Hyperspectral Imaging : Techniques for Spectral Detection and Classification

個数:

Hyperspectral Imaging : Techniques for Spectral Detection and Classification

  • 提携先の海外書籍取次会社に在庫がございます。通常2週間で発送いたします。
    重要ご説明事項
    1. 納期遅延や、ご入手不能となる場合が若干ございます。
    2. 複数冊ご注文の場合、分割発送となる場合がございます。
    3. 美品のご指定は承りかねます。
  • 【重要:入荷遅延について】
    各国での新型コロナウィルス感染拡大により、洋書・洋古書の入荷が不安定になっています。
    弊社サイト内で表示している標準的な納期よりもお届けまでに日数がかかる見込みでございます。
    申し訳ございませんが、あらかじめご了承くださいますようお願い申し上げます。

  • 製本 Hardcover:ハードカバー版/ページ数 370 p.
  • 言語 ENG
  • 商品コード 9780306474835
  • DDC分類 621.3678

基本説明

An outgrowth of the research conducted over the years in the Remote Sensing Signal and Image Processing Laboratory (RSSIPL) at the University of Maryland, Baltimore County.

Full Description


Hyperspectral ImagingClassification is an outgrowth of the research conducted over the years in the Remote Sensing Signal and Image Processing Laboratory (RSSIPL) at the University of Maryland, Baltimore County. It explores applications of statistical signal processing to hyperspectral imaging and further develops non-literal (spectral) techniques for subpixel detection and mixed pixel classification. This text is the first of its kind on the topic and can be considered a recipe book offering various techniques for hyperspectral data exploitation. In particular, some known techniques, such as OSP (Orthogonal Subspace Projection) and CEM (Constrained Energy Minimization) that were previously developed in the RSSIPL, are discussed in great detail. This book is self-contained and can serve as a valuable and useful reference for researchers in academia and practitioners in government and industry.

Table of Contents

1 INTRODUCTION                                     1  (12)
1.1 BACKGROUND 2 (1)
1.2 OUTLINE OF THE BOOK 3 (10)
1.2.1 Stochastic Hyperspectral Measures 3 (1)
1.2.2 Subpixel Detection 4 (1)
1.2.3 Mixed Pixel classification (MPC) 5 (3)
1.2.3.1 Unconstrained MPC 7 (1)
1.2.3.2 Constrained MPC 7 (1)
1.2.3.3 Automatic Mixed Pixel 8 (1)
Classification (AMPC)
1.2.4 Hyperspectral Data to be Used in the 8 (2)
Book
1.2.5 Notations to Be Used in the Book 10 (3)
PART I: HYPERSPECTRAL MEASURES 13 (24)
2 HYPERSPECTRAL MEASURES FOR SPECTRAL 15 (22)
CHARACTERIZATION
2.1 MEASURES OF SPECTRAL VARIABILITY 15 (5)
2.1.1 Spectral Information Measure (SIM) 16 (1)
2.1.2 Hidden Markov Model(HMM)-Based 17 (3)
Measure
2.2 SPECTRAL SIMILARITY MEASURES 20 (3)
2.2.1 Commonly Used Measures 20 (1)
2.2.1.1 Distance-Based Measures 20 (1)
2.2.1.2 Orthogonal Projection-Based 20 (1)
Measures
2.2.2 Spectral Information Divergence 21 (2)
(SID)
2.2.3 Hidden Markov Model-Based 23 (1)
Information Divergence (HMMID)
2.3 MEASURES OF SPECTRAL DISCRIMINABILITY 23 (3)
2.3.1 Relative Spectral Discriminatory 24 (1)
ProBability (RSDPB)
2.3.2 Relative Spectral Discriminatory 24 (1)
PoWer (RSDPW)
2.3.3 Relative Spectral Discriminatory 25 (1)
Entropy (RSDS)
2.4 EXPERIMENTS 26 (8)
2.4.1 AVIRIS Data 26 (5)
2.4.2 HYDICE Data 31 (3)
2.5 CONCLUSIONS 34 (3)
PART II: SUBPIXEL DETECTION 37 (102)
3 TARGET ABUNDANCE-CONSTRAINED SUBPIXEL 39 (12)
DETECTION: PARTIALLY CONSTRAINED
LEAST-SQUARES METHODS
3.1 INTRODUCTION 39 (1)
3.2 LINEAR SPECTRAL MIXTURE MODEL 40 (1)
3.3 ORTHOGONAL SUBSPACE PROJECTION (OSP) 41 (3)
3.4 SUM-TO-ONE CONSTRAINED LEAST SQUARES 44 (1)
METHOD (SCLS)
3.5 NONNEGATIVITY CONSTRAINED LEAST SQUARES 45 (3)
METHOD (NCLS)
3.6 HYPERSPECTRAL IMAGE EXPERIMENTS 48 (2)
3.7 CONCLUSIONS 50 (1)
4 TARGET SIGNATURE-CONSTRAINED SUBPIXEL 51 (22)
DETECTION: LINEARLY CONSTRAINED MINIMUM
VARIANCE (LCMV)
4.1 INTRODUCTION 51 (2)
4.2 LCMV TARGET DETECTOR 53 (3)
4.2.1 Constrained Energy Minimization 54 (1)
(CEM)
4.2.2 Target-Constrained 55 (1)
Interference-Minimized Filter (TCIMF)
4.3 RELATIONSHIP AMONG OSP, CEM AND TCIMF 56 (2)
4.4 A COMPARARTIVE ANALYSIS BETWEEN CEM AND 58 (5)
TCIMF
4.4.1 Computer Simulations 58 (3)
4.4.2 Hyperspectral Image Experiments 61 (2)
4.5 SENSITIVITY OF CEM AND TCIMF TO LEVEL 63 (5)
OF TARGET INFORMATION
4.5.1 Computer Simulations 64
4.5.2 Hyperspectral Image Experiments 61 (7)
4.6 REAL-TIME PROCESSING 68 (3)
4.7 CONCLUSIONS 71 (2)
5 AUTOMATIC SUBPIXEL DETECTION: UNSUPERVISED 73 (16)
SUBPIXEL DETECTION
5.1 INTRODUCTION 73 (1)
5.2 UNSUPERVISED VECTOR QUANTIZATION 74 (1)
(UVQ)-BASED ALGORITHM
5.3 UNSUPERVISED TARGET GENERATION PROCESS 75 (3)
(UTGP)
5.4 UNSUPERVISED NCLS (UNCLS) ALGORITHM 78 (2)
5.5 EXPERIMENTS 80 (7)
5.6 CONCLUSIONS 87 (2)
6 AUTOMATIC SUBPIXEL DETECTION: ANOMALY 89 (16)
DETECTION
6.1 INTRODUCTION 89 (2)
6.2 RXD 91 (3)
6.3 LPTD AND UTD 94 (3)
6.4 RELATIONSHIP BETWEEN CEM AND RXD 97 (2)
6.5 REAL-TIME PROCESSING 99 (3)
6.6 CONCLUSIONS 102(3)
7 SENSITIVITY OF SUBPIXEL DETECTION 105(34)
7.1 INTRODUCTION 105(2)
7.2 SENSITIVITY OF TARGET KNOWLEDGE 107(4)
7.3 SENSITIVITY OF NOISE 111(18)
7.3.1 TSCSD 111(5)
7.3.2 Hyperspectral Image Experiments 116(28)
7.3.2.1 AVIRIS Data 116(9)
7.3.2.2 HYDICE Data 125(4)
7.4 SENSITIVITY OF ANOMALY DETECTION 129(8)
7.5 CONCLUSIONS 137(2)
PART III: UNCONSTRAINED MIXED PIXEL 139(40)
CLASSIFICATION
8 UNCONSTRAINED MIXED PIXEL CLASSIFICATION: 141(20)
LEASTSQUARES SUBSPACE PROJECTION
8.1 INTRODUCTION 141(3)
8.2 A POSTERIORI OSP 144(6)
8.2.1 Signature Subspace Projection (SSP) 144(2)
Classifier
8.2.2 Target Subspace Projection (TSP) 146(1)
Classifier
8.2.3 Oblique Subspace Projection (OBSP) 147(1)
Classifier
8.2.4 Unconstrained Maximum Likelihood 148(2)
Estimation Classifier
8.3 ESTIMATION ERROR EVALUATED BY ROC 150(3)
ANALYSIS
8.3.1 Signature Subspace Projection (SSP) 151(2)
Classifier
8.3.2 Oblique Subspace Projection (OBSP) 153(1)
Classifier
8.4 COMPUTER SIMULATIONS AND HYPERSPECTRAL 153(6)
IMAGE EXPERIMENTS
8.4.1 Computer Simulations 154(2)
8.4.2 Hyperspectral Data 156(3)
8.5 CONCLUSIONS 159(2)
9 A QUANTITATIVE ANALYSIS OF MIXED-TO-PURE 161(18)
PIXEL CONVERSION (MPCV)
9.1 INTRODUCTION 162(1)
9.2 CONVERSION OF MPC TO PPC 162(7)
9.2.1 Mixed-to-Pure Pixel Converter (MPCV) 163(1)
9.2.2 Minimum Distance-Based 164(2)
Classification
9.2.3 Fisher's Linear Discriminant 166(3)
Analysis (LDA)
9.2.4 Unsupervised Classification 169(1)
9.3 CRITERIA FOR TARGET DETECTION AND 169(2)
CLASSIFICATION
9.4 COMPARATIVE PERFORMANCE ANALYSIS 171(1)
9.5 CONCLUSIONS 171(8)
PART IV: CONSTRAINED MIXED PIXEL CLASSIFICATION 179(64)
10 TARGET ABUNDANCE-CONSTRAINED MIXED PXIEL 181(26)
CLASSIFICATION (TACMPC)
10.1 INTRODUCTION 181(2)
10.2 FULLY CONSTRAINED LEAST-SQUARES 183(1)
APPROACH
10.2.1 Fully Constrained Least-Squares 183(1)
Method (FCLS)
10.2.2 Unsupervised FCLS Method (UFCLS) 183(1)
10.3 MODIFIED FULLY CONSTRAINED 184(2)
LEAST-SQUARES (MFCLS) APPROACH
10.4 COMPUTER SIMULATIONS AND REAL 186(15)
HYPERSPECTRAL IMAGE EXPERIMENTS
10.4.1 Computer Simulations 186(2)
10.4.2 AVIRIS Image Experiments 188(5)
10.4.3 HYDICE Image Experiments 193(8)
10.5 NEAR REAL-TIME IMPLEMENTATION 201(4)
10.6 CONCLUSIONS 205(2)
11 TARGET SIGNATURE-CONSTRAINED MIXED PIXEL 207(22)
CLASSIFICATION (TSCMPC): LCMV CLASSIFIERS
11.1 INTRODUCTION 207(1)
11.2 LCMV CLASSWER 208(1)
11.3 BOWLES ET AL.'S FILTER VECTORS (FV) 209(2)
ALGORITHM
11.4 COLOR ASSIGNMENT OF LCMV CLASSIFIERS 211(2)
11.5 EXTENSION OF CEM FILTER TO CLASSIFIERS 213(1)
11.5.1 Winner-Take-All CEM (WTACEM) 213(1)
Classifier
11.5.2 Sum CEM (SCEM) Classifier 213(1)
11.5.3 Multiple-Target CEM (MTCEM) 213(1)
Classifier
11.5.4 Target-Constrained 214(1)
Interference-Minimized (TCIM) Classifier
11.6 COMPUTER SIMULATIONS 214(4)
11.7 HYPERSPECTRAL IMAGE EXPERIMENTS 218(5)
11.8 REAL-TIME IMPLEMENTATION FOR LCMV 223(4)
CLASSIFIERS
11.9 CONCLUSIONS 227(2)
12 TARGET SIGNATURE-CONSTRAINED MIXED PIXEL 229(14)
CLASSIFICATION (TSCMPC): LINEARLY CONSTRAINED
DISCRIMINANT ANALYSIS (LCDA)
12.1 INTRODUCTION 229(1)
12.2 LCDA 230(3)
12.3 WHITENING PROCESS FOR LCDA 233(1)
12.4 BOWLES ET AL.'S FILTER VECTORS (FV) 234(1)
ALGORITHM
12.5 COMPUTER SIMULATIONS リ HYPERSPECTRAL 235(5)
IMAGE EXPERIMENTS
12.6 CONCLUSIONS 240(3)
PART V: AUTOMATIC MIXED PIXEL CLASSIFICATION 243(106)
(AMPC)
13 AUTOMATIC MIXED PIXEL CLASSIFICATION 245(12)
(AMPC): UNSUPERVISED MIXED PIXEL
CLASSIFICATION
13.1 INTRODUCTION 245(1)
13.2 UNSUPERVISED MPC 246(1)
13.3 DESIRED TARGET DETECTION AND 246(7)
CLASSIFICATION
13.4 AUTOMATIC TARGET DETECTION AND 253(2)
CLASSIFICATION
13.5 CONCLUSIONS 255(2)
14 AUTOMATIC MIXED PIXEL CLASSIFICATION 257(20)
(AMPC): ANOMALY CLASSIFICATION
14.1 INTRODUCTION 257(1)
14.2 TARGET DISCRIMINATION MEASURES 258(2)
14.3 ANOMALY CLASSIFICATION 260(1)
14.4 AUTOMATIC THRESHOLDING METHOD 260(5)
14.5 ANALYSIS ON TARGET CORRELATION USING 265(5)
TARGET DISCRIMINATION MEASURES
14.6 ON-LINE IMPLEMENTATION 270(4)
14.7 CONCLUSIONS 274(3)
15 AUTOMATIC MIXED PIXEL CLASSIFICATION 277(28)
(AMPC): LINEAR SPECTRAL RANDOM MIXTURE
ANALYSIS (LSRMA)
15.1 INTRODUCTION 277(2)
15.2 INDEPENDENT COMPONENT ANALYSIS (ICA) 279(1)
15.3 ICA-BASED LSRMA 280(4)
15.3.1 Relative Entropy-Based Measure for 281(1)
ICA
15.3.2 Learning Algorithm to Find 282(2)
Separating Matrix W
15.4 EXPERIMENTS 284(11)
15.4.1 AVIRIS Image Experiments 284(5)
15.4.2 HYDICE Image Experiments 289(6)
15.5 3-D ROC ANALYSIS FOR LSRMA 295(7)
15.6 CONCLUSIONS 302(3)
16 AUTOMATIC MIXED PIXEL CLASSIFICATION 305(14)
(AMPC): PROJECTION PURSUIT
16.1 INTRODUCTION 305
16.2 PROJECTION PURSUIT 301(7)
16.3 EVOLUTIONARY ALGORITHM (EA) 308(2)
16.4 THRESHOLDING OF PROJECTION IMAGES 310(1)
USING ZERODETECTION
16.5 EXPERIMENTS 311(7)
16.5.1 AVIRIS Data Experiments 311(2)
16.5.2 HYDICE Data Experiments 313(5)
16.6 CONCLUSIONS 318(1)
17 ESTIMATION FOR VIRTUAL DIMENSIONALITY OF 319(16)
HYPERSPECTRAL IMAGERY
17.1 INTRODUCTION 319(2)
17.2 NEYMAN-PEARSON DETECTION THEORY-BASED 321(2)
EIGENTHRESHOLDING ANALYSIS (HFC METHOD)
17.3 ESTIMATION OF NOISE COVARIANCE MATRIX 323(3)
17.3.1 Residual Analysis (Roger, 1996) 323(2)
17.3.2 Inter/Intra-Band Prediction Noise 325(1)
Estimation: Spatial/Spectral Prediction
Noise Estimation (Roger and Arnold, 1996)
17.4 NOISE ESTIMATION-BASED 326(2)
EIGEN-THRESHOLDING
17.4.1 Noise-Whitened HFC (NWHFC) Method 326(1)
17.4.2 Noise Subspace Projection (NSP) 326(1)
17.4.3 AIC and MDL 327(1)
17.5 COMPUTER SIMULATIONS AND HYPERSPECTRAL 328(5)
IMAGE EXPERIMENTS
17.5.1 Computer Simulations 328(2)
17.5.2 AVIRIS and HYDICE Image Experiments 330(3)
17.6 CONCLUSIONS 333(2)
18 CONCLUSIONS AND FURTHER TECHNIQUES 335(14)
18.1 FUNCTIONAL TAXONOMY OF TECHNIQUES 335(2)
18.2 MATHEMATICAL TAXONOMY OF TECHNIQUES 337(2)
18.3 EXPERIMENTS 339(1)
18.4 ROC ANALYSIS FOR SUBPIXEL DETECTION 340(1)
AND MIXED PIXEL CLASSIFICATION
18.5 SENSITIVITY ISSUES 341(1)
18.5.1 Sensitivity to Level of Target 341(1)
Information
18.5.2 Sensitivity to Noise 341(1)
18.6 REAL-TIME IMPLEMENTATION 341(1)
18.7 FURTHER TECHNIQUES 342(5)
18.7.1 Generalized Orthogonal Subspace 342(1)
Projection
18.7.2 Convex Cone Analysis 343(1)
18.7.3 Kalman Filter-Based Linear Unmixing 344(1)
18.7.4 Interference-Annihilated 344(1)
Eigen-Analysis
18.7.5 Band Selection 345(1)
18.7.6 Linear Mixture Analysis-Based Data 346(1)
Compression
18.7.7 Radial Basis Function Neural 347(1)
Network Approach
18.8 APPLICATIONS TO MAGNETIC RESONANCE 347(2)
IMAGING
GLOSSARY 349(4)
REFERENCES 353(12)
INDEX 365