Circuits and Systems for Video Technology,2013,23(4):621-635.
[20] CHELLAPPA R,CHATTERJEE S.Classification of textures using Gaussian Markov random fields[J].IEEE Transaction on Acoustics,Speech,and Signal Processing,1985,33(4):959-963.
[21] XU Shengjun,HAN Jiuqiang,ZHAO Liang,et al.Algorithm of minimizing local region energy for image segmentation[J].Journal of Xi'an Jiaotong University,2011,45(8):7-12.(in Chinese)徐胜军,韩九强,赵亮,等.用于图像分割的局部区域能量最小化算法[J].西安交通大学学报,2011,45(8):7-12.
[22] FENG Feng,XU Shengjun,MENG Yuebo,et al.Image segmentation based on high-order MRF model with robust local spatial information[C]//Proceedings of 2018 Chinese Automation Congress.Washington D.C.,USA:IEEE Press,2018:1-12.
[23] NIKOU C,GALATSANOS N,LIKAS A.A class-adaptive spatially variant mixture model for image segmentation[J].IEEE Transactions on Image Processing,2007,16(4):1121-1130.
[24] ARBELAEZ P,FOWLKES C,MARTIN D.The Berkeley segmentation dataset and benchmark[EB/OL].[2019-09-10].https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/bsds/.
[25] UNNIKRISHNAN R,PANTOFARU C,HEBERT M.A measure for objective evaluation of image segmentation algorithms[C]//Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2005:34-41.

选择文件类型/文献管理软件名称

选择包含的内容