医学细胞图象分割与分析方法研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
本文讨论了数字图象处理技术在肺部细胞图象上的应用。首先进行细胞图象增强处理。细胞图象增强主要目的是改善细胞图象的质量,突出细胞图象的整体或局部特征,提高细胞的视觉效果和识别特征。细胞图象增强主要采用灰度变换、直方图修正、图象平滑、图象锐化等方法。灰度变换调整改变细胞图象的灰度,突出了感兴趣的区域。通过直方图修正得到均衡化或规定化等所需的不同的处理效果。采用有效的图象平滑方法对细胞图象进行平滑降噪处理,消除图象数字化和传输时所混入的噪声,提高了图象的视觉效果。利用图象锐化处理突出细胞的边缘信息,加强细胞的轮廓特征,以便于一些基本特征的提取,并比较各种图象锐化方法处理细胞图象时的结果。
     其次,对图象的分割方法包括细胞图象的分割方法进行了较详细的综述,指出现有方法的实现原理与特点。本文重点研究了数学形态学在二值细胞图象和灰度细胞图象分割处理中的应用。对二值图象数学形态学和灰度图象数学形态学的基本理论进行了较详细的介绍,利用二值图象数学形态学的边缘形态梯度检测出二值细胞图象的边缘,并对该方法进行了一定的研究;对于灰度图象,利用形态梯度、Top-Hat变换等进行细胞的边缘检测。对灰度图象数学形态学的边缘检测方法进行了改进,应用多结构多尺度结构算子的方法,达到既能检测细胞边缘又能有效降低噪声的目的。针对细胞之间存在的粘连和重叠现象,对基于数学形态学中的开闭运算和流域分割的方法进行了研究以便于进行细胞之间的分离。
     论文最后讨论了细胞图象的分析。利用局部邻域或图象标号等方法进行细胞计数。通过形态特征的提取和选择,利用最小距离法和形态法对肺部细胞进行了初步的识别分类。
This paper mainly discussed the digital image processing techniques in application of lung cell image. Firstly, to process the cell image so as to enhance the quality of the image, the aim of enhancing the cell image is to give prominence to the whole and portion feature of the cell image and improve the effect of vision and the feature of recognition. The method of enhancing the cell image include gray-scale transformation, histogram modification, image smoothing and image sharpening etc. gray-scale transformation adjust the gray-scale of the cell's image and give prominence to the interested area. By use of the method of histogram modification can get the results of equipoise and prescript. The method of image smoothing to process the cell image can eliminate the noises that caused by digital process and transmission and improve the effect of vision. The method of image sharpening give prominence to the edge information of the cell's image and strengthen the feature of contour so as to the extraction of ba
    sic feature, at last by comparison the results that were processed by some kinds of method of image sharpening can get which method is suit for this type of cell's image.
    Secondly, there is a survey about image segmentation method including the cell's image segmentation method, the survey indicate the principle and the characteristic of the segmentation methods. In this paper, the emphasis is the segmentation application of mathematical morphology to binary cell's image and gray-scale cell's image. There is a detailed presentation about the basic theory of binary image mathematical morphology and gray-scale image mathematical morphology. Using the edge modal detection method of binary image mathematical morphology can detect the edge of binary cell's image. In this paper, there is a research about the edge modal detection method of binary image mathematical morphology. As for gray-scale image, using the method of modal grads and Top-Hat transformation can detect the edge of cells. The edge detection method of gray-scale image mathematical morphology was be improved
    
    
    because of the disadvantage. Using the multi-structure and multi-scale operator to process the cell image can not only detect the edge of cell but also effectively reduce the noise. Aim at the conglutination and overlap in cells, researched the method of the opening and closing operation and valley segmentation based on mathematical morphology.
    Finally, discussed the cell image's analysis. By using the local neighborhood algorithm or the image label algorithm can take count of the number of cells. By extraction and selection of the cell shape feature, using the least distance method and shape method primarily identify and classify the lung cell.
引文
1.夏德深,傅德胜.现代图像处理技术与应用[M],南京:东南大学出版社,1997,42~43
    2.阮秋琦,数字图像处理学[M],北京:电子工业出版社,2001,200
    3.[美]R C.冈萨雷斯,P.温茨.数字图象处理[M],北京:科学出版社,1981,162
    4.王润生.图象理解[M],长沙:国防科技大学出版,1995
    5.章毓晋.图象分割评价技术分类和比较[J],中国图形图象学报,1996,1(2)
    6. V Chalana and Y Kim. A Methodology for Evaluation of Boundary Detection Algorithms on Medical Images[J], IEEE Trans, On Medical Imaging, 1997, 16(5): 642~652
    7. S D Yanowitz and A M Bruckstein.A New Method for Image Segmentation[J], CVGIP, 1989, 46
    8. K V Mardin and T J Hainsworth.A Spatial Thresholding Method for Image Segmentation[J], IEEE Trans.on PAMI-10, 1988
    9.李翊华,胡匡枯.细胞显微图像灰度梯度双阈值的快速分割[J],模式识别与人工智能,1995,8(4):357~362
    10.刘雷健,杨静宇,陆建峰.肺癌细胞识别彩色图像处理系统[J],计算机应用与软件,1996,(4):34~44
    11. Wu HS, Barba J, Gil J.A parametric fitting algorithm for segmentation of cellimages[J], IEEE Transactions on Biomedical Engineering, 1998, 3(45): 400
    12. Liedtke CE, Gahm T, Kappei F et al.Segmentation of microscopic cell scenes[J], Analyt Quant Cytol Histol, 1987, (9): 197~211
    13. Dinstein, A c Fong. Fast Distribution Between Homogeneous and Textured Region[J], Proceedings of 7th Conference on PR, 1984
    14. N Babaguchi, K Yamada, K Kisc and T.Tezuka. Connectionist Model Binarization[J], Processing of 10th ICPR, 1990, 51~56
    15. Theo Pavlidis & Yuh-tay Liow. Integrating region Growing and Edge Detection[J], IEEE Trans. on PAMI, 1990, 12(3)
    16. Wu HS, Barba J. An Efficient semi-automatic algorithm for cell contour extraction[J], J microsc, 1995, 179(Pt3): 270~276
    17.崔屹.图象处理与分析—数学形态学方法及应用[M],北京:科学出版社,2000
    18. Keng Wu. Live cell image segmentation[J], IEEE Trans On Biomedical
    
    Engineering, 1995, 42(1): 1~12
    19. Dwi Anoragaingrum. Cell segmentation with median filter and mathematical morphology operation[J], In Proc of the IEEE 10th International Conference on Imange Analysis and Processing9ICIAP, 1999, 1034~1046
    20.马东,曹培杰,潘凯丽,程敬之.分割重叠细胞核的方法及比较研究[J],北京生物医学工程,1999,18(3)
    21.陆建峰,杨静宇,唐振民等.重叠细胞图像分离算法的设计[J],计算机研究与发展[J],2000,37(2):228~232
    22.刘相滨,邹北骥,胡峰松.基于边界剥离的细胞图象分离算法[J],中国图象图形学报,2002,3:234~239
    23. Vincent L, Soille P.Watershed in digital spaces:an efficient algorithm based on immersion simulation[J], IEEE Trans. PAMI, 1991, 13(6): 583
    24.周孝宽.实用微机图象处理[M],北京:北京航空航天大学出版社,1994,102
    25.郑富盛.细胞形态立体计量学[M],北京:北京医科大学协和医科大学联合出版社,1990,3~40
    26.周洪堂.细胞自动计数中的一种图象标号算法[J],生物医学工程学杂志,1992,9(3):307~310
    27.杨晓敏,罗立民.白细胞自动分类中的纹理分析方法研究[J],中国生物医学工程学报,1993,12(3):124~128
    28.杨晓敏,罗立民,韦钰.血液白细胞计算机分类中的特征提取研究[J],应用科学学报,1994,12(2):132~136
    29.冒宇清,李宁,陆新泉等.肺癌早期诊断系统中形态学识别的研究与实现[J],计算机工程,1999,25(8)
    30. Chao Ch,Dhawan AP.Edge dectection Using a Hopfield neural network[J], Optical Engineering, 1994, 33:3739
    31.周维忠,赵海洋等.基于多尺度数学形态学的边缘检测[J],数据采集与处理,2000,15(3):316~319
    32.戴青云,余英林.一种基于小波与形态学的车牌图像的分割方法[J],中国图像图形学报,2000,5(5):411
    33.黄凤岗,杨国,宋克欧.柔性(soft)形态学在图像边缘检测中的应用[J],中国图像图形学报,2000,5(4):284
    34.李洪举.基于图象的建模和绘制研究[D],北京:中国科学院软件研究所,1998
    35. Lee J S J, Haralick R M.Morphology edge detection[J], IEEE Trans on Robotics Automat, 1987, (3): 140~156
    
    
    36.刘宁宁.医学图象处理与分析中交互技术的研究及应用[D],北京:中科院自动化研究所,1998
    37.吕维雪.医学图象处理[M],北京:高等教育出版社,1989,21~146
    38.容观澳.计算机图象处理[M],北京:清华大学出版社,2000,23
    39.郭峻,赵荣椿等.口腔颌面部涎腺肿瘤细胞计算机图象分析[J],中国生物医学工程,1997,16(1):94~95
    40.张晨红.数学形态学在图象处理中的若干应用研究[D],南京:南京航空航天大学,1999
    41 易东,曹佳等.微核计算机图像自动化检测研究现状及展望[J],国外医学生物医学工程分册,1997,20(1):46~50
    42.汤耀法,顾云娣,白咸勇.细胞形态参数的全自动检测[J],计算机应用与软件,1996,(1):17~21
    43.王凯,张建正.尿显微图像中细胞自动识别与分类的研究与分析系统的实现[J].仪器仪表学报,2000,21(6)648~650
    44.许军.基于计算机神经网络的细胞诊断系统研究[J],西安科技学院,2001,21(3):243~245
    45.周浩,李天牧,尉洪.基于数学形态学的血液细胞图像边缘提取[J],北京生物医学工程,2002,21(2):89~91
    46.周洪堂,俞可大.组织形态测量中的图象分析技术[J],生物医学工程杂志,1994,11(3):226~228
    47.郝葆青,罗教明,尹光福等.组织细胞图象简易分割方法的研究[J],四川大学学报,2000,32(3)
    48.宋今丹.医学细胞生物学[M],北京:人民卫生出版社,1997
    49.戴青云,余英林.数学形态学在图象处理中的应用进展[J],控制论与应用,2001,18(4)
    50.李晶晖,戚飞虎.基于数学形态学的强鲁棒性边缘检测方法研究[J],上海交通大学学报,1995,29(6):96~101

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700