Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/4870
標題: 多頻譜腦部磁振造影像分析使用權重式高斯核心之支援向量機
Multispectral Brain Magnetic Resonance Image Analysis Using Weighted Radial Basis Function Kernels for Support Vector Machine
作者: 林伯鴻
Lin, Bo-Hong
關鍵字: Weighted-RBF Kernels;權重式高斯核心;SVM;支援向量機
出版社: 通訊工程研究所
引用: [1] G.. A. Wright, “Magnetic resonance imaging,” IEEE Signal Processing Magazine, 14(1): 56-66, 1997 [2] C.-I Chang, Hyperspectral Imaging: Techniques for Spectral Detection and Classification, Kluwer Academic Publishers, 2003. [3] J. W. Boardman, F. A. Kruse and R. O. Green, “Mapping target signatures via partial unmixing of AVIRIS data,” Summaries of JPL Airborne Earth Science Workshop, Pasadena, CA, 1995 [4] L. Clarke, R. Vethuizen, M. Camacho, J. Heine, M. Vaidyanathan, L.O. Hall, R.W. Thatcher and M.L. Silbiger, “MRI segmentation: methods and applications,” Magnetic Resonance Imaging, 13(3): 343-368, 1995. [5] J. C. Bezdek, L. O. Hall and L. Clarke, “Review of MRI segmentation techniques using pattern recognition,” Medical Physics, 20: 1033-1048, 1993. [6] A. Hyvarinen, J. Karhunen and E. Oja, Independent Component Analysis, New York: John Wiley & Sons, 2001. [7] T. Nakai, S. Muraki, E. Bagarinao, Y. Miki, Y. Takehara, K. Matsuo, C. Kato, H. Sakahara, H. Isoda, “Application of independent component analysis to magnetic resonance imaging for enhancing the contrast of gray and white matter,” NeuroImage, 21: 251-260, 2004. [8] T. Kim, I. Lee, T.W. Lee, “Independent vector analysis: definition and algorithms.” Proc. of 40th Asilomar Conference on Signals, Systems and Computers, 2006. ACSSC ''06; 1393-1396. [9] T. Nakai, S. Muraki, E. Bagarinao, Y. Miki, Y. Takehara, K. Matsuo, C. Kato, H. Sakahara, H. Isoda, “Application of independent component analysis to magnetic resonance imaging for enhancing the contrast of gray and white matter,” NeuroImage, 21: 251-260, 2004. [10] Y. C. Ouyang, H. M. Chen, J. W. Chai, C. C. C. Chen, C. C. Chen, S. K. Poon, C. W. Yang, and S. K. Lee,“Independent Component Analysis for Magnetic Resonance Image Analysis.” EURASIP Journal on Applied Signal Processing, 780656:1-14, 2008. [11] Y. C. Ouyang, H. M. Chen, J. W. Chai, C. C. C. Chen, S. K. Poon, C. W. Yang, S. K. Lee, and C. I . Chang, “Band Expansion-Based Over-Complete Independent Component Analysis for Multispectral Processing of Magnetic Resonance Images.” IEEE Transactions on Biomedical Engineering, 55(6): 1666-1677, 2008J.W. Boardman, F.A. Kruse and R.O. Green, “Mapping target signatures via partial unmixing of AVIRIS data,” Summaries of JPL Airborne Earth Science Workshop, Pasadena, CA, 1995. [12] Y. J Chiou, H. M Chen, J. W Chai, C. C. C Chen, Y. C. Ouyang, W. C Su, C. W Yang, S. K Lee, C. I Chang, “Brian Tissue Classification Using Independent Vector Analysis(IVA) For Magnetic Resonance Image” IEEE International Conference on Bioinformatics and Bioengineering, 52: 324-329, 2009 [13] C.-I Chang and Q. Du, "Estimation of number of spectrally distinct signal sources in hyperspectral imagery," IEEE Trans. on Geoscience and Remote Sensing, vol. 42, no. 3, pp. 608-619, March 2008. [14] A. Ifarraguerri and C.-I Chang, "Hyperspectral image segmentation with convex cones," IEEE Trans. on Geoscience and Remote Sensing, vol. 37, no 2, pp. 756-770, March 1999. [15] F. Chaudhry, C. Wu, W. Liu, C.-I Chang and A. Plaza, Pixel Purity Index-Based Algorithms for Endemember Extraction from Hyperspectral Imagery, Chapter 2, Recent Advances in Hyperspectral Signal and Image Processing, edited by C.-I Chang, Trivandrum, Kerala: Research Signpost, India, 2006. [16] C-I Chang and A. Plaza, “Fast iterative algorithm for implementation of pixel purity index,” IEEE Trans. on Geoscience and Remote Sensing Letters, vol. 3, no. 1, pp. 63-67, January 2006. [17] S. Haykin, Neural Networks: A Comprehensive Foundation, 2nd ed., Prentice-Hall, 1999, Chapter 6. [18] R. O. Duda and P.O. Hart, Pattern Classification and Scene Analysis, New York: John Wiley & Sons, 1973. [19] Boardman, 1992, SIPS User's Guide Spectral Image Processing System, Version 1.2, Center for the Study of Earth from Space, Boulder. [20] F. A., Kruse.; A. B., Lefkoff.; J. W., Boardman; Heiedbrecht; A. T.; P. J. , Barloon; Goetz, A. F. H.,1992. The Spectral Image Processing System (SIPS) - Software for Integrated Analysis of AVIRIS DataSummaries of the 4th Annual JPL Airborne Geoscience Workshop, JPL Pub-92-14, AVIRISWorkshop. Jet Propulsion Laboratory, Pasadena, CA, pp. 23-25. [21] H. Soltanian-Zadeh, J. P. Windham and D.J. Peck, “Optimal linear transformation for MRI feature extraction,” IEEE Transaction on Medical Imaging, 15(6): 749-767, 1996. [22] J. C. Harsanyi, Detection and Classification of Subpixel Spectral Signatures in Hyperspectral Image Sequences, Department of Electrical Engineering, University of Maryland Baltimore County, Baltimore, MD, August 1993. [23] R. C., Gonzalez, R. E., Woods, and Eddins, Digital Image Processing Using MATLAB, Prentice Hall, Upper Saddle River, NJ, 2004; 164-165. [24] http://www.bic.mni.mcgill.ca/brainweb, first date: May, 1997; last modified date: Jun, 2006. [25] S. Theodoridis, K. Koutroumbas. Pattern recognition. 2nd, New York, Academic Press 2003; 411-415. [26] A. Hyvarinen, and E. Oja, Independent Component Analysis: Algorithms and Applications, Neural Networks, 13(4-5):411-430, 2000. [27] A. Hyvarinen, The fixed-point algorithm and maximum likelihood estimation for independent component analysis, Neural Processing Letters, 1999. [28] V. N. Vapnik, Statistical Learning Theory, New York: John Wiley & Sons, 1998. [29] R. O. Duda and P. O. Hart, Pattern Classification and Scene Analysis, New York: John Wiley & Sons, 1973. [30] H. Ren, Q. Du, J. Wang, C.-I Chang and J. Jensen, “Automatic target recognition hyperspectral imagery using high order statistics,” IEEE Trans. on Aerospace and Electronic Systems, vol. 42, no. 4, pp. 1372-1385, Oct. 2006. [31] I. L. Chen, C. S. Huang, B.C. Kuo and C. H. Li “An automatic algorithm to choice a composite kernel,” The 23th Computer Vision, Graphics, and Image Processing, August, 15-17, 2010.
摘要: 
本論文主要將支援向量機(SVM)的高斯核心(RBF kernel)衍伸為權重式高斯核心(weighted-RBF kernel)。一般我們使用0.5的自相關矩陣作為RBF kernel 的係數,本研究將探討依影像本身之相關矩陣(correlation matrix)、共變異數矩陣(covariance matrix)及內變異數矩陣(within-class covariance matrix)運用在RBF kernel上,並以不同比例隨機選取訓練樣本點。其主要目的是可以依據影像本身之統計特性,選取適合的RBF kernel,對腦部磁振造影影像做出有效的分類。在SVM中有兩個重要的議題,第一是如何選擇kernel;第二是如何決定訓練樣本的數量。為解決第一項議題,我們使用了weighted-RBF kernel。第二項議題,我們以不同百分比數量的樣本來探討。最後,我們使用支援向量器(SVM)以及費雪線性辨別分析(FLDA)來提高分類效果。實驗結果顯示本論文所提出的演算法,對於腦部磁振造影影像做出有效的分類。

In this thesis, we will extend the RBF kernel of support vector machine (SVM) to weighted RBF kernel. The default coefficient of RBF kernel is an identity matrix with 0.5. This study will explore the correlation matrix, the covariance matrix and the covariance matrix by image itself on the use of the RBF kernel, and with different training samples which were randomly selected. Its main purpose is that we can select an appropriate RBF kernel who base on the statistical properties of image itself. Two major challenging issues arise in SVM: one is how to choice a proper kernel; the other is how to decide the quantity about training samples. To address the first issue, we propose the W-RBF kernel. To address the second issue, we discuss different number of percentage of samples. Finally, we combine the support vector control (SVM) and Fisher linear discriminant analysis (FLDA) to improve the classification results. Experimental results show that that the proposed algorithm has great promise in the brain magnetic resonance image.
URI: http://hdl.handle.net/11455/4870
其他識別: U0005-2707201121163000
Appears in Collections:通訊工程研究所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.