Combined Feature Based Hyperspectral Image Classification Technique Using Support Vector Machines
A spatial classification technique incorporating a State of Art Feature Extraction algorithm is proposed in this paper for classifying a heterogeneous classes present in hyper spectral images. The classification accuracy can be improved if and only if both the feature extraction and classifier selection are proper. As the classes in the hyper spectral images are assumed to have different textures, textural classification is entertained. Run Length feature extraction is entailed along with the Principal Components and Independent Components. A Hyperspectral Image of Indiana Site taken by AVIRIS is inducted for the experiment. Among the original 220 bands, a subset of 120 bands is selected. Gray Level Run Length Matrix (GLRLM) is calculated for the selected forty bands. From GLRLMs the Run Length features for individual pixels are calculated. The Principle Components are calculated for other forty bands. Independent Components are calculated for next forty bands. As Principal & Independent Components have the ability to represent the textural content of pixels, they are treated as features. The summation of Run Length features, Principal Components, and Independent Components forms the Combined Features which are used for classification. SVM with Binary Hierarchical Tree is used to classify the hyper spectral image. Results are validated with ground truth and accuracies are calculated.
[1] David Landgrebe, "Some Fundamentals and Methods for Hyper Spectral Image Data Analysis", SPIE Photonics, pp.1-10, Jan-1999.
[2] G. F. Hughes, "On the mean accuracy of statistical pattern recognizers", IEEE Trans. Inf. Theory, vol. IT-14, no. 1, pp. 55-63, Jan. 1968.
[3] Yulia Tarabalka, Jon Atli Beneditksson, Joycelyn Chanussot, "Spectral-Spatial Classification of Hyperspectral Imagery based on Partitional Clustering Techniques", IEEE Trans. On Geoscience and Remote sensing, Vol-47,No-8,Aug 2009, pp 2973¬2987.
[4] Galloway,"Texture information in Run Length Matrices",IEEE Transactions on Image Processing,Nov-1998,Vol.7,No.11, pp:1602- 1609.
[5] R. M. Haralick, K. Shanmugam, I. Dinstein, "Texture features for image classification." IEEE Trans. System Man Cybernetics, Vol.8, No.6, pp. 610-621, 1973.
[6] Horng-Hai Loh, Jia Guu Leu," The Analysis of natural Textures using Run Length Features", IEEE Transactions on Industrial Electronics, May 1988, Vol.35,No.2, pp:323-328
[7] Xiaoou Tang,"Dominant Run- Length Method for Image Classification" Department of Applied Ocean Physics and Engineering, Woods Hole Oceanographic Institution, Woods Hole, Report-MA 02543.
[8] Maneesha Singh and Sameer Singh,"Spatial Texture Analysis: A Comparative Study", IEEE conference on Image Processing, 2002, pp:676-679.
[9] Poonguzhali, and G.Ravindran," Automatic classification of focal Lesion in Ultrasound Liver Images using Combined Texture features, Information Technology Journal, 2008, pp:205-209.
[10] Wiseline Jiji and L.Ganesan," Unsupervised Texure Classification, Journal of Theoretical and Applied Information Technology,2009, pp:373-381
[11] Renuka Methra, and M.Ravindranath, " Exploring Spatial Information in Spectral features for Texture Image retrival", International Journal of Computer and network Security, Dec.2009,Vol 1,No.3, pp:46-49
[12] Lindsay I Smith, "A tutorial On Principal Components Analysis", ICIC International Journal, Febuary2002, Vol. 4, No.5
[13] Andreas Weingessel and Kurt Hornik, "Local PCA Algorithms", IEEE Transactions on Neural Networks, November 2000, Vol.11, No.6.
[14] J. A. K. Suykens, T. Van Gestel, J.Vandewalle, and B. De Moor, "A Support Vector Machine Formulation to PCA Analysis and Its Kernel Version", IEEE Transactions on Neural Networks, March 2003, Vol. 14, No. 2, pp:447-450
[15] Yuka Higashijima, Shigeru Takano, and Koichi Niijima, "Image Classification by Lifting Wavelet PCA", Journal of Kyushu University, Japan.
[16] Zhan-Li Sun,De-Shuang Huang,Yiu-Ming Cheung,"Using FCMC, FVS, and PCA techniques for feature Extraction of Multispectral Images, IEEE Transactions on Geosciences and Remotesensing Letters,Apr-2005, Vol.2,No.2, pp: 108-112
[17] M.Asuncion Vicente, Patrik O.Hoyer and Aapo Hyvarinen, "Equivalence of Some Common Linear Feature Extraction Techniques for Appearance-Based Object Recognition Tasks", IEEE Transactions on Pattern Analysis and Machine Intelligence, May 2007, Vol.29, No.5
[18] A. Hyvarinen, J. Karhunen, and E. Oja, "Independent Component Analysis"
[19] Hyvarinen, A. and E. Oja, "A fast fixed-point algorithm for independent component analysis", Journal of Neural Computation, vol.9, pp.1483-1492, 1997
[20] Mark D. Plumbley, and Erkki Oja, "A Nonnegative PCA Algorithm for Independent Component Analysis", IEEE Transactions on Neural Networks, January 2004, Vol.15, No. 1.
[21] C.A.Shah, P.Watanachaturaporn,p.K.varshney,M.K.Arora," Some recent results on Hyperspectral Image Classification", IEEE conference on Geosciences and remotesensing,2004, pp:346-353.
[22] Jon Aevar Palmason, Jon Atli Benedikson,Johannes
R.Sveinsson,Jocelyn Chanussot," Classification of Hyperspectral data from Urban Areas using Morphological Preprocessing and Independent Component Analysis, IEEE conference on Geosciences and Remotesensing,2005,Voll , pp :176-180
[23] Jian Yang, David Zhang, and JingYu Yang, "Constructing PCA Baseline Algorithms to Reevaluate ICA-Based Face-Recognition Performance", IEEE Transactions on Systems, Man, and Cybernetics—Part B: Cybernetics, August 2007, Vol.37, No. 4, pp:1015-1021
[24] Xiao Benlin Li Fangfang,Mao Xingliang, Jin Huazhong, " The International Archives of the photogrammetry, remote Sensing and Spatial Information sciences, Beijing 2008, Vol. XXXVII, Part B7.
[25] Weibo Zou, Yan Li King Chuen Lo, Zheru Chi," Improvement of Image Classification with Wavelet and Independent Component Analysis(ICA) based on a structrured Neural Networks, International Joint Conference on Neural Networks, July 2006, pp:3949-3954.
[26] David Landgrebe, "Information Extraction Principles and Methods for Multispectral and Hyper spectral Image Data", pp.1-30, November 1998.
[27] V. N. Vapnik, Statistical Learning Theory. New York: Wiley, 1998.
[28] Farid Melgani, Lorenzo Bruzzone, "Classification of Hyperspectral Remote Sensing Images With Support Vector Machines", IEEE Trans. Geosci. Remote Sensing, August 2004,vol. 42, no. 8, pp. 1778-1790
[29] G. Camps-Valls and L. Bruzzone, "Kernel-based methods for Hyperspectral image classification," IEEE Trans. Geosci. Remote Sens., Jun. 2005, vol. 43,no. 6, pp. 1351-1362
[30] M. Fauvel, "Spectral and spatial methods for the classification of urban remote sensing data," Ph.D.dissertation, Grenoble Inst. Technol., Grenoble, France, 2007.
Gualtieri and R. F. Cromp, "Support Vector Machines for Hyperspectral remote sensing classification," Proc. SPIE, vol. 3584, pp. 221-232, Jan. 1998.
[31] M. Fauvel, J. Chanussot, and J. A. Benediktsson,"Evaluation of Kernels for Multiclass Classification of Hyperspectral remote sensing data," in Proc. ICASSP, May2006, pp. II-813-11-816.
[32] Gustavo Camps-Valls, Lorenzo Bruzzone, "Kernel-Based Methods for Hyperspectral Image Classification", IEEE Trans. Geosci. Remote Sensing, June 2005, Vol.43, No.6; pp:1351-1363
[1] David Landgrebe, "Some Fundamentals and Methods for Hyper Spectral Image Data Analysis", SPIE Photonics, pp.1-10, Jan-1999.
[2] G. F. Hughes, "On the mean accuracy of statistical pattern recognizers", IEEE Trans. Inf. Theory, vol. IT-14, no. 1, pp. 55-63, Jan. 1968.
[3] Yulia Tarabalka, Jon Atli Beneditksson, Joycelyn Chanussot, "Spectral-Spatial Classification of Hyperspectral Imagery based on Partitional Clustering Techniques", IEEE Trans. On Geoscience and Remote sensing, Vol-47,No-8,Aug 2009, pp 2973¬2987.
[4] Galloway,"Texture information in Run Length Matrices",IEEE Transactions on Image Processing,Nov-1998,Vol.7,No.11, pp:1602- 1609.
[5] R. M. Haralick, K. Shanmugam, I. Dinstein, "Texture features for image classification." IEEE Trans. System Man Cybernetics, Vol.8, No.6, pp. 610-621, 1973.
[6] Horng-Hai Loh, Jia Guu Leu," The Analysis of natural Textures using Run Length Features", IEEE Transactions on Industrial Electronics, May 1988, Vol.35,No.2, pp:323-328
[7] Xiaoou Tang,"Dominant Run- Length Method for Image Classification" Department of Applied Ocean Physics and Engineering, Woods Hole Oceanographic Institution, Woods Hole, Report-MA 02543.
[8] Maneesha Singh and Sameer Singh,"Spatial Texture Analysis: A Comparative Study", IEEE conference on Image Processing, 2002, pp:676-679.
[9] Poonguzhali, and G.Ravindran," Automatic classification of focal Lesion in Ultrasound Liver Images using Combined Texture features, Information Technology Journal, 2008, pp:205-209.
[10] Wiseline Jiji and L.Ganesan," Unsupervised Texure Classification, Journal of Theoretical and Applied Information Technology,2009, pp:373-381
[11] Renuka Methra, and M.Ravindranath, " Exploring Spatial Information in Spectral features for Texture Image retrival", International Journal of Computer and network Security, Dec.2009,Vol 1,No.3, pp:46-49
[12] Lindsay I Smith, "A tutorial On Principal Components Analysis", ICIC International Journal, Febuary2002, Vol. 4, No.5
[13] Andreas Weingessel and Kurt Hornik, "Local PCA Algorithms", IEEE Transactions on Neural Networks, November 2000, Vol.11, No.6.
[14] J. A. K. Suykens, T. Van Gestel, J.Vandewalle, and B. De Moor, "A Support Vector Machine Formulation to PCA Analysis and Its Kernel Version", IEEE Transactions on Neural Networks, March 2003, Vol. 14, No. 2, pp:447-450
[15] Yuka Higashijima, Shigeru Takano, and Koichi Niijima, "Image Classification by Lifting Wavelet PCA", Journal of Kyushu University, Japan.
[16] Zhan-Li Sun,De-Shuang Huang,Yiu-Ming Cheung,"Using FCMC, FVS, and PCA techniques for feature Extraction of Multispectral Images, IEEE Transactions on Geosciences and Remotesensing Letters,Apr-2005, Vol.2,No.2, pp: 108-112
[17] M.Asuncion Vicente, Patrik O.Hoyer and Aapo Hyvarinen, "Equivalence of Some Common Linear Feature Extraction Techniques for Appearance-Based Object Recognition Tasks", IEEE Transactions on Pattern Analysis and Machine Intelligence, May 2007, Vol.29, No.5
[18] A. Hyvarinen, J. Karhunen, and E. Oja, "Independent Component Analysis"
[19] Hyvarinen, A. and E. Oja, "A fast fixed-point algorithm for independent component analysis", Journal of Neural Computation, vol.9, pp.1483-1492, 1997
[20] Mark D. Plumbley, and Erkki Oja, "A Nonnegative PCA Algorithm for Independent Component Analysis", IEEE Transactions on Neural Networks, January 2004, Vol.15, No. 1.
[21] C.A.Shah, P.Watanachaturaporn,p.K.varshney,M.K.Arora," Some recent results on Hyperspectral Image Classification", IEEE conference on Geosciences and remotesensing,2004, pp:346-353.
[22] Jon Aevar Palmason, Jon Atli Benedikson,Johannes
R.Sveinsson,Jocelyn Chanussot," Classification of Hyperspectral data from Urban Areas using Morphological Preprocessing and Independent Component Analysis, IEEE conference on Geosciences and Remotesensing,2005,Voll , pp :176-180
[23] Jian Yang, David Zhang, and JingYu Yang, "Constructing PCA Baseline Algorithms to Reevaluate ICA-Based Face-Recognition Performance", IEEE Transactions on Systems, Man, and Cybernetics—Part B: Cybernetics, August 2007, Vol.37, No. 4, pp:1015-1021
[24] Xiao Benlin Li Fangfang,Mao Xingliang, Jin Huazhong, " The International Archives of the photogrammetry, remote Sensing and Spatial Information sciences, Beijing 2008, Vol. XXXVII, Part B7.
[25] Weibo Zou, Yan Li King Chuen Lo, Zheru Chi," Improvement of Image Classification with Wavelet and Independent Component Analysis(ICA) based on a structrured Neural Networks, International Joint Conference on Neural Networks, July 2006, pp:3949-3954.
[26] David Landgrebe, "Information Extraction Principles and Methods for Multispectral and Hyper spectral Image Data", pp.1-30, November 1998.
[27] V. N. Vapnik, Statistical Learning Theory. New York: Wiley, 1998.
[28] Farid Melgani, Lorenzo Bruzzone, "Classification of Hyperspectral Remote Sensing Images With Support Vector Machines", IEEE Trans. Geosci. Remote Sensing, August 2004,vol. 42, no. 8, pp. 1778-1790
[29] G. Camps-Valls and L. Bruzzone, "Kernel-based methods for Hyperspectral image classification," IEEE Trans. Geosci. Remote Sens., Jun. 2005, vol. 43,no. 6, pp. 1351-1362
[30] M. Fauvel, "Spectral and spatial methods for the classification of urban remote sensing data," Ph.D.dissertation, Grenoble Inst. Technol., Grenoble, France, 2007.
Gualtieri and R. F. Cromp, "Support Vector Machines for Hyperspectral remote sensing classification," Proc. SPIE, vol. 3584, pp. 221-232, Jan. 1998.
[31] M. Fauvel, J. Chanussot, and J. A. Benediktsson,"Evaluation of Kernels for Multiclass Classification of Hyperspectral remote sensing data," in Proc. ICASSP, May2006, pp. II-813-11-816.
[32] Gustavo Camps-Valls, Lorenzo Bruzzone, "Kernel-Based Methods for Hyperspectral Image Classification", IEEE Trans. Geosci. Remote Sensing, June 2005, Vol.43, No.6; pp:1351-1363
@article{"International Journal of Electrical, Electronic and Communication Sciences:60437", author = "Mrs.K.Kavitha and S.Arivazhagan", title = "Combined Feature Based Hyperspectral Image Classification Technique Using Support Vector Machines", abstract = "A spatial classification technique incorporating a State of Art Feature Extraction algorithm is proposed in this paper for classifying a heterogeneous classes present in hyper spectral images. The classification accuracy can be improved if and only if both the feature extraction and classifier selection are proper. As the classes in the hyper spectral images are assumed to have different textures, textural classification is entertained. Run Length feature extraction is entailed along with the Principal Components and Independent Components. A Hyperspectral Image of Indiana Site taken by AVIRIS is inducted for the experiment. Among the original 220 bands, a subset of 120 bands is selected. Gray Level Run Length Matrix (GLRLM) is calculated for the selected forty bands. From GLRLMs the Run Length features for individual pixels are calculated. The Principle Components are calculated for other forty bands. Independent Components are calculated for next forty bands. As Principal & Independent Components have the ability to represent the textural content of pixels, they are treated as features. The summation of Run Length features, Principal Components, and Independent Components forms the Combined Features which are used for classification. SVM with Binary Hierarchical Tree is used to classify the hyper spectral image. Results are validated with ground truth and accuracies are calculated.
", keywords = "Multi-class, Run Length features, PCA, ICA, classification and Support Vector Machines.", volume = "4", number = "10", pages = "1515-8", }
