医学图像融合算法研究与应用
摘要
医学图像融合属于生物医学工程等多种学科的研究前沿和应用热点。本文以CT和MRI图像为例,对医学图像融合的有关理论和算法进行了研究。
首先分析了几种目前常用的医学图像的成像原理和特点,对CT和MRI图像的成像机理进行了深入分析,提出CT和MRI图像融合的意义。第二章简要介绍了医学图像融合的基本概念和原理以及图像融合的评价标准,分析CT和MRI图像融合模型。第三章介绍了医学图像处理的多分辩分析的理论,用Curvelet数据结构分析了CT和MRI图像,证明图像的各个尺度和各个方向上的融合规则的选取都很重要。第四章分析了现有的两类医学图像融合的融合规则的优缺点和互补性,提出了基于Curvelet和PCNN的融合算法,并且利用MATLAB编程,针对实际图像进行了融合实验。实验结果证明,该算法能很好地融合图像的边缘,并且获得较好的视觉效果。
综上,本文提出的方法能较好的融合CT和MRI图像,得到骨骼和软组织均清晰的图像,故具有一定的实用性,且对于同类课题具有借鉴价值。
The rising technology of medical image fusion is becoming hot point for biomedical engineering and many other disciplines. This paper gives a research on medical images fusion and related theory and algorithm, basing on the images of CT and MRI.
Firstly, several medical imaging theory and their respective characteristic are reviewed, especially CT and MRI, whose fusion significance is proposed. Then the basic concept and theory of medical images fusion are introduced, and the evaluation of images fusion is introduced too, and fusion model of CT and MRI images are analyzed. In the third chapter, Multi-resolution analysis are introduced, and it gives an analysis on CT and MRI images with the data structure of Curvelet, which proves that the choices of fusion rules are of the same importance in every direction and every scale. In the forth chapter, tow fusion rules and their excellences and disadvantages are discussed at the beginning, and then it finally comes into the images fusion algorithm based on Curvelet and PCNN(Pulse Coupled Neural Network). What's more, the experimentation about this rule is acted on the platform of MATLAB, which shows that this method is capable of fusing the edge information of CT and MRI images very well, and gives a better appearance than traditional rules.
Above all, the method proposed in this paper are capable of fusion the CT and MRI images more edge information and visual effect than the traditional fusion algorithms do. Thus, for developers and researchers in this field this dissertation should be of some use.
首先分析了几种目前常用的医学图像的成像原理和特点,对CT和MRI图像的成像机理进行了深入分析,提出CT和MRI图像融合的意义。第二章简要介绍了医学图像融合的基本概念和原理以及图像融合的评价标准,分析CT和MRI图像融合模型。第三章介绍了医学图像处理的多分辩分析的理论,用Curvelet数据结构分析了CT和MRI图像,证明图像的各个尺度和各个方向上的融合规则的选取都很重要。第四章分析了现有的两类医学图像融合的融合规则的优缺点和互补性,提出了基于Curvelet和PCNN的融合算法,并且利用MATLAB编程,针对实际图像进行了融合实验。实验结果证明,该算法能很好地融合图像的边缘,并且获得较好的视觉效果。
综上,本文提出的方法能较好的融合CT和MRI图像,得到骨骼和软组织均清晰的图像,故具有一定的实用性,且对于同类课题具有借鉴价值。
The rising technology of medical image fusion is becoming hot point for biomedical engineering and many other disciplines. This paper gives a research on medical images fusion and related theory and algorithm, basing on the images of CT and MRI.
Firstly, several medical imaging theory and their respective characteristic are reviewed, especially CT and MRI, whose fusion significance is proposed. Then the basic concept and theory of medical images fusion are introduced, and the evaluation of images fusion is introduced too, and fusion model of CT and MRI images are analyzed. In the third chapter, Multi-resolution analysis are introduced, and it gives an analysis on CT and MRI images with the data structure of Curvelet, which proves that the choices of fusion rules are of the same importance in every direction and every scale. In the forth chapter, tow fusion rules and their excellences and disadvantages are discussed at the beginning, and then it finally comes into the images fusion algorithm based on Curvelet and PCNN(Pulse Coupled Neural Network). What's more, the experimentation about this rule is acted on the platform of MATLAB, which shows that this method is capable of fusing the edge information of CT and MRI images very well, and gives a better appearance than traditional rules.
Above all, the method proposed in this paper are capable of fusion the CT and MRI images more edge information and visual effect than the traditional fusion algorithms do. Thus, for developers and researchers in this field this dissertation should be of some use.
引文
[1]田捷,包尚联,周明全.医学影像处理与分析.北京:电子工业出版社,2003
[2]赵家真.核素断层成像及其应用简介.医疗设备信息.2002,3期.22-23
[3]王静云,李绍林.医学影像图像融合技术的新进展.第四军医大学学报.2004,Vol.25,No.20.1918-1920
[4]吴世法.近代成象技术与图像处理.北京:国防工业出版社,1997,139-145,200-207
[5]阮春,李月卿.医学图像融合技术及其应用研究概况.医学影像学杂志.2001年,Vol.11.No.6.408-410
[6]王毅,倪国强,李勇亮.高性能双波段图像融合处理机.高科技通讯,2002,05:103-106
[7] Burt P J,Adelson E. The Laplacian Pyramid as a Compact Image Code. IEEE Transactions On Communications. 1983,31(4) .523-540
[8] Akerman. Pyramid Techniques for Multisensor Fusion. SPIE Sensor Fusion. 1992,1828.124-131
[9] Toet A. Image Fusion by a Ratio of Low-Pass Pyramid. Pattern Recognition Letters. 1989,9(4) .245-253
[10] Yocky David. Image Merging and Data Fusion by Means of the Discrete 2-D Wavelet Transform. Journal of Optical Society of America. 1995,12(9) .1834-1841
[11] Li H,Manjunath B S,Mitra S K. Multisensor Image Fusion Using the Wavelet Transform. Graphical Models and Image Processing. 1995,57(3) .235-245
[12] Xiaoyu Jiang,Liwei Zhou,Zhiyun Gao. Multispectral Image Fusion Using Wavelet Transform. Acta Electronica Sinica. 1997,25(8) .105-108
[13] Yu Lifeng,Zu Donglin,Wang Weidong,etal. Multi-Modality Medical Image Fusion Based on Wavelet Analysis and Quality Evaluation. Journal of Systems Engineering and Electronics. 2001,12(1) .42-48
[14] Wenzhong Shi,ChangQing Zhu,Yan Tian,etal. Wavelet-based Image Fusion and Quality Assessment. International Journal of Applied Earth Observation and Geoinformation. 2005,7(2) .241-251
[15]何友,王国宏.多传感器信息融合及应用.北京:电子工业出版社,2000
[16]康耀红.数据融合理论与应用.陕西:西安电子科技大学出版社,1997
[17] Lucien WAld. Some Terms of Reference in Data Fusion. IEEE Transacations on Geoscience and Remote Sensing. 1999,37(3) .1190-1193
[18] S.S Shen. Summary of types of data fusion methods utilized in workshop papers,Multisource Data Integration in Remote Sensing, Proceedings of Workshop,Marylandm, NASA Conference Publication 3099,1990
[19]霍宏涛,林小竹,何薇.数字图像处理.北京:北京理工大学出版社,2002
[20]朱志刚,石定机等译.数字图像处理.北京:电子工业出版社,1998
[21]那彦,焦李成.基于多分辨分析理论的图像融合方法.西安:西安电子科技大学出版社,2007
[22]张煜,刘哲星,李树祥等.医学图像信息融合技术的发展.国外医学生物医学工程分册.2000,Vol.23,No.4.103-105
[23] Levin DN,Hu Xiao-ping,Tan KK,etal. The brain:intergrated three-dimensional display of MR and PRT images. Radiology. 1989,172.783-789
[24] Vincken KL, Koster ASE, Viergever M,etal. Probabilistic multiscale image segmentation. IEEE Trans Patem Analysis and Machine Intelligence.1997,19.109-120
[25]郭利明.图像处理及图像融合.西安:西北工业大学,2006
[26] L.H.David. An introduction to multi-sensor fusion. Proceeding of the IEEE. New York. 1997,85(1).6-23
[27]陈煜.图像融合技术及其应用研究.南京:南京航空航天大学,2004
[28] Fechner. T. Godlewski. G. Optimal fusion of TV and infrared images using artificial neural networks. in Proc. of SPIE on Applications and Science of Artificial Neural Networks. Washington, USA: SPIE Press. 1995.919-925
[29] Li H, Manjunath B S, Mitra S K. Multisensor Image Fusion Using the Wavelet Transform. Graphical Models and Image Processing. 1995,27(3). 235-244
[30] Yocky D A. Image Merging and Data Fusion by Means of the Discrete Two Dimensional Wavelet Transform. J Opt Soc Am A.1995,12.( 9).1834-1841
[31]覃征,鲍复民,李爱国.数字图像融合.西安:西安交通大学出版社,2006
[32] Laure J.Chipman,Timothy M.and Lewis N.Graham.Wavelets and Image Fusion,’Proc ICIP’95. 1995.248-251
[33] Mallat S G.A .Theory for Multidimension Signal Decomposition:The Wavelet Representation.IEEE Trans .Pattern Analysis and Machine Intell.,1989,11:674-693
[34] Mallat S G. Multifrequency Channel Decomposition of Image and Wavelet Models. IEEE Trans.Acoust.,Speech,Signal Processing,1989,37.2091-2110
[35] Gemma Piella. New quality measure for image fusion. Polytechnical University ofCatalonia.
[36]刘贵喜,杨万海.基于小波分解的图像融合方法及性能评价.自动化学报.2002,11(28).927-934
[37]焦李成,谭山.图像的多尺度几何分析:回顾和展望.电子学报.2003,Vol.31,12A.43-50
[38] F.G.Meyer,R.Coifman. Directional image compression with brushlets. Proceedings of the IEEE-SP International Symposium on Time-Frequency and Time-Scale Analysis. 1996,189-1926
[39] E. Le. Pennec, S. Mallat. Image compression with geometrical wavelets . In Proc. of ICIP'2000, Vancouver, Canada. 2000,9. 661-664
[40] D. L. Donoho, M. Vetterli, R. A. DeVore. Data compression and harmonic analysis [J]. IEEE Trans.1998, Information Theory-44(6).2435–2476
[41] Candes E.J.,Donoho D.L.Curvelets. Tech.Report, Department of Statistics, Stanford University,1999
[42] E. Le. Pennec and S. Mallat. Bandelet representation for image compression. In Proc. Int. Conf. Image Processing. 2001,Oct
[43] M.N. Do and M. Vetterli. The contourlet Transform : An Efficient Directional Multiresolution Image Representation. IEEE Trans. on Image Processing. 2005, 14(12).2091-2106
[44] Fran?ois G. Meyer and Ronald R. Coifman, Brushlets: A Tool for Directional Image Analysis and Image Compression, Applied and Computational Harmonic Analysis, 1997,4.147-187
[45] Fran?ois G. Meyer and Ronald R. Coifman, Directional image compression with brushlets, Proceedings of the IEEE-SP International Symposium on, June 1996,18-21
[46] E. J. Candes. Ridgelet: theory and applications. Ph. D. dissertation. Stanford Univ,1998
[47] E. J. Candes, D. L. Donoho Ridgelets: A key to Higher-dimmensional Intermittency Technical report, Stanford Univ ,1999
[48] E. J. Candes Monoscale ridgelets for the representation of images with edges. Technical report, Stanford Univ,1999
[49]侯彪,刘芳,焦李成.基于脊波变换的直线特征检测.中国科学.2003, Vol. 33(1),Jan.65-73
[50]倪林,Y.Miao.一种更适合图像处理的多尺度变换.计算机工程与应用.2004,28.21-26
[51] Jianwei Liu, Qian yin, Ping Guo. A NEW STRATEGY TO IMPROVE IMAGE FUSION EFFECT .Proceeding of the Fifth International Conference on Machine Learining and Cybermetics, Dalian. 2006,8. 13-16
[52]丛琳.图像的方向纹理分析及其应用研究.西安电子科技大学硕士学位论文.2004,1
[53]顾晓东,余道衡.PCNN的原理及其应用.电路与系统学报.2001,6(3).45-50
[54]马义德,李廉,王亚馥等.脉冲耦合神经网络原理及其应用.北京:科学出版社,2006
[55] Broussard R P, Rogers S K, Oxley M E,etal. Physiologically motivated image fusion for object detection using a pulse coupled neural network. IEEE Transactions on Neural Networks.1999,10(3).554-563
[56]王科俊,张燕,唐墨等.PCNN在图像处理中的应用.哈尔滨工程大学学报.2006,Vol.27 (Suppl).183-184
[57] G.Kuntimad,H.S.Ranganath. Perfect Image Segmentation using Pulse Coupled Neuran Networks. IEEE Trans.on Neural Networks. 1999,5, Vol.10,No.3
[58] John.L,Johnson,Mary Lou Padgett. PCNN Models and Applications. IEEE Trans on Neural Netwroks. 1999,5, Vol.10,No.3
[59] Thomas Lindblad, Jason M.Kinser. Inherent feature of wavelets and Pulse Coupled Neural Networks. IEEE Trans,Neural Networks. 1999,5, Vol.10,No.3.607-614
[60] A.R.Gillespie,A.B.Kahle,R.E.Walker. Color enhancement of highly correlated images-II channel ratio and‘chromaticity’transformation techniques. Remote Sens. Environ. 1982,22.343-356
[2]赵家真.核素断层成像及其应用简介.医疗设备信息.2002,3期.22-23
[3]王静云,李绍林.医学影像图像融合技术的新进展.第四军医大学学报.2004,Vol.25,No.20.1918-1920
[4]吴世法.近代成象技术与图像处理.北京:国防工业出版社,1997,139-145,200-207
[5]阮春,李月卿.医学图像融合技术及其应用研究概况.医学影像学杂志.2001年,Vol.11.No.6.408-410
[6]王毅,倪国强,李勇亮.高性能双波段图像融合处理机.高科技通讯,2002,05:103-106
[7] Burt P J,Adelson E. The Laplacian Pyramid as a Compact Image Code. IEEE Transactions On Communications. 1983,31(4) .523-540
[8] Akerman. Pyramid Techniques for Multisensor Fusion. SPIE Sensor Fusion. 1992,1828.124-131
[9] Toet A. Image Fusion by a Ratio of Low-Pass Pyramid. Pattern Recognition Letters. 1989,9(4) .245-253
[10] Yocky David. Image Merging and Data Fusion by Means of the Discrete 2-D Wavelet Transform. Journal of Optical Society of America. 1995,12(9) .1834-1841
[11] Li H,Manjunath B S,Mitra S K. Multisensor Image Fusion Using the Wavelet Transform. Graphical Models and Image Processing. 1995,57(3) .235-245
[12] Xiaoyu Jiang,Liwei Zhou,Zhiyun Gao. Multispectral Image Fusion Using Wavelet Transform. Acta Electronica Sinica. 1997,25(8) .105-108
[13] Yu Lifeng,Zu Donglin,Wang Weidong,etal. Multi-Modality Medical Image Fusion Based on Wavelet Analysis and Quality Evaluation. Journal of Systems Engineering and Electronics. 2001,12(1) .42-48
[14] Wenzhong Shi,ChangQing Zhu,Yan Tian,etal. Wavelet-based Image Fusion and Quality Assessment. International Journal of Applied Earth Observation and Geoinformation. 2005,7(2) .241-251
[15]何友,王国宏.多传感器信息融合及应用.北京:电子工业出版社,2000
[16]康耀红.数据融合理论与应用.陕西:西安电子科技大学出版社,1997
[17] Lucien WAld. Some Terms of Reference in Data Fusion. IEEE Transacations on Geoscience and Remote Sensing. 1999,37(3) .1190-1193
[18] S.S Shen. Summary of types of data fusion methods utilized in workshop papers,Multisource Data Integration in Remote Sensing, Proceedings of Workshop,Marylandm, NASA Conference Publication 3099,1990
[19]霍宏涛,林小竹,何薇.数字图像处理.北京:北京理工大学出版社,2002
[20]朱志刚,石定机等译.数字图像处理.北京:电子工业出版社,1998
[21]那彦,焦李成.基于多分辨分析理论的图像融合方法.西安:西安电子科技大学出版社,2007
[22]张煜,刘哲星,李树祥等.医学图像信息融合技术的发展.国外医学生物医学工程分册.2000,Vol.23,No.4.103-105
[23] Levin DN,Hu Xiao-ping,Tan KK,etal. The brain:intergrated three-dimensional display of MR and PRT images. Radiology. 1989,172.783-789
[24] Vincken KL, Koster ASE, Viergever M,etal. Probabilistic multiscale image segmentation. IEEE Trans Patem Analysis and Machine Intelligence.1997,19.109-120
[25]郭利明.图像处理及图像融合.西安:西北工业大学,2006
[26] L.H.David. An introduction to multi-sensor fusion. Proceeding of the IEEE. New York. 1997,85(1).6-23
[27]陈煜.图像融合技术及其应用研究.南京:南京航空航天大学,2004
[28] Fechner. T. Godlewski. G. Optimal fusion of TV and infrared images using artificial neural networks. in Proc. of SPIE on Applications and Science of Artificial Neural Networks. Washington, USA: SPIE Press. 1995.919-925
[29] Li H, Manjunath B S, Mitra S K. Multisensor Image Fusion Using the Wavelet Transform. Graphical Models and Image Processing. 1995,27(3). 235-244
[30] Yocky D A. Image Merging and Data Fusion by Means of the Discrete Two Dimensional Wavelet Transform. J Opt Soc Am A.1995,12.( 9).1834-1841
[31]覃征,鲍复民,李爱国.数字图像融合.西安:西安交通大学出版社,2006
[32] Laure J.Chipman,Timothy M.and Lewis N.Graham.Wavelets and Image Fusion,’Proc ICIP’95. 1995.248-251
[33] Mallat S G.A .Theory for Multidimension Signal Decomposition:The Wavelet Representation.IEEE Trans .Pattern Analysis and Machine Intell.,1989,11:674-693
[34] Mallat S G. Multifrequency Channel Decomposition of Image and Wavelet Models. IEEE Trans.Acoust.,Speech,Signal Processing,1989,37.2091-2110
[35] Gemma Piella. New quality measure for image fusion. Polytechnical University ofCatalonia.
[36]刘贵喜,杨万海.基于小波分解的图像融合方法及性能评价.自动化学报.2002,11(28).927-934
[37]焦李成,谭山.图像的多尺度几何分析:回顾和展望.电子学报.2003,Vol.31,12A.43-50
[38] F.G.Meyer,R.Coifman. Directional image compression with brushlets. Proceedings of the IEEE-SP International Symposium on Time-Frequency and Time-Scale Analysis. 1996,189-1926
[39] E. Le. Pennec, S. Mallat. Image compression with geometrical wavelets . In Proc. of ICIP'2000, Vancouver, Canada. 2000,9. 661-664
[40] D. L. Donoho, M. Vetterli, R. A. DeVore. Data compression and harmonic analysis [J]. IEEE Trans.1998, Information Theory-44(6).2435–2476
[41] Candes E.J.,Donoho D.L.Curvelets. Tech.Report, Department of Statistics, Stanford University,1999
[42] E. Le. Pennec and S. Mallat. Bandelet representation for image compression. In Proc. Int. Conf. Image Processing. 2001,Oct
[43] M.N. Do and M. Vetterli. The contourlet Transform : An Efficient Directional Multiresolution Image Representation. IEEE Trans. on Image Processing. 2005, 14(12).2091-2106
[44] Fran?ois G. Meyer and Ronald R. Coifman, Brushlets: A Tool for Directional Image Analysis and Image Compression, Applied and Computational Harmonic Analysis, 1997,4.147-187
[45] Fran?ois G. Meyer and Ronald R. Coifman, Directional image compression with brushlets, Proceedings of the IEEE-SP International Symposium on, June 1996,18-21
[46] E. J. Candes. Ridgelet: theory and applications. Ph. D. dissertation. Stanford Univ,1998
[47] E. J. Candes, D. L. Donoho Ridgelets: A key to Higher-dimmensional Intermittency Technical report, Stanford Univ ,1999
[48] E. J. Candes Monoscale ridgelets for the representation of images with edges. Technical report, Stanford Univ,1999
[49]侯彪,刘芳,焦李成.基于脊波变换的直线特征检测.中国科学.2003, Vol. 33(1),Jan.65-73
[50]倪林,Y.Miao.一种更适合图像处理的多尺度变换.计算机工程与应用.2004,28.21-26
[51] Jianwei Liu, Qian yin, Ping Guo. A NEW STRATEGY TO IMPROVE IMAGE FUSION EFFECT .Proceeding of the Fifth International Conference on Machine Learining and Cybermetics, Dalian. 2006,8. 13-16
[52]丛琳.图像的方向纹理分析及其应用研究.西安电子科技大学硕士学位论文.2004,1
[53]顾晓东,余道衡.PCNN的原理及其应用.电路与系统学报.2001,6(3).45-50
[54]马义德,李廉,王亚馥等.脉冲耦合神经网络原理及其应用.北京:科学出版社,2006
[55] Broussard R P, Rogers S K, Oxley M E,etal. Physiologically motivated image fusion for object detection using a pulse coupled neural network. IEEE Transactions on Neural Networks.1999,10(3).554-563
[56]王科俊,张燕,唐墨等.PCNN在图像处理中的应用.哈尔滨工程大学学报.2006,Vol.27 (Suppl).183-184
[57] G.Kuntimad,H.S.Ranganath. Perfect Image Segmentation using Pulse Coupled Neuran Networks. IEEE Trans.on Neural Networks. 1999,5, Vol.10,No.3
[58] John.L,Johnson,Mary Lou Padgett. PCNN Models and Applications. IEEE Trans on Neural Netwroks. 1999,5, Vol.10,No.3
[59] Thomas Lindblad, Jason M.Kinser. Inherent feature of wavelets and Pulse Coupled Neural Networks. IEEE Trans,Neural Networks. 1999,5, Vol.10,No.3.607-614
[60] A.R.Gillespie,A.B.Kahle,R.E.Walker. Color enhancement of highly correlated images-II channel ratio and‘chromaticity’transformation techniques. Remote Sens. Environ. 1982,22.343-356