结合SSFCM与随机游走的半监督图像分割算法
|
摘要
针对基于颜色特征空间的半监督聚类分割算法适合分割结果包含多个颜色特征相似目标的应用场合,但对高噪声图像却无法获得理想的分割结果,而基于随机游走理论的半监督图像分割算法需要用户对目标逐一进行标记的问题,提出一种半监督图像分割算法.首先根据用户标记采用半监督模糊C均值聚类(SSFCM)算法对图像颜色特征进行建模;然后引入一个确信度函数,并根据SSFCM算法得到的隶属度数据计算确信度函数值,再将像素分为2类,分别作为随机游走图像分割算法的已标记点和未标记点;最后采用随机游走算法完成最终的分割.实验结果表明,该算法对图像中的噪声具有良好的抑制作用,且无需用户对目标逐一进行标记.
Algorithms based on semi-supervised clustering are suitable to segment images containing a large amount of objects with the similar color features,but they cannot gain ideal effects to images containing noises;semi-supervised image segmentation algorithm based on random walks theory requires the user to label all objects contained in the image.A semi-supervised image segmentation algorithm to solve this problem is presented,which is based on integration of semi-supervised fuzzy c-means clustering algorithms with random walks.It models the image's color feature through semi-supervised c-means clustering algorithm(SSFCM) based label data,then it defines a reliability function based on the membership calculated by SSFCM,and the pixels are classified into two types that are considered as labeled and unlabeled pixels of Random Walks.The experimental results indicate that the algorithm not only reduces the noise sensitivity of SSFCM but also avoids cumbersome operations that the user labels the seed points of all objects for Random Walks.
Algorithms based on semi-supervised clustering are suitable to segment images containing a large amount of objects with the similar color features,but they cannot gain ideal effects to images containing noises;semi-supervised image segmentation algorithm based on random walks theory requires the user to label all objects contained in the image.A semi-supervised image segmentation algorithm to solve this problem is presented,which is based on integration of semi-supervised fuzzy c-means clustering algorithms with random walks.It models the image's color feature through semi-supervised c-means clustering algorithm(SSFCM) based label data,then it defines a reliability function based on the membership calculated by SSFCM,and the pixels are classified into two types that are considered as labeled and unlabeled pixels of Random Walks.The experimental results indicate that the algorithm not only reduces the noise sensitivity of SSFCM but also avoids cumbersome operations that the user labels the seed points of all objects for Random Walks.
引文
[1]Blake A,Rother C,Brown M,et at.Interactive image segmentation using an adaptive GMMRF model[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2004.3021:428-441
[2]Boykov Y Y,Jolly M P.Interactive graph cuts for optimal boundary and region segmentation of objects in ND images[C]//Proceedings of the 8th International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2001,1:105-112
[3]Chuang Y Y,Curless B,Salesin D H,et al.A Bayesian approach to digital matting[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los ALamitos:IEEE Computer Society Press,2001,2:264-271
[4]Li Y,Sun J,Tang C K,et al.Lazy snapping[J].ACM Transactions on Graphics,2004,23(3):303-308
[5]Perez P,Blake A,Gangnet M.JetStream:probabilistic contour extraction with particles[C]//Proceedings of the 8th IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2001,2:524-531
[6]Rother C,Kolmogorov V,Blake A.Grabcut:interactive foreground extraction using iterated graph cuts[J].ACM Transactions on Graphics,2004,23(3):309-314
[7]Ruzon M A,Tomasi C.Alpha estimation in natural images[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2000,1:18-25
[8]Sun J,Jia J Y,Tang C K,et al.Poisson matting[J].ACM Transactions on Graphics,2004,23(3):315-321
[9]Wang J,Cohen M F.An iterative optimization approach for unified image segmentation and matting[C]//Proceedings of the 10th IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2005,2:936-943
[10]Mortensen E N,Barrett W A.Intelligent scissors for image composition[C]//Computer Graphics Proceedings,Annaul Conference Series,ACM SIGGRAPH.New York:ACM Press,1995:191-198
[11]Mortensen E N,Barrett W A.Toboggan-based intelligent scissors with a four-parameter edge model[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,1999,2:452-458
[12]Mortensen E N,Barrett W A.Interactive segmentation with intelligent scissors[J].Graphical Models and Image Processing,1998,60(5):349-384
[13]Gleicher M.Image snapping[C]//Computer Graphics Proceedings,Annual Conference Series,ACM SIGGRAPH.New York:ACM Press,1995:183-190
[14]Kass M,Witkin A,Terzopoulos D.Snakes:active contour models[J].International Journal of Computer Vision,1988,1(4):321-331
[15]Luo Xiping,Tian Jie.A modified interactive segmentation of medical image series[J],Chinese Journal of Electronics,2003,31(1):29-32(in Chinese)(罗希平,田捷.一种改进的交互式医学图像序列分割方法[J].电子学报,2003,31(1):29-32)
[16]Wang Zheng,Yang Xin,Li Jun,et al.An interactive model for segmentation of medical image[J].Pattern Recognition and Artificial Intelligence,2002,15(4):392-396(in Chinese)(王峥,杨新,李俊,等.基于水平集人机交互模型的医学图像分割[J].模式识别与人工智能,2002,15(4):392-396)
[17]Xiang S M,Nie F P,Zhang C X,et al.Interactive natural image segmentation via spline regression[J],IEEE Transactions on Image Processing,2009,18(7):1623-1632
[18]Xiang S M,Nie F P,Zhang C S.Semi-supervised classification via local spline regression[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(11):2039-2053
[19]Delong A,Gorelick L,Schmidt F R,et al.Interactive segmentation with super-labels[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2011,6819;147—162
[20]Liu D D,Pulli K,Shapiro L G,et al.Fast interactive image segmentation by discriminative clustering[C]//Proceedings of the ACM Multimedia Workshop on Mobile Cloud Media Computing.New York:ACM Press,2010:47-52
[21]Zhang X,TianJ,Xiang D H,et al.Interactive segmentation method with graph cut and SVMs[C]//Proceedings of the SPIE.Bellingham:Society of Photo-Optical Instrumentation Engineers,2010,7623:76234M-76234M-8
[22]Yang Nanhai,Huang Mingming,He Ran.et al.Robust semi-supervised learning algorithm based on maximum correntropy Criterion[J].Chinese Journal of Software,2012,23(2):279-288(in Chinese)(杨南海,黄明明,赫然,等.基于最大相关熵准则的鲁棒半监督学习算法[J].软件学报,2012,23(2):279-288)
[23]Freedman D,Zhang T.Interactive graph cut based segmentation with shape priors[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2005,1:755-762
[24]Vicente S,Kolmogorov V,Rother C.Graph cut based image segmentation with connectivity priors[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2008:1-8
[25]Liu Chen,Li Fengxia,Zhang Yan.An interactive object cutout algorithm based on graph-cut and generalized shape prior[J].Journal of Computer-Aided Design&Computer Graphic,2009,21(12):1753-1760(in Chinese)(刘陈,李凤霞,张艳.基于图割与泛形信息的对象分割方法[J].计算机辅助设计与图形学学报,2009,21(12):1753-1760)
[26]Zheng Jiaming,Chen Zhaojiong.Connectivity constrained Graph-Cut for fast interactive image segmentation[J].Journal of Computer-Aided Design&Computer Graphic,(郑加明,陈昭炯.带连通性约束的快速交互式Graph-Cut算法[J].计算机辅助设计与图形学学报,2011,23(3):399-405)
[27]Grady L.Random walks for image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(11):1768-1783
[28]Grady L,Funka-Lea G.Multi-label image segmentation for medical applications based on graph-theoretic electrical potentials[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2004,3117:230-245
[29]Grady L,Schiwietz T,Aharon S,et al.Random walks for interactive organ segmentation in two and three dimensions:Implementation and validation[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2005,3750:773-780
[30]Zhang L,Ji Q.A Bayesian network model for automatic and interactive image segmentation[J].IEEE Transactions on Image Processing,2011,20(9):2582-2593
[31]Shentai N,Hertz T,Bar-Hilel A,et al.Computing Gaussian mixture models with EM using equivalence constraints[M]//Advances in Neural Information Processing Systems,16.Cambridge:MIT Press,2003
[32]Qian Y T,Si W W.A semi-supervised color image segmentation method[C]//Proceedings of IEEE International Conference on Image Processing.Los Alamitos:IEEE Computer Society Press,2005,2:1194-1197
[33]Pedrycz W.Algorithms of fuzzy clustering with partial supervision[J].Pattern Recognition Letters,1985,3(1):13-20
[34]Bouchachia A,Pedrycz W.Data clustering with partial supervision[J].Data Mining and Knowledge Discovery,2006,12(1):47-78
[35]Pedrycz W,Waletzky J.Fuzzy clustering with partial supervision[J].IEEE Transactions on Systems,Man,and Cybernetics,Part B:Cybernetics,1997,27(5):787-795
[36]Pedrycz W,Waletzky J.Neural-network front ends in unsupervised learning[J].IEEE Transactions on Neural Networks,1997,8(2):390-401
[37]Doyle P G,Snell J L.Random walks and electric networks[M].Ann Arbor:The Mathematical Association of America,1984
[2]Boykov Y Y,Jolly M P.Interactive graph cuts for optimal boundary and region segmentation of objects in ND images[C]//Proceedings of the 8th International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2001,1:105-112
[3]Chuang Y Y,Curless B,Salesin D H,et al.A Bayesian approach to digital matting[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los ALamitos:IEEE Computer Society Press,2001,2:264-271
[4]Li Y,Sun J,Tang C K,et al.Lazy snapping[J].ACM Transactions on Graphics,2004,23(3):303-308
[5]Perez P,Blake A,Gangnet M.JetStream:probabilistic contour extraction with particles[C]//Proceedings of the 8th IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2001,2:524-531
[6]Rother C,Kolmogorov V,Blake A.Grabcut:interactive foreground extraction using iterated graph cuts[J].ACM Transactions on Graphics,2004,23(3):309-314
[7]Ruzon M A,Tomasi C.Alpha estimation in natural images[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2000,1:18-25
[8]Sun J,Jia J Y,Tang C K,et al.Poisson matting[J].ACM Transactions on Graphics,2004,23(3):315-321
[9]Wang J,Cohen M F.An iterative optimization approach for unified image segmentation and matting[C]//Proceedings of the 10th IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2005,2:936-943
[10]Mortensen E N,Barrett W A.Intelligent scissors for image composition[C]//Computer Graphics Proceedings,Annaul Conference Series,ACM SIGGRAPH.New York:ACM Press,1995:191-198
[11]Mortensen E N,Barrett W A.Toboggan-based intelligent scissors with a four-parameter edge model[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,1999,2:452-458
[12]Mortensen E N,Barrett W A.Interactive segmentation with intelligent scissors[J].Graphical Models and Image Processing,1998,60(5):349-384
[13]Gleicher M.Image snapping[C]//Computer Graphics Proceedings,Annual Conference Series,ACM SIGGRAPH.New York:ACM Press,1995:183-190
[14]Kass M,Witkin A,Terzopoulos D.Snakes:active contour models[J].International Journal of Computer Vision,1988,1(4):321-331
[15]Luo Xiping,Tian Jie.A modified interactive segmentation of medical image series[J],Chinese Journal of Electronics,2003,31(1):29-32(in Chinese)(罗希平,田捷.一种改进的交互式医学图像序列分割方法[J].电子学报,2003,31(1):29-32)
[16]Wang Zheng,Yang Xin,Li Jun,et al.An interactive model for segmentation of medical image[J].Pattern Recognition and Artificial Intelligence,2002,15(4):392-396(in Chinese)(王峥,杨新,李俊,等.基于水平集人机交互模型的医学图像分割[J].模式识别与人工智能,2002,15(4):392-396)
[17]Xiang S M,Nie F P,Zhang C X,et al.Interactive natural image segmentation via spline regression[J],IEEE Transactions on Image Processing,2009,18(7):1623-1632
[18]Xiang S M,Nie F P,Zhang C S.Semi-supervised classification via local spline regression[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(11):2039-2053
[19]Delong A,Gorelick L,Schmidt F R,et al.Interactive segmentation with super-labels[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2011,6819;147—162
[20]Liu D D,Pulli K,Shapiro L G,et al.Fast interactive image segmentation by discriminative clustering[C]//Proceedings of the ACM Multimedia Workshop on Mobile Cloud Media Computing.New York:ACM Press,2010:47-52
[21]Zhang X,TianJ,Xiang D H,et al.Interactive segmentation method with graph cut and SVMs[C]//Proceedings of the SPIE.Bellingham:Society of Photo-Optical Instrumentation Engineers,2010,7623:76234M-76234M-8
[22]Yang Nanhai,Huang Mingming,He Ran.et al.Robust semi-supervised learning algorithm based on maximum correntropy Criterion[J].Chinese Journal of Software,2012,23(2):279-288(in Chinese)(杨南海,黄明明,赫然,等.基于最大相关熵准则的鲁棒半监督学习算法[J].软件学报,2012,23(2):279-288)
[23]Freedman D,Zhang T.Interactive graph cut based segmentation with shape priors[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2005,1:755-762
[24]Vicente S,Kolmogorov V,Rother C.Graph cut based image segmentation with connectivity priors[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2008:1-8
[25]Liu Chen,Li Fengxia,Zhang Yan.An interactive object cutout algorithm based on graph-cut and generalized shape prior[J].Journal of Computer-Aided Design&Computer Graphic,2009,21(12):1753-1760(in Chinese)(刘陈,李凤霞,张艳.基于图割与泛形信息的对象分割方法[J].计算机辅助设计与图形学学报,2009,21(12):1753-1760)
[26]Zheng Jiaming,Chen Zhaojiong.Connectivity constrained Graph-Cut for fast interactive image segmentation[J].Journal of Computer-Aided Design&Computer Graphic,(郑加明,陈昭炯.带连通性约束的快速交互式Graph-Cut算法[J].计算机辅助设计与图形学学报,2011,23(3):399-405)
[27]Grady L.Random walks for image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(11):1768-1783
[28]Grady L,Funka-Lea G.Multi-label image segmentation for medical applications based on graph-theoretic electrical potentials[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2004,3117:230-245
[29]Grady L,Schiwietz T,Aharon S,et al.Random walks for interactive organ segmentation in two and three dimensions:Implementation and validation[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2005,3750:773-780
[30]Zhang L,Ji Q.A Bayesian network model for automatic and interactive image segmentation[J].IEEE Transactions on Image Processing,2011,20(9):2582-2593
[31]Shentai N,Hertz T,Bar-Hilel A,et al.Computing Gaussian mixture models with EM using equivalence constraints[M]//Advances in Neural Information Processing Systems,16.Cambridge:MIT Press,2003
[32]Qian Y T,Si W W.A semi-supervised color image segmentation method[C]//Proceedings of IEEE International Conference on Image Processing.Los Alamitos:IEEE Computer Society Press,2005,2:1194-1197
[33]Pedrycz W.Algorithms of fuzzy clustering with partial supervision[J].Pattern Recognition Letters,1985,3(1):13-20
[34]Bouchachia A,Pedrycz W.Data clustering with partial supervision[J].Data Mining and Knowledge Discovery,2006,12(1):47-78
[35]Pedrycz W,Waletzky J.Fuzzy clustering with partial supervision[J].IEEE Transactions on Systems,Man,and Cybernetics,Part B:Cybernetics,1997,27(5):787-795
[36]Pedrycz W,Waletzky J.Neural-network front ends in unsupervised learning[J].IEEE Transactions on Neural Networks,1997,8(2):390-401
[37]Doyle P G,Snell J L.Random walks and electric networks[M].Ann Arbor:The Mathematical Association of America,1984