模糊C-均值算法在彩色牛乳体细胞图像分割中的应用
|
摘要
牛乳体细胞图像分割算法是基于图像处理的体细胞自动分析、计数系统的关键部分,对提高牛乳质量检测和奶牛乳腺炎诊断的效率和准确性具有重要意义。为此,主要针对彩色牛乳体细胞图像的分割进行了分析和研究。
首先采用方差分析方法和对比度方法进行彩色空间的选择。经实验表明,在RGB、HSI和Lab中,RGB空间较适合于牛乳体细胞图像的分割。用方差方法分析了RGB的三个颜色分量R、G、B,得出分量R和G区分能力较好,因此本文采用R和G分量信息进行图像分割与融合。
由于模糊C-均值算法在3维彩色空间对图像分割时运算量大、速度慢、不利于实时应用,提出了将模糊C-均值算法及融合技术方法应用到1维空间R与G空间实施图像分割,再将分割结果利用信息融合技术进行图像融合,从而完成图像的分割。经与直接在3维空间上进行的分割及与在2维RB空间和1维空间分割并融合的方法进行比较,表明这种方法分割正确率高、时间耗费少,在实际应用中可采用R与G 2个颜色值分割并进行融合的方法对牛乳体细胞图像进行分割。
在C++ Builder下实现了牛乳体细胞图像的分割。
The accurate segmentation of milk somatic cells in microscope images may contribute to development of a successful system which automaticly analyze,detect and count cells in microscope images. It is important to improve the milk quality detection and diagnose bovine mastitis. Therefore, the thesis mainly concentrates on researching the segmentation the segmentation of milk somatic cell color images.
Variance analysis and contrast method are applied to select color space. The results showed that RGB is more suitable for milk somatic cell image in RGB, HSI and Lab. Three color components R, G and B are analysed by variance analysis, The results showed that R , G color values is used in image segmentation.
Because the image segmentation methods of 3-D color spaces consume large amount of computation and have slow speed. They do not suit to real-time applications. For those reasons, an approach for color image segmentation, called fuzzy c-means algorithm and images fusion are presented. For RGB color images, the band subsets are chosen as RB, R and G. segmentation results are then combined to obtain a final result using image fusion technology. The results are compared to segmentation in 3-D and segmentation in 2-D and 1D fusion, The results showed that this method correct rate is high and less time-consuming. In practical application, milk somatic cell color images were segmented by the method using the R and G color values is used. Milk somatic cell image is realized under C++ Builder.
首先采用方差分析方法和对比度方法进行彩色空间的选择。经实验表明,在RGB、HSI和Lab中,RGB空间较适合于牛乳体细胞图像的分割。用方差方法分析了RGB的三个颜色分量R、G、B,得出分量R和G区分能力较好,因此本文采用R和G分量信息进行图像分割与融合。
由于模糊C-均值算法在3维彩色空间对图像分割时运算量大、速度慢、不利于实时应用,提出了将模糊C-均值算法及融合技术方法应用到1维空间R与G空间实施图像分割,再将分割结果利用信息融合技术进行图像融合,从而完成图像的分割。经与直接在3维空间上进行的分割及与在2维RB空间和1维空间分割并融合的方法进行比较,表明这种方法分割正确率高、时间耗费少,在实际应用中可采用R与G 2个颜色值分割并进行融合的方法对牛乳体细胞图像进行分割。
在C++ Builder下实现了牛乳体细胞图像的分割。
The accurate segmentation of milk somatic cells in microscope images may contribute to development of a successful system which automaticly analyze,detect and count cells in microscope images. It is important to improve the milk quality detection and diagnose bovine mastitis. Therefore, the thesis mainly concentrates on researching the segmentation the segmentation of milk somatic cell color images.
Variance analysis and contrast method are applied to select color space. The results showed that RGB is more suitable for milk somatic cell image in RGB, HSI and Lab. Three color components R, G and B are analysed by variance analysis, The results showed that R , G color values is used in image segmentation.
Because the image segmentation methods of 3-D color spaces consume large amount of computation and have slow speed. They do not suit to real-time applications. For those reasons, an approach for color image segmentation, called fuzzy c-means algorithm and images fusion are presented. For RGB color images, the band subsets are chosen as RB, R and G. segmentation results are then combined to obtain a final result using image fusion technology. The results are compared to segmentation in 3-D and segmentation in 2-D and 1D fusion, The results showed that this method correct rate is high and less time-consuming. In practical application, milk somatic cell color images were segmented by the method using the R and G color values is used. Milk somatic cell image is realized under C++ Builder.
引文
1 薛河儒. 牛乳体细胞图像的分割方法的研究[D]. 内蒙古农业大学博士学位论文, 2007
2 刘俊丽. 牛乳体细胞图像处理及计数方法的研究[D].内蒙古农业大学硕士学位论文,2007
3 刘海霞. 牛乳体细胞数的检测方法[J].中国乳品工业,2004,Vol.32:31～34
4 储明星, 师守坊. 奶牛体形和体细胞数(乳房炎)之间关系的研究进展[J]. 畜牧与兽医,1995, Vol. 27(1): 35～37
5 侯振杰. 动物骨髓细胞图像分割方法的研究[D]. 内蒙古农业大学博士学位论文, 2004
6 冈萨雷斯. 数字图像处理(第二版). 电子工业出版社. 470～489
7 马英辉,韩炎. 彩色图像分割方法综述[J]. 科技情报开发与经济, 2006, vol.16(14)
8 张晓芸. 彩色图像分割算法的研究与实现[D]. 重庆大学硕士学位论文, 2005
9 KurugolluF, SankurB, HarmanciAE. Color image segmentation using histogram multithresholding and fusion[J]. Image and Vision Computing, 2001, 19(13): 915～
928
10 张博. 基于边缘检测的细胞图像分割方法研究与实现[D]. 武汉理工大学硕士学位论文,2006
11 G.B.Coleman, H.C.Andrews. Image segmentation by clustering[J].Proc IEEE, 1979,
5(67): 773-785
12 T.L. Huntsberger. Iterative fuzzy image segmentation[J].Patern Recognition, 1985, 2(18): 131-137
13 J.C.Bezdek. Patern recognition with fuzzy objective function algorithms[J].New York: Plenum Press,1981
14 Schachter, BJ, Davis, LS, Rosenfeld, A. Some experiments in image segmentation by clustering of local feature values[J].PR(11),1979,1:19-28
15 Silverman, JF, Cooper, DB. Bayesian clustering for unsupervised estimation of surface and texture models[J].TPAM(10),1988,482-495
16 Vinod, VV, Chaudhury, S. A Connectionist Approach for Clustering with Applications in Image-Analysis[J].SMC(24),1994,3:365-383
17 JM Jolion, P.Meer, S.Bataouche. Robust clustering with application in computer vision[J].IEEE Trans.Pattern and Machine Intell,1991,13(8):791-802
18 R.Hofmann, A.Jain.Segmentation and classification of range images[J]. IEEE Trans. On Patt.Anal.Machine Intell,1987,9(5):608-620
19 P.Flynn, A.Jain bonsal. 3D Object recognition using constrained search[J].PAMI,1991,13(10):1066-1075
20 裴继红,谢维信.直方图模糊约束 FCM 聚类自适应多阴值图像分割算法[J].电子学报,1999,27(10):38-42
21 林开林,林林林,林林辉.快速模糊 C-均值聚类彩色图像分割方法[J].中国图像图形学报,2004,9(2):159-163
22 林怡,李龙澍,李学俊. 基于粗糙熵和 K 均值聚类算法的图像分割[J],华东理工大学学报(自然科学版),2007,4(33):255-258
23 Rahimi.S, Zargham.M, Thakre.A. A parallel fuzzy C-mean algorithm for image segmentation[J], Fuzzy Information,2004(1):234-237
24 Janne Koljonen, Jarmo T. Genetic algorithm for optimizing fuzzy image pattern matching[R], the 12th Finnish Artificial Intelligence Society FAIS,2006
25 Tolisas YA, Panas S M. Image segmentation by a fuzzy clustering algorithm using adaptive spatially constrained membership function[J].IEEE Trans Syst, Man, Cybernet Part,1998,28(3):359-369
26 Liew AWC,Leung SH,Lau WH.Fuzzy image clustering incorporating spatial continuity[J].IEEE Proc Vis Image Signal Proc 2000,147(2):185-192
27 Pham DL. Fuzzy clustering with spatial constraints[c].in:Proceedings of the IEEE International Conference on Image Processing, New York,USA,August,2002
28 Krishnapuran, Keller J M. A possibilistic c-means algorithm[J],IEEE Trans Fuzzy Syst,1993,2:100-112
29 Wang X ,Wang Y, Wang L. Improving fuzzy c-means clustering based on feature-weight learning[J].Pattern Recognit Lett 2004:25:1123-1132
30 Dao-Qiang Zhang, Song-Can Chen. A novel kernelized fuzzy C-means algorithm with application in medical image segmentation[J].Artifical Intelligence in Medicine 2004 32:37-50
31 Surya Pemmaraju,Multiresolution walelet decompositon and neuro-fuzzy clustering for segmentation of radiographic images.Eighth IEEE Symposium on Computer-Based Medical Systems,1995:142-149
32 Ke Chen, De-liang Wang. A dynamically coupled neural oscillator net work for image segmentation. Neural Networks,2002:15:423-439
33 沈明霞,李秀智,姬长英.水果品质检测中的模糊阈值分割方法[J],农业机械学报,2003,5(34):113-115
34 杨勇,郑崇勋,林盘,潘晨,顾建文. 基于改进的模糊 C 均值聚类图像分割新算法[J],光电子激光,2005,9(16):1118-1122
35 M.M.Trivedi, J.C.Bezdek. Low-level segmentation of aerial image with fuzzy clustering[J].IEEE SMC,1996,16(4):589-598
36 R.Porter, N. Canagarajah. A robust automatic clustering scheme for imagesegmentation using wavelets[J].IEEE Image Processing 1996,5(4):662-665
37 S.W.Chen, C.F.Chen, M.S.Chen. Neural-fuzzy classification for segmentation of remotely sensed images[J].IEEE Signal Processing,1997,45(11):2639-2654
38 B.B.Chaudhuri, N.Sarkar.Texture segmentation using fractal dimension[J].IEEE PAMI, 2002 17(1):72-77
39 刘珍. 呼和浩特地区临床型奶牛乳腺炎病原分离鉴定及其对小白鼠致病性的研究[D].内蒙古农业大学硕士学位论文, 2002: 20
40 薛河儒,麻硕士,裴喜春.一种基于数学形态学及融合技术的彩色图像分割方法[J].中国图像图形学报,2006, 11(12):1764～1767
41 J. Gauch, Chi-Wan Hsia. A Comparison of three color image segmentation algorithm in four color spaces. Visual Communications and Image Processing92.1992.1168-1181
42 钟志光,卢君. 数字图像处理实例与解析[M]. 北京:清华大学出版社, 2003: 63～64
43 叶家鸣. 彩色城市交通地图道路信息的识别与提取[D]. 中国科学技术大学硕士学位, 2003: 11～12
44 Liu J Q,Yang Y H.Multiresolution color image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1994,16(7):689-700
45 史勇红. 基于模糊聚类的彩色图像分割算法的研究[D].昆明理工大学的硕士学位论文,2002
46 王彬. 基于模糊算法的彩色血液图像分割方法的研究[D]. 四川大学硕士学位论文, 2006
47 于桂玲. 基于计算机辅助结核病检测及图像处理系统[D].北京化工大学硕士学位论文, 2006: 18～23
48 I T Young, J J Gerbrands, J van Vliet. Image processing fundamentals[EB/OL]. http://www.ph.tn.tudelft.nl/Courses/FTP,2004
49 郭戈. 图像分割算法研究及其在癌细胞诊断中的应用[D]. 解放林信息工程大学硕士学位论文, 2005: 26～27
50 王林. 基于三基色原理的彩色图像拼接算法改进[D]. 内蒙古工业大学硕士学位论文, 2005: 42～44
51 Kosko B. Neural network and fuzzy system: A dynamical systems approach to machine intelligence. Prentice-Hall Englewood Cliffs,NJ,1996
52 Zadeh L A.Fuzzy sets. Information and Control, 8(1965):338-353
53 闵珊华,贺仲雄.懂一点模糊数学.北京:中国青年出版社,1985
54 LA.Zadeh,Fuzzy sets,Information and Control,1965,8:338353
55 林薇.图像处理中的模糊技术,现代电子技术,2001:28-30.
56 孙光灵.基于模糊阈值的图像分割方法研究[D].合肥工业大学硕士学位论文,2005,32-40
57 刘晓龙.基于模糊聚类图像分割方法研究[D].合肥工业大学硕士学位论文,2006,19-33
58 鲍正益.模糊聚类算法及其有效性研究,厦门大学硕士学位论文.2006,41-45
59 丁震,胡钟山等.FCM 算法用于灰度图像分割的研究.电子学报,1997,25(5):39-43
60 Bezdek JC. Pattern recognition with objective function algorithms. New York: Plenum Press,1981
61 Gao XB, Li J, Xie WX. Parameter optimization in FCM clustering algorithms. International Conference on Signal Processing,2000.Vol.3:1457-1461
62 高 新 波 , 裴 继 红 , 谢 维 信 . 模 糊 c- 均 值 聚 类 算 法 中 加 权 指 数 m 的 研 究 . 电报,2000,28(4):80-83
63 宫改云,高新波,伍忠东.FCM 聚类算法中模糊加权指数 m 的优选方法.模糊系统与数学,2005,19(1):143-148
64 李云松.改进的模糊 C-均值聚类对噪声图像的分割[D].兰州理工大学硕士学位论文
65 杨润玲.基于 FCM 类型算法的自动图像分割方法研究[D].西安电子科技大学硕士学位论文,2006
66 Kehrli M E, Shuster D E. Factors affecting milk somatic cells and their role in health of the bovine mammary gland [J].J Dairy Science,1994,Vol.77:619～627
67 Cheng H D, Jiang X H, Sun Y, et al. Co lo r image segmentation: advances and prospects[J]. Pattern Recognition,2001, 34 (12) : 2259～ 2281
68 范玮琦,张田文. 血细胞图像的计数方法研究[J]. 计算机应用与软件, 1998, Vol. 17(5): 61～64
69 赵宁. 血液细胞彩色图像的微机自动计数研究[J]. 林州医学院学报, 1998, Vol. 18(4): 390
2 刘俊丽. 牛乳体细胞图像处理及计数方法的研究[D].内蒙古农业大学硕士学位论文,2007
3 刘海霞. 牛乳体细胞数的检测方法[J].中国乳品工业,2004,Vol.32:31～34
4 储明星, 师守坊. 奶牛体形和体细胞数(乳房炎)之间关系的研究进展[J]. 畜牧与兽医,1995, Vol. 27(1): 35～37
5 侯振杰. 动物骨髓细胞图像分割方法的研究[D]. 内蒙古农业大学博士学位论文, 2004
6 冈萨雷斯. 数字图像处理(第二版). 电子工业出版社. 470～489
7 马英辉,韩炎. 彩色图像分割方法综述[J]. 科技情报开发与经济, 2006, vol.16(14)
8 张晓芸. 彩色图像分割算法的研究与实现[D]. 重庆大学硕士学位论文, 2005
9 KurugolluF, SankurB, HarmanciAE. Color image segmentation using histogram multithresholding and fusion[J]. Image and Vision Computing, 2001, 19(13): 915～
928
10 张博. 基于边缘检测的细胞图像分割方法研究与实现[D]. 武汉理工大学硕士学位论文,2006
11 G.B.Coleman, H.C.Andrews. Image segmentation by clustering[J].Proc IEEE, 1979,
5(67): 773-785
12 T.L. Huntsberger. Iterative fuzzy image segmentation[J].Patern Recognition, 1985, 2(18): 131-137
13 J.C.Bezdek. Patern recognition with fuzzy objective function algorithms[J].New York: Plenum Press,1981
14 Schachter, BJ, Davis, LS, Rosenfeld, A. Some experiments in image segmentation by clustering of local feature values[J].PR(11),1979,1:19-28
15 Silverman, JF, Cooper, DB. Bayesian clustering for unsupervised estimation of surface and texture models[J].TPAM(10),1988,482-495
16 Vinod, VV, Chaudhury, S. A Connectionist Approach for Clustering with Applications in Image-Analysis[J].SMC(24),1994,3:365-383
17 JM Jolion, P.Meer, S.Bataouche. Robust clustering with application in computer vision[J].IEEE Trans.Pattern and Machine Intell,1991,13(8):791-802
18 R.Hofmann, A.Jain.Segmentation and classification of range images[J]. IEEE Trans. On Patt.Anal.Machine Intell,1987,9(5):608-620
19 P.Flynn, A.Jain bonsal. 3D Object recognition using constrained search[J].PAMI,1991,13(10):1066-1075
20 裴继红,谢维信.直方图模糊约束 FCM 聚类自适应多阴值图像分割算法[J].电子学报,1999,27(10):38-42
21 林开林,林林林,林林辉.快速模糊 C-均值聚类彩色图像分割方法[J].中国图像图形学报,2004,9(2):159-163
22 林怡,李龙澍,李学俊. 基于粗糙熵和 K 均值聚类算法的图像分割[J],华东理工大学学报(自然科学版),2007,4(33):255-258
23 Rahimi.S, Zargham.M, Thakre.A. A parallel fuzzy C-mean algorithm for image segmentation[J], Fuzzy Information,2004(1):234-237
24 Janne Koljonen, Jarmo T. Genetic algorithm for optimizing fuzzy image pattern matching[R], the 12th Finnish Artificial Intelligence Society FAIS,2006
25 Tolisas YA, Panas S M. Image segmentation by a fuzzy clustering algorithm using adaptive spatially constrained membership function[J].IEEE Trans Syst, Man, Cybernet Part,1998,28(3):359-369
26 Liew AWC,Leung SH,Lau WH.Fuzzy image clustering incorporating spatial continuity[J].IEEE Proc Vis Image Signal Proc 2000,147(2):185-192
27 Pham DL. Fuzzy clustering with spatial constraints[c].in:Proceedings of the IEEE International Conference on Image Processing, New York,USA,August,2002
28 Krishnapuran, Keller J M. A possibilistic c-means algorithm[J],IEEE Trans Fuzzy Syst,1993,2:100-112
29 Wang X ,Wang Y, Wang L. Improving fuzzy c-means clustering based on feature-weight learning[J].Pattern Recognit Lett 2004:25:1123-1132
30 Dao-Qiang Zhang, Song-Can Chen. A novel kernelized fuzzy C-means algorithm with application in medical image segmentation[J].Artifical Intelligence in Medicine 2004 32:37-50
31 Surya Pemmaraju,Multiresolution walelet decompositon and neuro-fuzzy clustering for segmentation of radiographic images.Eighth IEEE Symposium on Computer-Based Medical Systems,1995:142-149
32 Ke Chen, De-liang Wang. A dynamically coupled neural oscillator net work for image segmentation. Neural Networks,2002:15:423-439
33 沈明霞,李秀智,姬长英.水果品质检测中的模糊阈值分割方法[J],农业机械学报,2003,5(34):113-115
34 杨勇,郑崇勋,林盘,潘晨,顾建文. 基于改进的模糊 C 均值聚类图像分割新算法[J],光电子激光,2005,9(16):1118-1122
35 M.M.Trivedi, J.C.Bezdek. Low-level segmentation of aerial image with fuzzy clustering[J].IEEE SMC,1996,16(4):589-598
36 R.Porter, N. Canagarajah. A robust automatic clustering scheme for imagesegmentation using wavelets[J].IEEE Image Processing 1996,5(4):662-665
37 S.W.Chen, C.F.Chen, M.S.Chen. Neural-fuzzy classification for segmentation of remotely sensed images[J].IEEE Signal Processing,1997,45(11):2639-2654
38 B.B.Chaudhuri, N.Sarkar.Texture segmentation using fractal dimension[J].IEEE PAMI, 2002 17(1):72-77
39 刘珍. 呼和浩特地区临床型奶牛乳腺炎病原分离鉴定及其对小白鼠致病性的研究[D].内蒙古农业大学硕士学位论文, 2002: 20
40 薛河儒,麻硕士,裴喜春.一种基于数学形态学及融合技术的彩色图像分割方法[J].中国图像图形学报,2006, 11(12):1764～1767
41 J. Gauch, Chi-Wan Hsia. A Comparison of three color image segmentation algorithm in four color spaces. Visual Communications and Image Processing92.1992.1168-1181
42 钟志光,卢君. 数字图像处理实例与解析[M]. 北京:清华大学出版社, 2003: 63～64
43 叶家鸣. 彩色城市交通地图道路信息的识别与提取[D]. 中国科学技术大学硕士学位, 2003: 11～12
44 Liu J Q,Yang Y H.Multiresolution color image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1994,16(7):689-700
45 史勇红. 基于模糊聚类的彩色图像分割算法的研究[D].昆明理工大学的硕士学位论文,2002
46 王彬. 基于模糊算法的彩色血液图像分割方法的研究[D]. 四川大学硕士学位论文, 2006
47 于桂玲. 基于计算机辅助结核病检测及图像处理系统[D].北京化工大学硕士学位论文, 2006: 18～23
48 I T Young, J J Gerbrands, J van Vliet. Image processing fundamentals[EB/OL]. http://www.ph.tn.tudelft.nl/Courses/FTP,2004
49 郭戈. 图像分割算法研究及其在癌细胞诊断中的应用[D]. 解放林信息工程大学硕士学位论文, 2005: 26～27
50 王林. 基于三基色原理的彩色图像拼接算法改进[D]. 内蒙古工业大学硕士学位论文, 2005: 42～44
51 Kosko B. Neural network and fuzzy system: A dynamical systems approach to machine intelligence. Prentice-Hall Englewood Cliffs,NJ,1996
52 Zadeh L A.Fuzzy sets. Information and Control, 8(1965):338-353
53 闵珊华,贺仲雄.懂一点模糊数学.北京:中国青年出版社,1985
54 LA.Zadeh,Fuzzy sets,Information and Control,1965,8:338353
55 林薇.图像处理中的模糊技术,现代电子技术,2001:28-30.
56 孙光灵.基于模糊阈值的图像分割方法研究[D].合肥工业大学硕士学位论文,2005,32-40
57 刘晓龙.基于模糊聚类图像分割方法研究[D].合肥工业大学硕士学位论文,2006,19-33
58 鲍正益.模糊聚类算法及其有效性研究,厦门大学硕士学位论文.2006,41-45
59 丁震,胡钟山等.FCM 算法用于灰度图像分割的研究.电子学报,1997,25(5):39-43
60 Bezdek JC. Pattern recognition with objective function algorithms. New York: Plenum Press,1981
61 Gao XB, Li J, Xie WX. Parameter optimization in FCM clustering algorithms. International Conference on Signal Processing,2000.Vol.3:1457-1461
62 高 新 波 , 裴 继 红 , 谢 维 信 . 模 糊 c- 均 值 聚 类 算 法 中 加 权 指 数 m 的 研 究 . 电报,2000,28(4):80-83
63 宫改云,高新波,伍忠东.FCM 聚类算法中模糊加权指数 m 的优选方法.模糊系统与数学,2005,19(1):143-148
64 李云松.改进的模糊 C-均值聚类对噪声图像的分割[D].兰州理工大学硕士学位论文
65 杨润玲.基于 FCM 类型算法的自动图像分割方法研究[D].西安电子科技大学硕士学位论文,2006
66 Kehrli M E, Shuster D E. Factors affecting milk somatic cells and their role in health of the bovine mammary gland [J].J Dairy Science,1994,Vol.77:619～627
67 Cheng H D, Jiang X H, Sun Y, et al. Co lo r image segmentation: advances and prospects[J]. Pattern Recognition,2001, 34 (12) : 2259～ 2281
68 范玮琦,张田文. 血细胞图像的计数方法研究[J]. 计算机应用与软件, 1998, Vol. 17(5): 61～64
69 赵宁. 血液细胞彩色图像的微机自动计数研究[J]. 林州医学院学报, 1998, Vol. 18(4): 390