
系統識別號 
U00260812200910453971 
論文名稱(中文) 
小波轉換應用於高光譜影像光譜特徵萃取之研究 
論文名稱(英文) 
Spectral Feature Extraction of Hyperspectral Images using Wavelet Transform 
校院名稱 
成功大學 
系所名稱(中) 
測量工程學系碩博士班 
系所名稱(英) 
Department of Geomatics 
學年度 
91 
學期 
2 
出版年 
92 
研究生(中文) 
徐百輝 
研究生(英文) 
PaiHui Hsu 
學號 
p6885101 
學位類別 
博士 
語文別 
英文 
論文頁數 
146頁 
口試委員 
口試委員王蜀嘉 口試委員陳哲俊 指導教授宮鵬 指導教授曾義星 口試委員蔡展榮 口試委員史天元

中文關鍵字 
高光譜影像
小波轉換
特徵萃取

英文關鍵字 
Hyperspectral Images
Feature Extraction
Wavelet Transform

學科別分類 

中文摘要 
成像光譜儀所擷取的高光譜影像具有豐富且細緻的地物光譜資訊，此特性對於提升地物辨識能力及土地使用分類精度將有所助益。然而若以傳統的統計分類方法對高光譜影像進行分類時，其高維度的特性要求必須有大量的訓練樣本數目，同時大批的資料量亦降低了計算效率，分類精度也沒有顯著地提升。主要的問題在於當統計方法應用於高維度資料時所產生的「維度詛咒」現象。為了解決這個問題，本研究著重在光譜資料維度的降低，特徵萃取即為降低資料維度的一種方式，其基本觀念為去除較不重要的光譜資訊，只保留有用的特徵以降低高光譜資料的維度。本研究主要提出數種以小波理論為基礎的特徵萃取方法，以獲得對高光譜影像分類有用之光譜特徵。小波轉換為新一代的數學分析工具，其多層解析度及時不變的特性，使得其具有偵測局部訊號結構的能力。首先，我們先對高光譜影像進行離散小波或小波包轉換以獲得一組小波係數，之後再根據所設計的判斷法則選出有利於分類的特徵。因為特徵選取時係在維度較低的子空間中進行，因此有限樣本數目及「維度詛咒」等問題皆可以獲得有效的解決。此外，高光譜影像所提供的吸收帶特徵與光譜反射量等資訊亦可與地物的物理及化學特性產生關連，以往的光譜吸收帶特徵係由專家以人工方式確定，因此本研究亦提出以連續小波轉換自動偵測光譜吸收特徵的方法。由於小波轉換的多尺度特性，我們可以很精確地確定光譜吸收特徵之局部結構。當每一個單獨的吸收帶特徵位置被確定之後，即可以連續統消除演算法去除其他的外在效應，如背景吸收帶所造成的輻射量變化及光譜曲線的傾斜。進一步我們定義並計算描述單一吸收帶特徵之參數，未來將可作為光譜比對或地物辨識之用。本研究以兩組高光譜資料來測試所提小波轉換應用於光譜特徵萃取之有效性。實驗結果顯示以小波理論所萃取出的光譜特徵確實可以有效降低高光譜影像的資料維度，同時保持影像分類之精度；此外連續小波方法亦可以有效地萃取出大氣吸收帶特徵的位置。

英文摘要 
The rich and detailed spectral information provided by hyperspectral images can be used to identify and quantify a large range of surface materials which cannot be identified by multispectral images. However, the classification methods that have been successfully applied to multispectral data in the past are not as effective as for hyperspectral data. The main problem is that the training data set does not increase corresponding to the increase of dimensionality of hyperspectral data. Actually, the problem of the “curse of dimensionality” emerges when a statistical classification method applied to hyperspectral data. A simpler, but sometimes very effective way of dealing with hyperspectral data is to reduce the number of dimensionality. This can be done by feature extraction that a small number of salient features are extracted from the hyperspectral data when confronted with a limited set of training samples. The goal of feature extraction is to reduce the number of dimensionality substantially without sacrificing useful information. In this study, several methods based on the wavelet transforms are developed to extract useful features for classification. Firstly, wavelet or wavelet packet transforms are implemented on the hyperspectral images and a sequence of wavelet coefficients is produced. Then, a simple feature selection procedure associated with a criterion is used to select the effective features for classification. Because the waveletbased feature extraction optimizes the criterion in a lower dimensional space, the problems of limited training sample size and the curse of dimensionality can be avoided. On the other hand, the laboratorylike spectral curves also describe diagnostic absorption and reflection features which are valuable to understand the physical or chemical properties of materials. The other objective of this study is to propose an automatic procedure for the detection of absorption features. By the zooming procedure of continuous wavelet transform (CWT), these localized structures of absorption features can be accurately characterized. While the position of an individual absorption feature is determined, the continuum removal algorithm can be implemented to reduce some additive effects, such as level changes and slopes due to other absorbing or emitting materials. Consequently the parameters of a single absorption feature can be calculated for spectral matching or material identification. Finally, two AVIRIS data sets are used to test the performance of the proposed waveletbased methods. The experiment results show that the waveletbased methods perform well for dimensionality reduction and also be effective for classification.

論文目次 
中 文 摘 要 I
ABSTRSCT II
ACKNOWLEDGEMENTS III
CONTENTS IV
LIST OF FIGURES VII
LIST OF TABLES X
CHAPTER 1 INTRODUCTION  1 
1.1 IMAGING SPECTROMETERS  1 
1.2 MOTIVATIONS  4 
1.3 RESEARCH OBJECTIVES  6 
1.4 THESIS ORGANIZATION  7 
CHAPTER 2 PROPERTIES OF HYPERSPECTRAL IMAGES  9 
2.1 INTRODUCTION  9 
2.2 CHEMICAL AND PHYSICAL PROPERTIES  10 
2.2.1 Absorption Bands  10 
2.2.2 Atmospheric Correction  14 
2.2.3 Summary  17 
2.3 STATISTICAL AND GEOMETRICAL PROPERTIES  18 
2.3.1 Spectral Signatures and Distributions  18 
2.3.2 Basic Statistics of Distributions  21 
2.3.3 Geometric Properties in HighDimensional Space  23 
2.3.4 Summary  26 
2.4 CURSE OF DIMENSIONALITY  27 
2.4.1 Statistical Pattern Recognition  27 
2.4.2 Curse of Dimensionality  28 
2.4.3 Summary  30 
2.5 VISUALIZATION OF HYPERSPECTRAL IMAGES  31 
2.5.1 Introduction  31 
2.5.2 SpatialSpectral Space  32 
2.5.3 Spectral Space  33 
2.5.4 Scatter Plots  36 
2.5.5 Parallel Coordinates  39 
2.5.6 Statistics Images  41 
2.5.7 Symbolic Representation  43 
CHAPTER 3 DIMENSIONALITY REDUCTION OF HYPERSPECTERAL IMAGE  45 
3.1 INTRODUCTION  45 
3.2 FEATURE SELECTION  46 
3.2.1 Criterion Function  47 
3.2.2 Optimal Search Procedure  48 
3.2.3 Suboptimal Search Procedure  51 
3.2.4 Experiments  53 
3.3 FEATURE EXTRACTION  54 
3.3.1 The Definition of Features  55 
3.3.2 Ideal Features for Classification  56 
3.4 FEATURE EXTRACTION ALGORITHM OVERVIEW  57 
3.4.1 Principal Components Analysis  58 
3.4.2 Linear Discriminant Analysis  59 
3.4.3 Decision Boundary Feature Extraction Algorithm  62 
3.5 EXPERIMENTS  63 
3.5.1 Experiment I  63 
3.5.2 Experiment II  67 
3.6 SUMMARY  72 
CHAPTER 4 WAVELETBASED FEATURE EXTRACTION  73 
4.1 INTRODUCTION  73 
4.2 FOURIERBASED FEATURE EXTRACTION  74 
4.3 WAVELETBASED FEATURE EXTRACTION  77 
4.3.1 Wavelet Transform  77 
4.3.2 Fast Orthogonal Wavelet Transform  80 
4.3.3 Linear Wavelet Feature Extraction  82 
4.3.4 Experiment I  88 
4.3.5 Nonlinear Wavelet Feature Extraction  91 
4.3.6 BestBasis Feature Extraction  94 
4.3.7 Local Discriminant Basis Feature Extraction  100 
4.4 EXPERIMENTS  103 
4.4.1 Experiment I  103 
4.4.2 Experiment II  112 
4.5 SUMMARY  114 
CHAPTER 5 ABSOPRTION FEATURE DETECTION OF HYPERSPECTRAL IMAGES  116 
5.1 INTRODUCTION  116 
5.2 ABSORPTION FEATURE EXTRACTION  117 
5.2.1 Definition of Absorption Features  117 
5.2.2 Continuum Removal of Absorption Features  118 
5.2.3 Summary of Methods for Absorption Features Extraction  120 
5.2.4 Fingerprints  121 
5.3 WAVELETBASED FEATURE EXTRACTION  123 
5.3.1 Continuous Wavelet Transform  123 
5.3.2 Absorption Feature Detection using Modulus Maxima Wavelet Transform  124 
5.4 EXPERIMENTS  133 
5.4.1 Chlorophyll Absorption Detection  133 
5.4.2 Absorption Information Image  135 
5.5 SUMMARY  136 
CHAPTER 6 CONCLUSION  138 
6.1 Summary  138 
6.2 SuggestIONs for Further Work  140 
BIBLIOGRAPHY  142 

參考文獻 
Bellman, R., Adaptive Control Processes: A Guided Tour: Princeton University Press, 1961.
Benediktsson, J. A., J. R. Sveinsson, and K. Arnason, "Classification and Feature Extraction of AVIRIS Data," IEEE Transactions on Geoscience and Remote Sensing, vol. 33, pp. 11941205, 1995.
Boardman, J. W., "Automating Spectral Unmixing of AVIRIS Data Using Convex Geometry Concepts," presented at Summeries of the Fourth Annual JPL Ariborne Geoscience Workshop, Pasadena, CA, 1993.
Boardman, J. W., F. A. Kruse, and R. O. Green, "Mapping Target Signatures via Partial Unmixing of AVIRIS Data," presented at Summaries of Fifth JPL Airborne Earth Science Workshop, Pasadena, CA, 1995.
Bruce, L. M. and J. Li, "Wavelets for Computationally Efficient Hyperspectral Derivative Analysis," IEEE Transactions on Geoscience and Remote Sensing, vol. 39, pp. 15401546, 2001.
Burns, R. G., Mineralogical Applications of Crystal Field Theory, Second ed. New York: Cambridge University Press, 1993.
Clark, R. N. and T. L. Roush, "Reflectance Spectroscopy: Quantitative Analysis Techniques for Remote Sensing Applications," Journal of Geophys. Res., vol. 89, pp. 63296340, 1984.
Clark, R. N., G. A. Swayze, and A. Gallagher, "Mapping the Mineralogy and Lithology of Canyonlands, Utah with Imaging Spectrometer Data and the Multiple Spectral Feature Mapping Algorithm," presented at Summaries of the Third Annual JPL Airborne Geosciences Workshop, 1992.
Clark, R. N., G. Swayze, K. Heidebrecht, A. F. H. Goetz, and R. O. Green, "Comparison of Methods for Calibrating AVIRIS Data to Ground Reflectance," presented at Summeries of the Fourth Annual JPL Ariborne Geoscience Workshop, Pasadena, CA, 1993.
Clark, R. N. and G. A. Swayze, "Mapping Minerals, Amorphous Materials, Environmental Materials, Vegetation, Water, Ice and Snow, and Other Materials: The USGS Tricorder Algorithm," presented at Summeries of the Fifth Annual JPL Airborne Earth Science Workshop, 1995a.
Clark, R. N., T. V. V. Kingm, C. Ager, and G. A. Swayze, "Initial Vegetation Species and Senescence/stress Mapping in the San Luis Calley, Colorado using Imaging Spectrometer data," presented at Summaries of the Fifth Annual JPL Airborne Earth Science Workshop, 1995b.
Clark, R. N., "Spectroscopy of Rocks and Minerals, and Principles of Spectroscopy," in Manual of Remote Sensing, A. Rencz, Ed. New York: John Wiley and Sons, Inc, 1999.
Clark, R. N., G. A. Swayze, K. E. Livo, R. F. Kokaly, T. King, V. V., J. B. Dalton, J. S. Vance, B. W. Rockwell, T. Hoefen, and R. R. McDougal, "Surface Reflectance Calibration of Terrestrial Imaging Spectroscopy Data: a Tutorial Using AVIRIS," presented at AVIRIS Workshop Proceedings, 2002.
Clevers, J. G. P. W., "The Use of Imaging Spectromatry for Agricultural Applications," ISPRS Journal of Photogrammetry & Remote Sensing, vol. 54, pp. 299304, 1999.
Coifman, R. R., Y. Meyer, and M. V. Wickerhauser, "Wavelet Analysis and Signal Processing," in Wavelets and Their Applications, M. B. Ruskai, Ed. Boston: Jones and Barlett, 1992, pp. 153178.
Coifman, R. R. and M. V. Wickerhauser, "EntropyBased Algorithms for Best Basis Selection," IEEE Transactions on Information Theory, vol. 38, pp. 713718, 1992.
Conel, J. E., R. O. Green, G. Vane, C. J. Bruegge, and R. E. Alley, "AIS2 Radiometry and a Comparison of Methods for the Recovery of Ground Reflectance," presented at Proceedings of the Third Airborne Imaging Spectrometer Data Analysis Workshop, Pasadena, CA, 1987.
Cover, T. M., "The Best Two Independent Measurements are not the Two Best," IEEE Transactions on Systems, Man, and Cybernetics, vol. 4, pp. 116117, 1974.
Daubechies, I., "Orthonormal Bases of Compactly Supported Wavelets," Communications on Pure and Applied Mathematics, vol. 41, pp. 909996, 1988.
Daubechies, I., Ten Lectures on Wavelets. Philadelphia, PA: Society of Industrial and Applied Mathematics, 1992.
DemetriadesShah, T. H., M. D. Steven, and J. A. Clark, "High Resolution Derivative Spectra in Remote Sensing," Remote Sensing of Environment, vol. 33, pp. 5564, 1990.
Fau, Y.H., M. O. Ward, and E. A. Rundensteiner, "Hierarchical Parallel Coordinates for Exploration of Large Data Sets," presented at proceedings of IEEE Conerence on Visualization '99, San Francisco, CA, 1999.
Fu, K. S. and T. Young, Handbook of pattern recognition and image processing. Orlando: Academic Press, 1986.
Fukunaga, K., Introduction to Statistical Pattern Recognition, Second ed. San Diego: Academic Press, Inc., 1990.
Galvao, L. S., M. A. Pizarro, and J. C. N. Epiphanio, "Variations in Reflectance of Tropical Soils: SpectralChemical Composition Relationships from AVIRIS Data," Remote Sensing Environment, vol. 75, pp. 245255, 2001.
Gao, B.C. and A. F. H. Goetz, "Derivation of Scaled Surface Reflectances from AVIRIS Data," Remote Sensing of Environment, vol. 44, pp. 145163, 1993.
Goetz, A. F. H., G. Vane, J. E. Solomon, and B. N. Rock, "Imaging Spectrometry for Earth Remote Sensing," Science, vol. 228, pp. 11471153, 1985.
Goetz, A. F. H., J. W. Boardman, B. Kindel, and K. B. Heidebrecht, "Atmospheric Corrections: On Deriving Surface Reflectance from Hyperspectral Imagers," presented at Proceedings SPIE Annual Meeting, 3118, 1997.
Gong, P., R. Pu, and B. Yu, "Conifer Species Recongnition: An Exploratory Analysis of In Situ Hyperspectral Data," Remote Sensing Environment, vol. 62, pp. 189200, 1997.
Grossmann, A. and J. Morlet, "Decomposition of Hardy Functions into Square Intergrable Wavelets of Constant Shapes," SIAM Journal of Math. Anal., vol. 15, pp. 723736, 1984.
Hsieh, P.F. and D. A. Landgrebe, "Classification of High Dimensional Data," in School of Electrical & Computer Engineering. West Lafayette: Purdue University, 1998
Hsu, P.H. and Y.H. Tseng, "Feature Extraction for Hyperspectral Images," presented at 20th Asian Conference on Remote Sensing, Hong Kong, 1999.
Hsu, P.H. and Y.H. Tseng, "Multiscale Analysis of Hyperspectral Data Using Wavelets for Spectral Feature Extraction," presented at 21th Asia Conference on Remote Sensing, Taipei, 2000.
Hughes, G. F., "On the Mean Accuracy of Statistical Pattern Recognizers," IEEE Transactions on Information Theory, vol. IT14, pp. 5563, 1968.
Hunt, G. R., "Spectral Signatures of Particulate Minerals, in the Visible and NearInfrared," Geophysics, vol. 44, pp. 19741986, 1977.
Inselberg, A., "NDimensional Graphics, Part I  Lines and Hyperplanes," Technical Report G3202711, IBM Los Angleles Sicentific Center, Los Angles, CA 1981 1981.
Inselberg, A., "A Tool for Visualizing Multidimensional Geometry," presented at IEEE Visualization 1990, 1990.
Jain, A. K., R. P. W. Duin, and J. Mao, "Statistical Pattern Recognition: A Review," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, pp. 437, 2000.
Jimenez, L. O. and D. A. Landgrebe, "High Dimensional Feature Reduction Via Projection Pursuit," in School of Electrical & Computer Engineering. West Lafayette: Purdue University, pp. 137, 1995
Jimenez, L. O. and D. A. Landgrebe, "Supervised Classification in HighDimensional Space: Geometrical, Statistical, and Asymptotical Properties of Multivariate Data," IEEE Transactions on Geoscience and Remote Sensing, vol. 28, pp. 3954, 1998.
Kruse, F. A., "Mapping Hydrothermally Altered Rocks in the Northern Grapevine Mountains, Nevada and California with the Airborne Imaging Spectrometer," presented at Proceedings of the Third Airborne Imaging Spectrometer Data Analysis Workshop, Pasadena, CA, 1987.
Kulkarni, S. R., G. Lugosi, and S. S. Venkatesh, "Learning Pattern Classification  A Survey," IEEE Transactions on Information Theory, vol. 44, pp. 21782206, 1998.
Landgrebe, D. A., "On Information Extraction Principles for Hyperspectral Data (A White Paper)," School of Electrical & Computer Engineering, Purdue University, West Lafayette 1997.
Landgrebe, D. A., "Analysis of Multispectral and Hyperspectral Image Data," in Introduction to Modern Photogrammetry, E. M. Mikhail and J. S. Bethel, Eds. Chris McGlone: John Wiley & Sons, Inc, 2001.
Lee, C. and D. A. Landgrebe, "Analyzing HighDimensional Multisepctral Data," IEEE Transactions on Geoscience and Remote Sensing, vol. 31, pp. 792800, 1993.
Lillesand, T. M. and R. W. Kiefer, Remote Sensing and Image Interpretation, Fourth ed. New York: John Wiley & Sons;, 2000.
Mallat, S. and W. L. Hwang, "Singularity Detection and Processing with Wavelets," IEEE Transactions on Information Theory, vol. 38, pp. 617643, 1992.
Mallat, S., A Wavelet Tour of Signal Processing, Second Edition ed. San Diego: Academic Press, 1999.
Mallat, S. G., "A Theory for Multiresolution Signal Decomposition: The Wavelet Representation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 11, pp. 674693, 1989a.
Mallat, S. G., "Multiresolution Approximations and Wavelet Orthonormal Bases of L2(R)," Transactions of the American Mathematical Society, vol. 315, pp. 6987, 1989b.
Marsh, S. E. and J. B. McKeon, "Integrated Analysis of HighResolution Field and Airborne Spectroradiometer Data for Alteration Mapping," Economic Geology, vol. 78, pp. 618632, 1983.
Narendra, P. and K. Fukunaga, "A Branch and Bound Algorithm for Feature Selection," IEEE Transactions on Computers, vol. 26, pp. 917922, 1977.
Philpot, W. D., "The Derivative Ratio Algorithm: Avoiding Atmospheric Effects in Remote Sensing," IEEE Transactions on Geoscience and Remote Sensing, vol. 29, pp. 350357, 1991.
Piech, M. A. and K. R. Piech, "Symbolic Representation of Hyperspectral Data," Applied Optic, vol. 26, pp. 40184026, 1987.
Piech, M. A. and K. R. Piech, "Hyperspectral Interactions: Invariance and Scaling," Applied Optic, vol. 28, pp. 481489, 1989.
Pittner, S. and S. V. Kamarthi, "Feature Extraction from Wavelet Coefficients for Pattern Recognition Tasks," IEEE Transactions on Pattern Analysis and Machine Intelligency, vol. 21, pp. 8388, 1999.
Richards, J. A., Remote Sensing Digital Image Analysis: An Introduction, 2nd ed. New York: SpringerVerlag, 1993.
Roberts, D. A., Y. Yamaguchi, and J. P. Lyon, "Comparison of Various Techniques for Calibration of AIS Data," presented at Proceedings of the Second Airborne Imaging Spectrometer Data Analysis Workshop, Pasadena, CA, 1986.
Saito, N. and R. R. Coifman, "Local Discriminant Bases," presented at Proceeding of SPIE, 1994.
Schowengerdt, R. A., Remote Sensing, Models and Methods for Image Processing, Second ed. San Diego: Academic Press, 1997.
Scott, D. W., Multivariate Density Estimation. New York: Wiley, 1992.
Shaw, G. and D. Manolakis, "Signal Processing for Hyperspectral Image Exploitation," IEEE Signal Processing Magazine, vol. 19, pp. 1216, 2002.
Stone, C. J., "Optimal Rates of Convergence for Nonparametric Estimators," Annals of Statistics, vol. 8, pp. 13481360, 1980.
Strang, G. and T. Nguyen, Wavelets and Filter Banks. Wellesley, MA: WellesleyCambridge Press, 1996.
Swain, P. H. and S. M. Davis, Remote Sensing: The Quantitative Approach. New York: McGrawHill, 1978.
Tadjudin, S. and D. A. Landgrebe, "Classification of high dimensional data with limited training samples," in School of Electrical & Computer Engineering. West Lafayette: Purdue University, pp. 123, 1998
Tsai, F. and W. Philpot, "Derivative Analysis of Hyperspectral Data for Detecting Spectral Features," presented at 1997 IEEE International Geoscience and Remote Sensing Symposium, Singapore, 1997.
Tsai, F. and W. Philpot, "Derivative Analysis of Hyperspectral Data," in Remote Sensing Environment, vol. 66, pp. 4151, 1998
Vane, G., J. E. Duval, and J. B. Wellman, "Imaging Spectroscopy of the Earth and other Solar System Bodies," in Remote Geochemical Analysis: Elemental and Mineralogical Composition, C. M. Pieters and P. A. J. Englert, Eds. Cambridge: Cambridge University Press, 1993, pp. 121143.
Ward, M. O., "XmdvTool: Integrating Multiple Methods for Visualizing Multivariate Data," presented at Proceedings of IEEE Conference on Visualization '94, Washington, DC, 1994.
Witkin, A. P., "ScaleSpace Filtering," presented at 4th International Joint Conference on Artifical Intelligence, 1983.

論文全文使用權限 
同意授權校內瀏覽/列印電子全文服務，於20030828起公開。同意授權校外瀏覽/列印電子全文服務，於20030828起公開。 


