基于内容的医学图像检索中特征提取技术研究
|
摘要
随着医学数字化影像设备在临床工作中日益广泛的应用，大量的医学图像数据随之产生。传统的医学档案管理系统采用简单的基于标注的图像数据库甚至完全人工的方法来管理图像数据，严重影响了图像在诊断过程中作用的发挥。如何有效地组织、管理和检索医学图像成为当前迫切需要解决的问题。基于内容的图像检索技术(CBIR，Content-Based Image Retrieval)是利用图像的视觉特征来进行检索，直接对图像内容进行分析并抽取特征，在临床、教学、科研以及医学图像归档和通信系统(PACS)中都有着重要的作用。
由于医学领域许多图像并不包含彩色或在有限条件下采用，因此，在视觉特征中，相比颜色或灰度特征而言，提取纹理和形状特征对医学图像检索显得较为重要。由此，本文重点研究了医学图像的形状特征提取和纹理特征提取。
对于纹理特征，本文利用空间域统计纹理特征描述符灰度共生矩阵实现了对医学图像纹理特征提取，并且将统计法和结构法有机的结合，利用灰度—基元共生矩阵对其进行算法改进。灰度—基元共生矩阵既考虑了图像像素的分布情况，又考虑了像素点周围邻域结构的分布情况。通过实验，验证了该方法鲁棒性好，对旋转不敏感，查询性能得到了改善。
对于形状特征，本文利用小波对图像边缘检测的有效性和相关边界矩对图像的区域和结构描述的统一性，提出了小波多尺度模极大值和相关矩的形状特征提取方法。首先对灰度图像进行小波模极大值变换，得到多尺度的边界图像，再利用6个相关边界矩提取每一个尺度的边界图像的特征，组成图像的特征向量，用欧几里得距离对归一化的特征矢量进行相似性度量，不同的尺度赋予不同的权值。通过实验，验证了该方法可以以一定的精度很好的描述目标物体的形状特征，抗干扰能力强，通用性较高，有效的解决了因图像的平移、尺度、旋转变换等带来的问题。同时针对由Mallat小波多尺度边缘检测方法得到的图像边缘较粗，细节过多的弊端，本文提出了模糊算法进行改进，剔除了冗余数据，得到精确的目标外形，将检索误差降低到最小程度。
最后，本文建立了一个基于内容的医学图像检索实验性原型系统，以一些医学图像为例对所研究的上述算法进行了验证。
With the rapid improvement of image-based clinical technique, large amounts of digital images are produced everyday. The management of medical image database and how to use those images in Clinical Diagnose Process (CDP) becomes an especially challenge in medical area. Content-Based Image Retrieval plays an important role in clinic, teaching, research and PACS, etc.
Many medical images don't contain color information or are adopted under the limited conditions. Texture and shape features are more important in the medical image retrieval, compared with the features of color or gray. In the context, we focus on the image's texture and shape features extraction.
For texture feature, we propose an algorithm of Gray-Primitive Co-occurrence Matrix that combines the space-statistical texture algorithm of Gray Level Co-occurrence Matrix with the structure distribution of surrounding pixels. Experiment results verify this method is robust, relatively feeble sensitivity to rotation and better query performance.
For texture feature, making use of the validity of wavelet edge detection and relative moment depiction of the region and structure, we propose a shape feature extraction based on wavelet and relative moments. First, it transforms the luminance images with wavelet edge detection to get multi-scale edge images, then calculates the six relative moments of every scale. Similarity is given by the Wikipedia distance between two images' normalized moment vectors. Experiment results verify this method is application independent and effective to solve the problem brought out by image translation, scaling, rotation. Meanwhile, the paper propose blurry algorithm to improve the defect of thick edge and overabundance details. It eliminates redundancy data, obtains accurate object shape and reduces the query error to the least degree.
At last, we develop a simple experimental medical image retrieval system based on example image and certificate the following algorithm.
由于医学领域许多图像并不包含彩色或在有限条件下采用，因此，在视觉特征中，相比颜色或灰度特征而言，提取纹理和形状特征对医学图像检索显得较为重要。由此，本文重点研究了医学图像的形状特征提取和纹理特征提取。
对于纹理特征，本文利用空间域统计纹理特征描述符灰度共生矩阵实现了对医学图像纹理特征提取，并且将统计法和结构法有机的结合，利用灰度—基元共生矩阵对其进行算法改进。灰度—基元共生矩阵既考虑了图像像素的分布情况，又考虑了像素点周围邻域结构的分布情况。通过实验，验证了该方法鲁棒性好，对旋转不敏感，查询性能得到了改善。
对于形状特征，本文利用小波对图像边缘检测的有效性和相关边界矩对图像的区域和结构描述的统一性，提出了小波多尺度模极大值和相关矩的形状特征提取方法。首先对灰度图像进行小波模极大值变换，得到多尺度的边界图像，再利用6个相关边界矩提取每一个尺度的边界图像的特征，组成图像的特征向量，用欧几里得距离对归一化的特征矢量进行相似性度量，不同的尺度赋予不同的权值。通过实验，验证了该方法可以以一定的精度很好的描述目标物体的形状特征，抗干扰能力强，通用性较高，有效的解决了因图像的平移、尺度、旋转变换等带来的问题。同时针对由Mallat小波多尺度边缘检测方法得到的图像边缘较粗，细节过多的弊端，本文提出了模糊算法进行改进，剔除了冗余数据，得到精确的目标外形，将检索误差降低到最小程度。
最后，本文建立了一个基于内容的医学图像检索实验性原型系统，以一些医学图像为例对所研究的上述算法进行了验证。
With the rapid improvement of image-based clinical technique, large amounts of digital images are produced everyday. The management of medical image database and how to use those images in Clinical Diagnose Process (CDP) becomes an especially challenge in medical area. Content-Based Image Retrieval plays an important role in clinic, teaching, research and PACS, etc.
Many medical images don't contain color information or are adopted under the limited conditions. Texture and shape features are more important in the medical image retrieval, compared with the features of color or gray. In the context, we focus on the image's texture and shape features extraction.
For texture feature, we propose an algorithm of Gray-Primitive Co-occurrence Matrix that combines the space-statistical texture algorithm of Gray Level Co-occurrence Matrix with the structure distribution of surrounding pixels. Experiment results verify this method is robust, relatively feeble sensitivity to rotation and better query performance.
For texture feature, making use of the validity of wavelet edge detection and relative moment depiction of the region and structure, we propose a shape feature extraction based on wavelet and relative moments. First, it transforms the luminance images with wavelet edge detection to get multi-scale edge images, then calculates the six relative moments of every scale. Similarity is given by the Wikipedia distance between two images' normalized moment vectors. Experiment results verify this method is application independent and effective to solve the problem brought out by image translation, scaling, rotation. Meanwhile, the paper propose blurry algorithm to improve the defect of thick edge and overabundance details. It eliminates redundancy data, obtains accurate object shape and reduces the query error to the least degree.
At last, we develop a simple experimental medical image retrieval system based on example image and certificate the following algorithm.
引文
[1] V N. Gudivada, V Raghavan.Content-based image retrieval systems. IEEE Computer Magazine, 28(9):18-22, September 1995.
[2] Y Rui, TS.Huang, S-F Chang. Image retrieval: Current techniques, promising directions, and open issues. Journal of Visual Communication and Image Representation, 10(1):39-62, March 1999.
[3] W M. Smeulders, M. Worring. Content-based image retrieval at the end of the early years. IEEE Trans. On Pattern Analysis and Machine Intelligence, 22(12):1349-1380, December 2000.
[4] 周杰，医学图像特征的自动获取与基于内容检索的方法研究，博士学位论文，第一军医大学，2004．
[5] Mojsilovis, J. Gomes. "Semantic based image categorization, browsing and retrieval in medical image databases",IEEE Int'l Conf. on Image Processing(ICIP'2000), Rochester, NY, USA, 2000.
[6] D. Keysers, J. Dahmen, H. Ney, B. B. Wein. "A statistical framework for model-based image retrieval in medical applications",Journal of Electronic Imaging, 12(1), 2003, pp.59-68.
[7] P. Schmidt-Saugeon, J. Guillod, J.-P. Thiran. "Towards a computer-aided diagnosis system for pigmented skin lesions", Computerized Medical Imaging and Graphics, vol. 27, 2003, pp.65-78.
[8] R. Pompl, W. Bunk, A. Horsch, W. Stolz. "MELDOQ: Ein System zur Unterst utzung der Fr uherkennung des malignen Melanoms durch digitale Bildverarbeitung", Proceedings of the Workshop Bildverarbeitung fur die Medizin,Munich, Germany, 2000.
[9] Sbober, C. Eccher, E. Blanzieri, P. Bauer. "A multiple classifier system for early melanoma diagnosis," Artificial Intelligence in Medicine, vol. 27, 2003, pp.29-44.
[10] K. Veropoulos, C. Campbell, G. Learnmonth. "Image processing and neural computing used in the diagnosis of tuberculosis", Proceedings of the Colloquium on Intelligent Methods in Healthcare and Medical Applications (IMHMA), York, England, 1998.
[11] M.-C. Jaulent, C. Le Bozec, Y. Cao. "A property concept frame representation for flexible image content retrieval in histopathology databases", Proceedings of the Annual Symposium of the American Society for Medical Informatics (AMIA), Los Angeles, CA, USA, 2000.
[12] M.R. Ogiela, R. Tadeusiewicz. "Semantic-oriented syntactic algorithms for content recognition and understanding of images in medical databases", Proceedings of the second Int'l Conf. on Multimedia and Exposition (ICME'2001), IEEE Computer Society, IEEE Computer Society, Tokyo, Japan, 2001, pp.621-624.
[13] G. P. Robinson, H.D.Targare, J. S. Duncan. "Medical image collection indexing: Shape-based retrieval using KD-trees," Computerized Medical Imaging and Graphics, 20 (4), 1996, pp.209-217.
[14] S. Constantinidis, M. C. Fairhurst, A. F. R. Rahman. "A new multi-expert decision combination algodthrn and its application to the detection of circumscribed masses in digital mammograms", Pattern Recognition, vol. 34, 2001, pp.1527-1537.
[15] P. Korn, N. Sidiropoulos, C. Faloutsos. "Fast and effective retrieval of medical tumor shapes", IEEE Trans.on Knowledge and Data Engineering, 10(6),1998, pp:889-904.
[16] S. Baeg, N. Kehtarnavaz. "Classification of breast mass abnormalities using denseness and architectural distorsion", Electronic Letters on Computer Vision and Image Analysis, 1 (1), 2002, pp.1-20.
[17] F. Schnorrenberg, C. S. Pattichis, C. N. Schizas, "Content-based retrieval of breast cancer biopsy slides", Technology and Health Care, vol.8, 2000, pp.291-297.
[18] W.J. Kuo, R.F. Chang. "Retrieval technique for the diagnosis solid breast tumors on sonogram", Ultrasound in Medicine and Biology, 28 (7), 2002, pp.903-909.
[19] C.R. Shyu, A. Kak, C. Brodley, L. S. Broderick. "Testing for human perceptual categories in a physician-in-the-loop CBIR system for medical imagery", IEEE Workshop on Content-based Access of Image and Video Libraries (CBAIVL'99), Fort Collins, Colorado, USA, 1999, pp.102-108.
[20] C.T. Liu, P.L. Tai, A. Y.J. Chen. "A content-based medical teaching file assistant for CT lung image retrieval," Proceedings of the IEEE Int'l Conf. On Electronics, Circuits, Systems (ICECS2000), Jouneih-Kaslik, Lebanon, 2000.
[21] C.T. Liu, PL. Tai, A. Y.J. Chen. "A content-based CT lung retrieval system for assisting differential diagnosis images collection," Proceedings of the second Int'l Conf. on Multimedia and Exposition (ICME'2001), IEEE Computer Society, Tokyo, Japan, 2001, pp.241-244.
[22] Schaefer Prokop, M. Prokop, D. Fleischmann, et al. "High-resolution CT of diffuse interstitial lung disease: key findings in common disorders," European Radiologist,vol. 11, 2001, pp.373-392.
[23] M.Hansel. " High-resolution CT of diffuse lung disease " ,Radiologic Clinics of North America, 39 (6), 2001, pp.1091-1113.
[24] H. Abe, H. MacMahon, R. Engelmann. "Computer-aided diagnosis in chest radiography: Results of large-scale observer tests at the 1996-2001 RSNA scientific assemblies", RadioGraphics, vol.23 (1), 2003, 255-265.
[25] 章毓晋，基于内容的视觉信息检索，科学出版社，2003，429-439．
[26] 胡小锋，赵辉，Visual C++／MATLAB图像处理与识别实用案例精选，人民邮 电出版社，2004，121-124．
[27] ZHENG Feng, FENG Ju-fu, SHI Qing-yun. Texture features based on local Fourier transform[A].In: Proceedings of International Conference on Image Processing(ICIP)[C], Thessaloniki, Greece, October 7-10,2001.
[28] 郭德军，宋蜇存，基于灰度共生矩阵的纹理图像分类研究，林业机械与木工设备，Vol．33，No．7，2005：45-47．
[29] 赵松年等，子波变换与子波分析，北京：电子工业出版社，1996，247-251．
[30] Mallat S, Hwang W L.1992a.Singularity detection and processing with wavelets[J]. IEEE IT,38(2):617-643.
[31] 王亮申，欧宗瑛．图像纹理分析的灰度-基元共生矩阵法，计算机工程，Vol．30，No 23：78-80．
[32] http://amazon.ece.utexas.edu/～qasim/cires.htm
[33] 吴国庆，基于小波变换和尺度共生矩阵的超声图像分割，电子工程师，Vol．31，No．12，Dec．2005：104-107．
[34] Mallat S, Zhang S.1992b.Characterization of signals from multi-scale edges[J]. IEEE PAMI,14(7): 617-643.
[35] 李丽宏等，相对边界矩在模式识别中的应用，微计算机信息，2005，21(7-3)：42-43．
[36] 傅蓉，许宏丽，基于小波多尺度分析的彩色图像检索方法，中国图象图形学报，Vol．9，No．11，Nov．2004：243-251．
[37] 杨晓强，魏生民，汪焰恩，一种基于知识模型的CT图像分割方法，计算机应用研究，No．8，2005：227-229．
[38] 杨加，吴祈耀，田捷等，几种图像分割算法在CT图像分割上的实现和比较，北京理工大学学报，Vol．20，No．6，Dec．2000：189-194．
[39] 丛培盛，张洪江等，用遗传算法C-均值聚类分割医学彩色图像，同济大学学报，Vol．31，No．7，Jul．2003：26-29．
[40] 姜晓彤，罗立民等，一种肺部肿瘤CT图像序列的自动分割方法，中国图象图形学报，Vol．8(A)，No．9，Sep．2003：138-143．
[41] 徐森，曹力，金属图像的自适应分割方法研究，南京航空航天大学学报，Vol．37，No．5，Oct．2005：132-138．
[42] 肖虎程，一种使用多特征融合的图像检索方法，硕士学位论文，襄樊大学，2005．
[43] 邵虹，崔文成等，低级特征和语义特征相结合的医学图像检索方法，中国图象图形学报，Vol．9，No．2，Feb．2004：114-120．
[44] 李殿赟，医学影像数据库的图像检索技术应用研究，重庆大学硕士学位论文，2004：76-79．
[2] Y Rui, TS.Huang, S-F Chang. Image retrieval: Current techniques, promising directions, and open issues. Journal of Visual Communication and Image Representation, 10(1):39-62, March 1999.
[3] W M. Smeulders, M. Worring. Content-based image retrieval at the end of the early years. IEEE Trans. On Pattern Analysis and Machine Intelligence, 22(12):1349-1380, December 2000.
[4] 周杰，医学图像特征的自动获取与基于内容检索的方法研究，博士学位论文，第一军医大学，2004．
[5] Mojsilovis, J. Gomes. "Semantic based image categorization, browsing and retrieval in medical image databases",IEEE Int'l Conf. on Image Processing(ICIP'2000), Rochester, NY, USA, 2000.
[6] D. Keysers, J. Dahmen, H. Ney, B. B. Wein. "A statistical framework for model-based image retrieval in medical applications",Journal of Electronic Imaging, 12(1), 2003, pp.59-68.
[7] P. Schmidt-Saugeon, J. Guillod, J.-P. Thiran. "Towards a computer-aided diagnosis system for pigmented skin lesions", Computerized Medical Imaging and Graphics, vol. 27, 2003, pp.65-78.
[8] R. Pompl, W. Bunk, A. Horsch, W. Stolz. "MELDOQ: Ein System zur Unterst utzung der Fr uherkennung des malignen Melanoms durch digitale Bildverarbeitung", Proceedings of the Workshop Bildverarbeitung fur die Medizin,Munich, Germany, 2000.
[9] Sbober, C. Eccher, E. Blanzieri, P. Bauer. "A multiple classifier system for early melanoma diagnosis," Artificial Intelligence in Medicine, vol. 27, 2003, pp.29-44.
[10] K. Veropoulos, C. Campbell, G. Learnmonth. "Image processing and neural computing used in the diagnosis of tuberculosis", Proceedings of the Colloquium on Intelligent Methods in Healthcare and Medical Applications (IMHMA), York, England, 1998.
[11] M.-C. Jaulent, C. Le Bozec, Y. Cao. "A property concept frame representation for flexible image content retrieval in histopathology databases", Proceedings of the Annual Symposium of the American Society for Medical Informatics (AMIA), Los Angeles, CA, USA, 2000.
[12] M.R. Ogiela, R. Tadeusiewicz. "Semantic-oriented syntactic algorithms for content recognition and understanding of images in medical databases", Proceedings of the second Int'l Conf. on Multimedia and Exposition (ICME'2001), IEEE Computer Society, IEEE Computer Society, Tokyo, Japan, 2001, pp.621-624.
[13] G. P. Robinson, H.D.Targare, J. S. Duncan. "Medical image collection indexing: Shape-based retrieval using KD-trees," Computerized Medical Imaging and Graphics, 20 (4), 1996, pp.209-217.
[14] S. Constantinidis, M. C. Fairhurst, A. F. R. Rahman. "A new multi-expert decision combination algodthrn and its application to the detection of circumscribed masses in digital mammograms", Pattern Recognition, vol. 34, 2001, pp.1527-1537.
[15] P. Korn, N. Sidiropoulos, C. Faloutsos. "Fast and effective retrieval of medical tumor shapes", IEEE Trans.on Knowledge and Data Engineering, 10(6),1998, pp:889-904.
[16] S. Baeg, N. Kehtarnavaz. "Classification of breast mass abnormalities using denseness and architectural distorsion", Electronic Letters on Computer Vision and Image Analysis, 1 (1), 2002, pp.1-20.
[17] F. Schnorrenberg, C. S. Pattichis, C. N. Schizas, "Content-based retrieval of breast cancer biopsy slides", Technology and Health Care, vol.8, 2000, pp.291-297.
[18] W.J. Kuo, R.F. Chang. "Retrieval technique for the diagnosis solid breast tumors on sonogram", Ultrasound in Medicine and Biology, 28 (7), 2002, pp.903-909.
[19] C.R. Shyu, A. Kak, C. Brodley, L. S. Broderick. "Testing for human perceptual categories in a physician-in-the-loop CBIR system for medical imagery", IEEE Workshop on Content-based Access of Image and Video Libraries (CBAIVL'99), Fort Collins, Colorado, USA, 1999, pp.102-108.
[20] C.T. Liu, P.L. Tai, A. Y.J. Chen. "A content-based medical teaching file assistant for CT lung image retrieval," Proceedings of the IEEE Int'l Conf. On Electronics, Circuits, Systems (ICECS2000), Jouneih-Kaslik, Lebanon, 2000.
[21] C.T. Liu, PL. Tai, A. Y.J. Chen. "A content-based CT lung retrieval system for assisting differential diagnosis images collection," Proceedings of the second Int'l Conf. on Multimedia and Exposition (ICME'2001), IEEE Computer Society, Tokyo, Japan, 2001, pp.241-244.
[22] Schaefer Prokop, M. Prokop, D. Fleischmann, et al. "High-resolution CT of diffuse interstitial lung disease: key findings in common disorders," European Radiologist,vol. 11, 2001, pp.373-392.
[23] M.Hansel. " High-resolution CT of diffuse lung disease " ,Radiologic Clinics of North America, 39 (6), 2001, pp.1091-1113.
[24] H. Abe, H. MacMahon, R. Engelmann. "Computer-aided diagnosis in chest radiography: Results of large-scale observer tests at the 1996-2001 RSNA scientific assemblies", RadioGraphics, vol.23 (1), 2003, 255-265.
[25] 章毓晋，基于内容的视觉信息检索，科学出版社，2003，429-439．
[26] 胡小锋，赵辉，Visual C++／MATLAB图像处理与识别实用案例精选，人民邮 电出版社，2004，121-124．
[27] ZHENG Feng, FENG Ju-fu, SHI Qing-yun. Texture features based on local Fourier transform[A].In: Proceedings of International Conference on Image Processing(ICIP)[C], Thessaloniki, Greece, October 7-10,2001.
[28] 郭德军，宋蜇存，基于灰度共生矩阵的纹理图像分类研究，林业机械与木工设备，Vol．33，No．7，2005：45-47．
[29] 赵松年等，子波变换与子波分析，北京：电子工业出版社，1996，247-251．
[30] Mallat S, Hwang W L.1992a.Singularity detection and processing with wavelets[J]. IEEE IT,38(2):617-643.
[31] 王亮申，欧宗瑛．图像纹理分析的灰度-基元共生矩阵法，计算机工程，Vol．30，No 23：78-80．
[32] http://amazon.ece.utexas.edu/～qasim/cires.htm
[33] 吴国庆，基于小波变换和尺度共生矩阵的超声图像分割，电子工程师，Vol．31，No．12，Dec．2005：104-107．
[34] Mallat S, Zhang S.1992b.Characterization of signals from multi-scale edges[J]. IEEE PAMI,14(7): 617-643.
[35] 李丽宏等，相对边界矩在模式识别中的应用，微计算机信息，2005，21(7-3)：42-43．
[36] 傅蓉，许宏丽，基于小波多尺度分析的彩色图像检索方法，中国图象图形学报，Vol．9，No．11，Nov．2004：243-251．
[37] 杨晓强，魏生民，汪焰恩，一种基于知识模型的CT图像分割方法，计算机应用研究，No．8，2005：227-229．
[38] 杨加，吴祈耀，田捷等，几种图像分割算法在CT图像分割上的实现和比较，北京理工大学学报，Vol．20，No．6，Dec．2000：189-194．
[39] 丛培盛，张洪江等，用遗传算法C-均值聚类分割医学彩色图像，同济大学学报，Vol．31，No．7，Jul．2003：26-29．
[40] 姜晓彤，罗立民等，一种肺部肿瘤CT图像序列的自动分割方法，中国图象图形学报，Vol．8(A)，No．9，Sep．2003：138-143．
[41] 徐森，曹力，金属图像的自适应分割方法研究，南京航空航天大学学报，Vol．37，No．5，Oct．2005：132-138．
[42] 肖虎程，一种使用多特征融合的图像检索方法，硕士学位论文，襄樊大学，2005．
[43] 邵虹，崔文成等，低级特征和语义特征相结合的医学图像检索方法，中国图象图形学报，Vol．9，No．2，Feb．2004：114-120．
[44] 李殿赟，医学影像数据库的图像检索技术应用研究，重庆大学硕士学位论文，2004：76-79．