基于二次分水岭和归一化割的彩色图像分割技术的研究
|
摘要
图像分割可看作是一种把已获得的图像区分成若干个子图的技术,子图被认为是有意义的区域并且区域与区域之间必然是互不重合的。
近年来,图论(Graph Theory)是研究图像分割的一个热点方面。它把图像分割的问题转化成更为灵活的图划分问题,成为全局最优的选取。归一化割(Normalized cut, Ncut)是基于图论的图像分割方法当中的一种,被看作规范化的准则。它可以融合输入图像的不同特征,同时可计算分割得到的不同目标区域之间的差异性和同一目标区域内部的相似性。但是,Ncut本身仍是存在某些问题,譬如说计算量大、过分割以及欠分割。其中,计算量大的实际问题会影响Ncut在现实中的应用。
分水岭是一种很成熟的分割方法,算法比较简单,有很快的计算速度。而且处理之后的物体轮廓线封闭,能准确地定位,对微弱边缘也能得到良好的响应。它的缺点是存在着过分割的现象。
针对两者的优点和缺点,本文提出一种将二次分水岭和Ncut相结合的图像分割方法。先利用二次分水岭对输入图像进行预处理,将得到的小区域看做像素点做为Ncut算法的输入。这种方法有效地解决了Ncut方法计算量大的问题。
本文重点研究内容：
(1)提出用二次分水岭和Ncut相结合的图像分割方法。用二次分水岭对输入图像进行预处理,将得到的每个小区域看做一个像素点。在此基础上构造加权图,做为Ncut算法的输入。实验证明,本文方法分割结果良好,速度很快。
(2)为有效地解决二次分水岭分割结果当中的“0”边界,本文提出几种去边界方法,并通过实验获得最优方案,即将标记为0的像素点划分到与其八邻域相似度最大的点所在的小区域。
(3)采用与欧氏距离联系密切的LUV颜色模型和空间位置信息构造权值矩阵,探索已有构造方法造成分割效果存在差异的原因,并通过实验验证。
(4)用二次分水岭和Ncut相结合的分割方法先后与传统的Ncut算法和基于meanshift和Ncut结合的分割方法对比。选取多组图像,实验结果表明,与传统的方法相比本文在运行时间和分割效果方面都有显著提高；与基于nean shift和Ncut结合的分割方法相比,分割效果相近,但是本文方法耗时更少。
Image segmentation is a technology that can divide an acquired image into a number of significant areas, and also, the areas are not overlapped.
In recent years, graph theory is a hot research area of image segmentation. It transforms image segmentation into a flexible graph partitioning problem that selects information form the global graph. Ncut (Normalized cut) based on graph theory is one of the image segmentation methods. It is a standardized criterion. It can merge the different characteristics of the input images, and calculate the difference of the nodes among different small areas and similarity of the nodes within the same target region. However, there are still some problems of Ncut. For instance, it has large computation, over-segmentation and under segmentation. Large computation will affect the application of Ncut in reality.
The watershed is a well developed algorithm which has the advantages of fast computing speed, and closed outline of objects. It also can accurately position and get a good response of the weak edge. However, it has a drawback of over-segmentation.
In this paper, a method combining second watershed and Ncut algorithm is proposed to make use of their advantages and overcome their shortcomings. The second watershed is used to preprocess the input image. In this case, small areas are got and set as the input of Ncut algorithm, so the computing time of the method is greatly reduced.
The main contents of this paper are as follows.
Firstly, we combine the second watershed and Ncut algorithm by considering each micro segment after the second watershed as a node in the graph. Since the number of nodes in the graph is reduced, the total running time of the segmentation is greatly reduced. In addition, experiments also show that the combined method also obtains better partitioning results compared with the Ncut algorithm.
Secondly, to solve the problem of border line (labeled by zeros) generated by the second watershed, several methods are proposed and fully discussed. We first look for the maximal similar pixel among the neighboring pixels (eight-neighborhood) of the zero's corresponding pixel, and then the zeros are classified as the areas which the maximal similar pixels belong to.
Thirdly, we construct the weight matrix based on the color information of LUV system and distance information. Moreover, the partitioning differences caused by existing weight matrix methods are analyzed and tested.
Fourthly, the combined scheme proposed in this paper is campared with the traditional Ncut algorithm and the mean shift Ncut method separately. Simulation results prove that our scheme not only greatly reduces the running time but also acquires good segmentation effect campared with the traditional Ncut algorithm. And our plan runs faster than the mean shift Ncut method as well.
近年来,图论(Graph Theory)是研究图像分割的一个热点方面。它把图像分割的问题转化成更为灵活的图划分问题,成为全局最优的选取。归一化割(Normalized cut, Ncut)是基于图论的图像分割方法当中的一种,被看作规范化的准则。它可以融合输入图像的不同特征,同时可计算分割得到的不同目标区域之间的差异性和同一目标区域内部的相似性。但是,Ncut本身仍是存在某些问题,譬如说计算量大、过分割以及欠分割。其中,计算量大的实际问题会影响Ncut在现实中的应用。
分水岭是一种很成熟的分割方法,算法比较简单,有很快的计算速度。而且处理之后的物体轮廓线封闭,能准确地定位,对微弱边缘也能得到良好的响应。它的缺点是存在着过分割的现象。
针对两者的优点和缺点,本文提出一种将二次分水岭和Ncut相结合的图像分割方法。先利用二次分水岭对输入图像进行预处理,将得到的小区域看做像素点做为Ncut算法的输入。这种方法有效地解决了Ncut方法计算量大的问题。
本文重点研究内容：
(1)提出用二次分水岭和Ncut相结合的图像分割方法。用二次分水岭对输入图像进行预处理,将得到的每个小区域看做一个像素点。在此基础上构造加权图,做为Ncut算法的输入。实验证明,本文方法分割结果良好,速度很快。
(2)为有效地解决二次分水岭分割结果当中的“0”边界,本文提出几种去边界方法,并通过实验获得最优方案,即将标记为0的像素点划分到与其八邻域相似度最大的点所在的小区域。
(3)采用与欧氏距离联系密切的LUV颜色模型和空间位置信息构造权值矩阵,探索已有构造方法造成分割效果存在差异的原因,并通过实验验证。
(4)用二次分水岭和Ncut相结合的分割方法先后与传统的Ncut算法和基于meanshift和Ncut结合的分割方法对比。选取多组图像,实验结果表明,与传统的方法相比本文在运行时间和分割效果方面都有显著提高；与基于nean shift和Ncut结合的分割方法相比,分割效果相近,但是本文方法耗时更少。
Image segmentation is a technology that can divide an acquired image into a number of significant areas, and also, the areas are not overlapped.
In recent years, graph theory is a hot research area of image segmentation. It transforms image segmentation into a flexible graph partitioning problem that selects information form the global graph. Ncut (Normalized cut) based on graph theory is one of the image segmentation methods. It is a standardized criterion. It can merge the different characteristics of the input images, and calculate the difference of the nodes among different small areas and similarity of the nodes within the same target region. However, there are still some problems of Ncut. For instance, it has large computation, over-segmentation and under segmentation. Large computation will affect the application of Ncut in reality.
The watershed is a well developed algorithm which has the advantages of fast computing speed, and closed outline of objects. It also can accurately position and get a good response of the weak edge. However, it has a drawback of over-segmentation.
In this paper, a method combining second watershed and Ncut algorithm is proposed to make use of their advantages and overcome their shortcomings. The second watershed is used to preprocess the input image. In this case, small areas are got and set as the input of Ncut algorithm, so the computing time of the method is greatly reduced.
The main contents of this paper are as follows.
Firstly, we combine the second watershed and Ncut algorithm by considering each micro segment after the second watershed as a node in the graph. Since the number of nodes in the graph is reduced, the total running time of the segmentation is greatly reduced. In addition, experiments also show that the combined method also obtains better partitioning results compared with the Ncut algorithm.
Secondly, to solve the problem of border line (labeled by zeros) generated by the second watershed, several methods are proposed and fully discussed. We first look for the maximal similar pixel among the neighboring pixels (eight-neighborhood) of the zero's corresponding pixel, and then the zeros are classified as the areas which the maximal similar pixels belong to.
Thirdly, we construct the weight matrix based on the color information of LUV system and distance information. Moreover, the partitioning differences caused by existing weight matrix methods are analyzed and tested.
Fourthly, the combined scheme proposed in this paper is campared with the traditional Ncut algorithm and the mean shift Ncut method separately. Simulation results prove that our scheme not only greatly reduces the running time but also acquires good segmentation effect campared with the traditional Ncut algorithm. And our plan runs faster than the mean shift Ncut method as well.
引文
[1]谭耀麟.图像信息系统原理.第1版.清华大学出版社,2006：1-3
[2]章毓晋.图像工程.第2版.清华大学出版社,2005：74-75
[3]McInerney Tim, Terzopoulos Demetri. Deformable models in medical image analysis. Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis, 1996:171-180
[4]付曼,黄祥林,高芸.文档图像分割技术研究.中国传媒大学学报(自然科学版).2006,13(4)：62-67
[5]楚亚蕴,詹小四,孙兆才等.一种结合方向信息的指纹图像二值化算法.中国图像图形学报.2006,11(6)：854-860
[6]朱胜利.Mean Shift及相关算法在视频跟踪中的研究.浙江大学博士学位论文.2006：13-30,13-24
[7]Tian B, Shaikh M A, Azimi-Sadjadi M R, etal. A study of cloud classification with neural networks using spectral and textural features. IEEE Trans. Neural Networks, 1999,10(1):138-151
[8]Xu Jinming, Zhao Xiaobo. Determination of fluid inclusion lines using edge-detection technique. Chinese Journal of Rock Mechanics and Engineering,2007,26:1132-1137
[9]S Osher, J A Sethian. Fronts propagating with curvature dependent speed:Algorithms based on Hamilton-Jacobi formulation. Journal of Computational Physics,1988,79 (1): 12-49
[10]N. Otsu. A Threshold Selection Method from Gray-Level Histogram. IEEE Trans. Systems Man, and Cybernetics,1979,9:62-66
[11]T Pun. A new method for gray-level picture thresholding using the entropy of the histogram. In the Signal Processing,1980,2:223-237
[12]J Kittler and J Illingworth. Minimum error thresholding. Pattern Recognition,1986, 19(1):41-47
[13]Kittler J, Illingworth J, Foglein J, Paler K. An automatic thresholding algorithm and its performance. In the Pattern recognition.1984:287-289
[14]Otsu N. A threshold selection method from gray-level histograms. IEEE Trans. Systems Man and Cybernetics,1979,9(1):62-66
[15]Abutaleb A S. Automatic thresholding of gray-level pictures using two-dimensional entropies. Pattern Recognition,1989,47(1):22-32
[16]王双,焦李成等.阈值法标记提取的分水岭SAR图像分割方法.中国专利：CN101556693,2009：10-14
[17]蒋艳会,李峰.基于混沌粒子群算法的多阈值图像分割.计算机工程与应用.2010,46(10)：175-176
[18]J Canny. A computational approach to edge detection. In the Pattern Analysis and Machine Intelligence, IEEE Transactions,1986:679-698
[19]Nilsson N J. Principles of Artificial Intelligence. Palo Alto:Tioga Publishing Co,1980: 61-94
[20]Kirsch R. Computer determination of the constituent structure of biological images. In the Comuter Biomedical Research,1971,4:315-328
[21]Marr D, Hildreth E. Theory of edge detection. Proc. R. Soc. London,1980, B207: 187-217
[22]张闯,迟健男等.基于边缘检测与双边滤波的彩色图像去噪.电子学报.2010,38(8)：1776-1783
[23]Kass M, Witkin A, Terzopoulos D. Snakes:Active contour models. International Journal of Computer Vision,1988,1(4):321-331
[24]R Adams, L Bischof. Seeded region growing. IEEE Transactions on Pattern Analysis and Machine Intelligence,1994,16(6):641-647
[25]Frank Y, Shih, Shouxian Cheng. Automatic seeded region growing for color image segmentation. Image and Vision Computing,2005(23):877-886
[26]A Mehnert, P Jackway. An improved seeded region growing algorithm. Pattern Recognition Letters,1997(18):1065-1071
[27]Vincent L, Soille P. Watersheds in digital spaces:An efficient algorithm based on immersion simulations. IEEE Transactions on Pattern Analysis and Machine Intelligence,1991,13(6):583-598
[28]孙惠泉.图论及其应用.科学出版社,2004：1-2
[29]keinosuke Fukunaga, Larry D. Hostetler. The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition. IEEE Transactions on information theory,1975,21(1)：32-40
[30]钱芸,张英杰.水平集的图像分割方法综述.中国图像图形学报.2008,13(1)：7-13
[31]耿素云,屈婉玲,张立昂.离散数学.北京：清华大学出版社,1999：147-162
[32]赵继东,鲁珂,吴跃.一种基于谱图理论的Web图像搜索方法.计算机应用研究.2008,25(5)：1598-1600
[33]谭志明.基于图论的图像分割及其嵌入式应用研究.上海交通大学博士学位论文.2007：16-24
[34]C Zahn. Graph Theoretical Methods for Detecting and Describing Gestalt Clusters. IEEE Transactions on Computation,1971,20:68-86
[35]Z Wu and R Leahy. An Optimal Graph Theoretic Approach to Data Clustering:Theory and Its Application to Image Segmentation. IEEE PAMI,1993,11:1101-1113
[36]R Urquhart. Graph Theoretical Clustering Based on Limited Neighborhood sets. Pattern Recognition,1982,15(3):173-187
[37]J. Shi and J. Malik. Normalized Cuts and Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):888-905
[38]Wenbing Tao, Hai Jin and Yimin Zhang, color image segmentation based on mean shift and normalized cuts. IEEE Transactions on Systems,2007,37(5):1382-1389
[39]http://www.cnblogs.com/whyandinside/articles/1517987.html
[40]http://baike.baidu.com/view/17423.htm
[41]http://en.wikipedia.org/wiki/CIE_XYZ
[42]http://en.wikipedia.org/wiki/CIELUV_color_space
[43]http://blog.csdn.net/kit_147/archive/2010/04/14/5485470.aspx
[44]Takamura S, Kobayashi N. Practical extension to CIELUV color space to improve uniformity. Image Processing,2002:393-396
[45]http://baike.baidu.com/view/189685.htm
[46]http://baike.baidu.com/view/189736.htm
[47]Rafael C. Gonzalez, Richard E.Woods, Steven L.Eddins. Digital Image Processing Using Matlab.阮秋琦等译.电子工业出版社,2005：315-320,172-176
[48]Serra J. Image Analysis and Mathematical Morphology. London:Academic Press, 1982
[49]崔屹.图像处理与分析—数学形态学方法及应用.第1版.科学出版社,2000： 113-122
[50]Rafael C. Gonzalez, Richard E. Woods.阮秋琦,阮宇智等译.数字图像处理.第2版.电子工业出版社,2003：423-427,500-505
[51]Ting Chen, Zhenqiang Shen. An adaptive image segmentation method using region growing. In the Computer Engineering and Technology,2010:78-80
[52]L. Vincent, P. Soille. Watersheds in digital spaces:An efficient algorithm based on immersion simulations. IEEE Trans on Pattern Analysis and Machine Intelligence, 1991,13(6):583-598
[53]http://baike.baidu.com/view/146890.htm
[54]S. Di Zenzo. A note on the gradient of a multi-image. Computer Vision, Graphics, and Image Processing,1986,33:116-125
[55]K. Fukunaga and L. D.Hostetler. The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition. IEEE Trans. Information Theory,1975,21: 32-40
[56]Y. Cheng. Mean shift, Mode Seeking, and Clustering. IEEE Trans. Pattern Analysis and Machine Intelligence,1995:17(8):790-799
[57]王新华,毕笃彦.Mean Shift算法在图像分割中的应用研究.微计算机信息.2009,25(3)：290-292
[58]Comaniciu D and Meer P. Robust Analysis of Feature Spaces:Color Image Segmentation. IEEE Conf. computer Vision and Patten Recongnition 1997:750-755
[59]L Euler. Solutio problematis ad geometriam situs pertinentis,1736
[60]张先迪,李正良.图论及其应用.第1版.高等教育出版社,2005：1-3,5-6,15-20
[61]刘建龙.基于图论的图像分割算法研究.哈尔滨工业大学硕士学位论文.2006：15-16
[62]Z. Wu and R. Leahy. An Optimal Graph Theoretic Approach to Data Clustering: Theory and Its Application to Image Segmentation. IEEE PAMI.1993,11:1101-1113
[63]陈彦至.Ncut与医学图像分割.东华大学硕士学位论文.2009：16-19
[64]Sun Feng, He Jin-Peng. A normalized cuts based image segmentation method. In the 2009 2nd International Conference on Information and Computing Science,2009, 2:333-336
[65]席秋波.基于Ncut的图像分割算法研究.电子科技大学硕士学位论文.2010
[66]陶文兵,金海.一种新的基于图谱理论的图像阈值分割方法.计算机学报.2007,30(1)：110-118
[67]De Bock Johan, De Smet Patrick, Philips Wilfried. Image segmentation using watersheds and normalized cuts. Proceedings of SPIE-The International Society for Optical Engineering,2005,5675:164-173
[68]Timoee Cour, Florence Benezit and Jianbo Shi. Spectral Segmentation with Multiscale Graph decomposition. In the Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2005:1124-1131
[69]Song Haiyu, Li Xiongfei, Wang Pengjie, Chen Jingrun. Block-based normalized-cut algorithm for image segmentation. In the Lecture Notes in Electrical Engineering,2010, 56:272-278
[70]Eriksson Anders, Olsson Carl, Kahl Fredrik. Normalized Cuts Revisited:A Reformulation for Segmentation with Linear Grouping Constraints. Mathematical Imaging and Vision,2010:1-17
[2]章毓晋.图像工程.第2版.清华大学出版社,2005：74-75
[3]McInerney Tim, Terzopoulos Demetri. Deformable models in medical image analysis. Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis, 1996:171-180
[4]付曼,黄祥林,高芸.文档图像分割技术研究.中国传媒大学学报(自然科学版).2006,13(4)：62-67
[5]楚亚蕴,詹小四,孙兆才等.一种结合方向信息的指纹图像二值化算法.中国图像图形学报.2006,11(6)：854-860
[6]朱胜利.Mean Shift及相关算法在视频跟踪中的研究.浙江大学博士学位论文.2006：13-30,13-24
[7]Tian B, Shaikh M A, Azimi-Sadjadi M R, etal. A study of cloud classification with neural networks using spectral and textural features. IEEE Trans. Neural Networks, 1999,10(1):138-151
[8]Xu Jinming, Zhao Xiaobo. Determination of fluid inclusion lines using edge-detection technique. Chinese Journal of Rock Mechanics and Engineering,2007,26:1132-1137
[9]S Osher, J A Sethian. Fronts propagating with curvature dependent speed:Algorithms based on Hamilton-Jacobi formulation. Journal of Computational Physics,1988,79 (1): 12-49
[10]N. Otsu. A Threshold Selection Method from Gray-Level Histogram. IEEE Trans. Systems Man, and Cybernetics,1979,9:62-66
[11]T Pun. A new method for gray-level picture thresholding using the entropy of the histogram. In the Signal Processing,1980,2:223-237
[12]J Kittler and J Illingworth. Minimum error thresholding. Pattern Recognition,1986, 19(1):41-47
[13]Kittler J, Illingworth J, Foglein J, Paler K. An automatic thresholding algorithm and its performance. In the Pattern recognition.1984:287-289
[14]Otsu N. A threshold selection method from gray-level histograms. IEEE Trans. Systems Man and Cybernetics,1979,9(1):62-66
[15]Abutaleb A S. Automatic thresholding of gray-level pictures using two-dimensional entropies. Pattern Recognition,1989,47(1):22-32
[16]王双,焦李成等.阈值法标记提取的分水岭SAR图像分割方法.中国专利：CN101556693,2009：10-14
[17]蒋艳会,李峰.基于混沌粒子群算法的多阈值图像分割.计算机工程与应用.2010,46(10)：175-176
[18]J Canny. A computational approach to edge detection. In the Pattern Analysis and Machine Intelligence, IEEE Transactions,1986:679-698
[19]Nilsson N J. Principles of Artificial Intelligence. Palo Alto:Tioga Publishing Co,1980: 61-94
[20]Kirsch R. Computer determination of the constituent structure of biological images. In the Comuter Biomedical Research,1971,4:315-328
[21]Marr D, Hildreth E. Theory of edge detection. Proc. R. Soc. London,1980, B207: 187-217
[22]张闯,迟健男等.基于边缘检测与双边滤波的彩色图像去噪.电子学报.2010,38(8)：1776-1783
[23]Kass M, Witkin A, Terzopoulos D. Snakes:Active contour models. International Journal of Computer Vision,1988,1(4):321-331
[24]R Adams, L Bischof. Seeded region growing. IEEE Transactions on Pattern Analysis and Machine Intelligence,1994,16(6):641-647
[25]Frank Y, Shih, Shouxian Cheng. Automatic seeded region growing for color image segmentation. Image and Vision Computing,2005(23):877-886
[26]A Mehnert, P Jackway. An improved seeded region growing algorithm. Pattern Recognition Letters,1997(18):1065-1071
[27]Vincent L, Soille P. Watersheds in digital spaces:An efficient algorithm based on immersion simulations. IEEE Transactions on Pattern Analysis and Machine Intelligence,1991,13(6):583-598
[28]孙惠泉.图论及其应用.科学出版社,2004：1-2
[29]keinosuke Fukunaga, Larry D. Hostetler. The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition. IEEE Transactions on information theory,1975,21(1)：32-40
[30]钱芸,张英杰.水平集的图像分割方法综述.中国图像图形学报.2008,13(1)：7-13
[31]耿素云,屈婉玲,张立昂.离散数学.北京：清华大学出版社,1999：147-162
[32]赵继东,鲁珂,吴跃.一种基于谱图理论的Web图像搜索方法.计算机应用研究.2008,25(5)：1598-1600
[33]谭志明.基于图论的图像分割及其嵌入式应用研究.上海交通大学博士学位论文.2007：16-24
[34]C Zahn. Graph Theoretical Methods for Detecting and Describing Gestalt Clusters. IEEE Transactions on Computation,1971,20:68-86
[35]Z Wu and R Leahy. An Optimal Graph Theoretic Approach to Data Clustering:Theory and Its Application to Image Segmentation. IEEE PAMI,1993,11:1101-1113
[36]R Urquhart. Graph Theoretical Clustering Based on Limited Neighborhood sets. Pattern Recognition,1982,15(3):173-187
[37]J. Shi and J. Malik. Normalized Cuts and Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):888-905
[38]Wenbing Tao, Hai Jin and Yimin Zhang, color image segmentation based on mean shift and normalized cuts. IEEE Transactions on Systems,2007,37(5):1382-1389
[39]http://www.cnblogs.com/whyandinside/articles/1517987.html
[40]http://baike.baidu.com/view/17423.htm
[41]http://en.wikipedia.org/wiki/CIE_XYZ
[42]http://en.wikipedia.org/wiki/CIELUV_color_space
[43]http://blog.csdn.net/kit_147/archive/2010/04/14/5485470.aspx
[44]Takamura S, Kobayashi N. Practical extension to CIELUV color space to improve uniformity. Image Processing,2002:393-396
[45]http://baike.baidu.com/view/189685.htm
[46]http://baike.baidu.com/view/189736.htm
[47]Rafael C. Gonzalez, Richard E.Woods, Steven L.Eddins. Digital Image Processing Using Matlab.阮秋琦等译.电子工业出版社,2005：315-320,172-176
[48]Serra J. Image Analysis and Mathematical Morphology. London:Academic Press, 1982
[49]崔屹.图像处理与分析—数学形态学方法及应用.第1版.科学出版社,2000： 113-122
[50]Rafael C. Gonzalez, Richard E. Woods.阮秋琦,阮宇智等译.数字图像处理.第2版.电子工业出版社,2003：423-427,500-505
[51]Ting Chen, Zhenqiang Shen. An adaptive image segmentation method using region growing. In the Computer Engineering and Technology,2010:78-80
[52]L. Vincent, P. Soille. Watersheds in digital spaces:An efficient algorithm based on immersion simulations. IEEE Trans on Pattern Analysis and Machine Intelligence, 1991,13(6):583-598
[53]http://baike.baidu.com/view/146890.htm
[54]S. Di Zenzo. A note on the gradient of a multi-image. Computer Vision, Graphics, and Image Processing,1986,33:116-125
[55]K. Fukunaga and L. D.Hostetler. The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition. IEEE Trans. Information Theory,1975,21: 32-40
[56]Y. Cheng. Mean shift, Mode Seeking, and Clustering. IEEE Trans. Pattern Analysis and Machine Intelligence,1995:17(8):790-799
[57]王新华,毕笃彦.Mean Shift算法在图像分割中的应用研究.微计算机信息.2009,25(3)：290-292
[58]Comaniciu D and Meer P. Robust Analysis of Feature Spaces:Color Image Segmentation. IEEE Conf. computer Vision and Patten Recongnition 1997:750-755
[59]L Euler. Solutio problematis ad geometriam situs pertinentis,1736
[60]张先迪,李正良.图论及其应用.第1版.高等教育出版社,2005：1-3,5-6,15-20
[61]刘建龙.基于图论的图像分割算法研究.哈尔滨工业大学硕士学位论文.2006：15-16
[62]Z. Wu and R. Leahy. An Optimal Graph Theoretic Approach to Data Clustering: Theory and Its Application to Image Segmentation. IEEE PAMI.1993,11:1101-1113
[63]陈彦至.Ncut与医学图像分割.东华大学硕士学位论文.2009：16-19
[64]Sun Feng, He Jin-Peng. A normalized cuts based image segmentation method. In the 2009 2nd International Conference on Information and Computing Science,2009, 2:333-336
[65]席秋波.基于Ncut的图像分割算法研究.电子科技大学硕士学位论文.2010
[66]陶文兵,金海.一种新的基于图谱理论的图像阈值分割方法.计算机学报.2007,30(1)：110-118
[67]De Bock Johan, De Smet Patrick, Philips Wilfried. Image segmentation using watersheds and normalized cuts. Proceedings of SPIE-The International Society for Optical Engineering,2005,5675:164-173
[68]Timoee Cour, Florence Benezit and Jianbo Shi. Spectral Segmentation with Multiscale Graph decomposition. In the Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2005:1124-1131
[69]Song Haiyu, Li Xiongfei, Wang Pengjie, Chen Jingrun. Block-based normalized-cut algorithm for image segmentation. In the Lecture Notes in Electrical Engineering,2010, 56:272-278
[70]Eriksson Anders, Olsson Carl, Kahl Fredrik. Normalized Cuts Revisited:A Reformulation for Segmentation with Linear Grouping Constraints. Mathematical Imaging and Vision,2010:1-17