書 名：Hyperspectral Imaging
著者名：Chein-I Chang
出版社：Kluwer Academic/ Plenum Publisher
【書籍の内容】
ハイパースペクトルでなければ出来ない分析手法に関する理論書である。
具体的には、ピクセルの中に含まれるスペクトル的特徴の抽出、またはその逆で、周囲のスペクトル的特徴からかけ離れた“異物”のスペクトルを抽出する手法（RXD等）が紹介されている。
【目次】
１ Introduction
1.1 Background
1.2 Outline of the book
PART I: Hyperspectral measures
2 Hyperspectral measures for spectral characterization
2.1 Measures of spectral variability
2.2 Spectral similarity measures
2.3 Measures of spectral discriminability
2.4 Experiments
2.5 Conclusions
PART II: Subpixel Detection
3 Target abundance-constrained subpixel detection: partially constrained least-squares
3.1 Introduction
3.2 Linear spectral mixture model
3.3 Orthogonal subspace projection (OSP)
3.4 Sum-to-one constrained least squares method
3.5 Non negativity constrained least squares method
3.6 Hyperspectral image experiments
3.7 Conclusions
4 Target-constrained subpixel detection: linearly minimum variance (LCMV)
4.1 Introduction
4.2 LCMV target detector
※CEM: Constrained energy minimization
※TCIMF: Target-constrained interference-minimized filter
4.3 Relationship among OSP, CEM and TCIMF
4.4 A comparative analysis between CEM and TCIME
4.5 Sensitivity of CEM and TCIMF to level of target
4.6 Real-time processing
4.7 Conclusions
5 Automatic subpixel detection: Unsupervised subpixel detection
5.1 Introduction
5.2 Unsupervised vector quantization (UVQ)-based algorism
5.3 Unsupervised target generation process (UTGP)
5.4 Unsupervised NCLS (UNCLS) algorism
5.5 Experiments
5.6 Conclusions
6 Automatic subpixel detection: Anomaly detection
6.1 Introduction
6.2 RXD
6.3 LPTD and UTD
6.4 Relationship between CEM and RXD
6.5 Real-time processing
6.6 Conclusions
7 Sensitivity of subpixel detection
7.1 Introduction
7.2 Sensitivity of target knowledge
7.3 Sensitivity of noise
7.4 Sensitivity of anomaly detection
7.5 Conclusions
PART III: Unconstrained mixed pixel classification
8 Unconstrained mixed pixel classification: Least-squares subspace projection
8.1 Introduction
8.2 A posteriori OSP
8.3 Estimation error evaluated by ROC analysis
8.4 Computer simulations and hyperspectral image experiments
8.5 Conclusions
9 A quantitative analysis of mixed-to-pure pixel
9.1 Introduction
9.2 Conversion of MPC to PPC
9.3 Criteria for target detection and classification
9.4 Comparative performance analysis
9.5 Conclusions
PART IV: Constrained mixed pixel classification
10 Target abundance-constrained mixed pixel classification (TACMPC)
10.1 Introduction
10.2 Fully constrained least-squares approach
10.3 Modified fully constrained least-squares (MFCLS) approach
10.4 Computer simulations and real hyperspectral image experiments
10.5 Near real-time implementation
10.6 Conclusions
11 Target signature-constrained mixed pixel classification (TSCMPC): LCMV classifiers
11.1 Introduction
11.2 LCMV classifier
11.3 Bowles et al. ’S Filters (FM) Algorism
11.4 Color assignment of LCMV classifiers
11.5 Experiments of CEM (MTCEM) classifiers
11.6 Computer simulations
11.7 Hyperspectral image experiments
11.8 Real-time implementation for LCMV classifiers
11.9 Conclusions
12 Target signature-constrained mixed pixel classification (TSCMPC): Linearly constrained discriminant analysis (LCDA)
12.1 Introduction
12.2 LCDA
12.3 Whitening process for LCDA
12.4 Bowles et al. ‘S Filter Vectors (FV) algorism
12.5 Computer simulations and hyperspectral image experiments
12.6 Conclusions
PART V: Automatic mixed pixel classification (AMPC)
13 Automatic mixed pixel classification (AMPC)
13.1 Introduction
13.2 Unsupervised MPC
13.3 Desired target detection and classification
13.4 Automatic target detection and classification
13.5 Conclusions
14 Automatic mixed pixel classification (AMPC): Anomaly classification
14.1 Introduction
14.2 Target discrimination measures
14.3 Anomaly classification
14.4 Automatic thresholding using target
14.5 Analysis on target correlation using target discrimination measures
14.6 On-line implementation
14.7 Conclusions
15 Automatic mixed pixel classification (AMPC): Linear spectral random mixture analysis (LSRMA)
15.1 Introduction
15.2 Independent component analysis (ICA)
15.3 ICA-based LSRMA
15.4 Experiments
15.5 3-D Roc analysis for LSRMA
15.6 Conclusions
16 Automatic mixed pixel classification (AMPC): Projection pursuit
16.1 Introduction
16.2 Projection pursuit
16.3 Evolutionary algorism (EA)
16.4 Thresholding of projection images using zero-detection
16.5 Experiments
16.6 Conclusions
17 Estimation for virtual dimensionality of hyperspectral imagery
17.1 Introduction
17.2 Neyman-Pearson detection theory-based eigen-thresholding analysis (HFC Method)
17.3 Estimation of noise covariance matrix
17.4 Noise estimation-based eigen-thresholding
17.5 Computer simulations and hyperspectral image experiments
17.6 Conclusions
18 Conclusions and future techniques
18.1 Functional taxonomy of techniques
18.2 Mathematical taxonomy of techniques
18.3 Experiments
18.4 Roc analysis for subpixel detection and mixed pixel classification
18.5 Sensitivity issues
18.6 Real-time implementation
18.7 Further techniques
18.8 Applications to magnetic resonance imaging
GLOSSARY
REFERENCES
INDEX
参考：http://hokkaido-sat.jugem.jp/
