Mitchell Image Fusion

Theories, Techniques and Applications
2010
ISBN: 978-3-642-11216-4
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 247 Seiten

The purpose of this book is to provide a practical introduction to the th- ries, techniques and applications of image fusion. The present work has been designed as a textbook for a one-semester ?nal-year undergraduate, or ?r- year graduate, course in image fusion. It should also be useful to practising engineers who wish to learn the concepts of image fusion and apply them to practical applications. In addition, the book may also be used as a supp- mentary text for a graduate course on topics in advanced image processing. The book complements the author's previous work on multi-sensor data [1] fusion by concentrating exclusively on the theories, techniques and app- cations of image fusion. The book is intended to be self-contained in so far as the subject of image fusion is concerned, although some prior exposure to the ?eld of computer vision and image processing may be helpful to the reader. Apart from two preliminary chapters, the book is divided into three parts.

Autoren/Hrsg.

Mitchell, H. B.

Inhaltsverzeichnis

1;Preface;6
2;Contents;8
3;Introduction;16
3.1;Synergy;16
3.2;Image Fusion Process;17
3.3;Common Representational Block;19
3.4;Image Fusion Block;19
3.5;Image Fusion Algorithms;21
3.6;Organization;22
3.7;Software;22
3.8;Further Reading;23
3.9;References;23
4;Image Sensors;24
4.1; Digital Camera;24
4.2; Optical System;24
4.2.1;Perspective Projection;25
4.2.2;Orthographic Projection;25
4.3; Recording Systems;25
4.3.1;Noise ;26
4.4;Sampling;26
4.4.1;Quantization;27
4.4.2;Bayer;27
4.5;Spatial vs. Spectral Resolution;29
4.5.1;Spatial Resolution;29
4.5.2;Spectral Resolution;30
4.6;Further Reading;32
4.7;References;32
5;Part I Theories;33
5.1;Common Representational Format;34
5.1.1;Introduction;34
5.1.2;Geographical Information System ;36
5.1.3;Choosing a Common Representational Format;36
5.1.3.1;Human Fusion;36
5.1.3.2;Sparseness;37
5.1.3.3;Object Recognition;38
5.1.3.4;Uncertainty;40
5.1.4;Textures;41
5.1.5;Multi-scale Representation;42
5.1.6;Sub-space Methods;43
5.1.7;Multiple Training Sets;44
5.1.8;Software;45
5.1.9;Further Reading;45
5.1.10;References;46
5.2;Spatial Alignment;47
5.2.1;Introduction;47
5.2.2;Pairwise Transformation;48
5.2.2.1;Thin-Plate Splines;49
5.2.3;Hierarchical Registration ;51
5.2.4;Mosaic Image;53
5.2.4.1;Stitching;55
5.2.5;Image Similarity Measures;55
5.2.6;Mutual Information;56
5.2.6.1;Normalized Mutual Information;56
5.2.6.2;Calculation;57
5.2.6.3;Histogram;57
5.2.6.4;Parzen Windows;57
5.2.6.5;Iso-intensity Lines;58
5.2.7;Partial Volume Interpolation;59
5.2.8;Artifacts;60
5.2.9;Software;62
5.2.10;Further Reading;62
5.2.11;References;62
5.3;Semantic Equivalence;64
5.3.1;Introduction;64
5.3.2;Probabilistic Scale;65
5.3.2.1;Plat Calibration;66
5.3.2.2;Histogram Calibration;67
5.3.2.3;Isotonic Calibration;67
5.3.3;Decision Labels;68
5.3.3.1;Assignment Matrix;69
5.3.3.2;Co-association Matrix;71
5.3.4;Software;72
5.3.5;Further Reading;72
5.3.6;References;72
5.4;Radiometric Calibration;74
5.4.1;Introduction;74
5.4.2;Histogram Matching;75
5.4.2.1;Exact Histogram Specification;76
5.4.3; Midway Image Equalization;77
5.4.4;Matching Second-Order Statistics;79
5.4.5;Ranking;79
5.4.6;Thresholding;80
5.4.7;Segmentation;81
5.4.8;Feature Map Normalization;82
5.4.9;Probabilistic Scale;83
5.4.10;Software;83
5.4.11;Further Reading;84
5.4.12;References;84
5.5;Pixel Fusion;85
5.5.1;Introduction;85
5.5.2;Addition;85
5.5.2.1;Robust Averaging;87
5.5.3;Subtraction;87
5.5.4;Multiplication;90
5.5.5;Division;90
5.5.6;Feature Map Fusion;91
5.5.7;Decision Fusion;93
5.5.7.1;Shape-Based Averaging;95
5.5.7.2;Similarity;97
5.5.7.3;Label Permutation;97
5.5.7.4;Co-associative Matrix;98
5.5.8;Software;99
5.5.9;References;99
6;Part II Techniques;101
6.1;Multi-resolution Analysis;102
6.1.1;Introduction;102
6.1.2;Discrete Wavelet Transform;103
6.1.3; Undecimated Discrete Wavelet Transform (UDWT);105
6.1.4;Wavelet Fusion ;107
6.1.5;Expectation-Maximization Algorithm;110
6.1.6;Multi-modal Wavelet Fusion;110
6.1.7;Pan-Sharpening;113
6.1.8;Software;114
6.1.9;Further Reading;114
6.1.10;References;114
6.2;Image Sub-space Techniques;116
6.2.1;Introduction;116
6.2.2;Principal Component Analysis (PCA);118
6.2.2.1;PCA Variants;121
6.2.2.2;Whitening;121
6.2.2.3;Two-Dimensional PCA;122
6.2.3;PCA Fusion;123
6.2.4;Non-negative Matrix Factorization (NMF);124
6.2.5;Linear Discriminant Analysis (LDA);125
6.2.5.1;Fisherface;126
6.2.5.2;Median LDA;127
6.2.5.3;Re-weighting LDA;127
6.2.5.4;Two-Dimensional LDA;128
6.2.6;Nearest Neighbor Discriminant Analysis (NNDA);129
6.2.6.1;K-Nearest Neighbor Discriminant Analysis;130
6.2.6.2;Two-Dimensional NNDA;130
6.2.7;Canonical Correlation Analysis (CCA);130
6.2.8;Software;131
6.2.9;Further Reading;131
6.2.10;References;131
6.3;Ensemble Learning;134
6.3.1;Ensemble Learning Methods;134
6.3.2;Diversity Measures;135
6.3.3;Multiple Image Transformations Ik;137
6.3.3.1;Multiple Subspace Transformations;138
6.3.3.2;Multiple Random Convolutions;138
6.3.3.3;Multiple Normalizations;139
6.3.3.4;Multiple Color Spaces;140
6.3.3.5;Multiple Thresholds;140
6.3.3.6;Multiple Segmentations;141
6.3.4;Re-sampling Methods;142
6.3.5;Image Fusion;142
6.3.6;Ensemble Thresholding;144
6.3.7;Ensemble Spatial Sampling;146
6.3.8;Ensemble Atlas Based Segmentation;148
6.3.9;Ensemble Nearest Neighbor Classification;149
6.3.10;Further Reading;150
6.3.11;Software;150
6.3.12;References;150
6.4;Re-sampling Methods;152
6.4.1;Introduction;152
6.4.2;Bootstrapping;152
6.4.3;Face Recognition with Bagging;153
6.4.4;Bagged Nearest Neighbor Classifier;153
6.4.5;Bagged K-means Clustering;154
6.4.6;Boosting;156
6.4.7;Viola-Jones Algorithm;158
6.4.8;Boosted Object Detection;158
6.4.9;Software;161
6.4.10;Further Reading;162
6.4.11;References;162
6.5;Image Thresholding;163
6.5.1;Global Thresholding;163
6.5.2;Statistical Algorithms;164
6.5.2.1;Ridler-Calvard;166
6.5.2.2;Otsu;166
6.5.2.3;Kittler-Illingworth;166
6.5.2.4;Kapur;167
6.5.2.5;Tsai;167
6.5.3;Local Thresholding;168
6.5.4;Software;168
6.5.5;Further Reading;168
6.5.6;References;169
6.6;Image Key Points;170
6.6.1;Scale-Invariant Feature Transform;170
6.6.1.1;Hyperspectral Images;171
6.6.2;Speeded-Up Robust Feature;172
6.6.3;Complex Wavelet Transform;172
6.6.4;Software;173
6.6.5;References;173
6.7;Image Similarity Measures;174
6.7.1;Introduction;174
6.7.2;Global Similarity Measures without Spatial Alignment;177
6.7.2.1;Probabilistic Similarity Measures;177
6.7.2.2;2 Distance Measure;179
6.7.2.3;Cross-Bin Distance Measures;181
6.7.3;Global Similarity Measures with Spatial Alignment;183
6.7.3.1;Mean Square Error and Mean Absolute Error ;183
6.7.3.2;Cross-Correlation Coefficient;184
6.7.3.3;Mutual Information;185
6.7.3.4;Ordinal Global Similarity Measures;185
6.7.4;Local Similarity Measures;187
6.7.4.1;Bhat-Nayar Distance Measure;187
6.7.4.2;Mittal-Ramesh Ordinal Measure;189
6.7.5;Binary Image Similarity Measure;189
6.7.5.1;Hausdorff Metric;190
6.7.6;Software ;191
6.7.7;Further Reading;191
6.7.8;References;191
6.8;Vignetting, White Balancing and Automatic Gain Control Effects;193
6.8.1;Introduction;193
6.8.2;Vignetting;194
6.8.2.1;Vignetting Correction;194
6.8.3;Radiometric Response Function;195
6.8.3.1;Automatic Gain Control ;195
6.8.4;White Balancing;197
6.8.5;Ensemble White Balancing;198
6.8.6;References;198
6.9;Color Image Spaces;200
6.9.1;Introduction;200
6.9.2;Perceptual Color Models;202
6.9.2.1;IHS;202
6.9.2.2;HSV;203
6.9.2.3;HLS;204
6.9.2.4;IHLS;205
6.9.2.5;Indirect IHS Transformation;205
6.9.2.6;Circular Statistics;206
6.9.3;Multiple Color Spaces;207
6.9.4;Software;208
6.9.5;Further Reading;208
6.9.6;References;209
6.10;Markov Random Fields;210
6.10.1;Markov Random Fields;210
6.10.2;Energy Function;212
6.10.3;Algorithm;213
6.10.4;Further Reading;214
6.10.5;References;214
6.11;Image Quality;215
6.11.1;Introduction;215
6.11.2;Reference-Based Quality Measures;215
6.11.3;Non-reference Based Quality Measures;216
6.11.4;Analysis;218
6.11.5;Software;218
6.11.6;Further Reading;218
6.11.7;References;219
7;Part III Applications;220
7.1;Pan-sharpening;221
7.1.1;Introduction;221
7.1.2;IHS Pan-sharpening;222
7.1.3;Spectral Distortion;224
7.1.3.1;Pan-sharpening Algorithm of Choi;225
7.1.3.2; Pan-sharpening Algorithm of Tu et al.;226
7.1.4;IKONOS;226
7.1.5;Wavelets;227
7.1.6;Sensor Spectral Response;228
7.1.7;References;229
7.2;Ensemble Color Image Segmentation;230
7.2.1;Introduction;230
7.2.2;Image Ensemble;231
7.2.3;K-Means Segmentation;231
7.2.4;K-Means Fusion Operator;232
7.2.5;Reference;233
7.3;STAPLE: Simultaneous Truth and Performance Level Estimation;234
7.3.1;Introduction;234
7.3.2;Expectation-Maximization Algorithm;234
7.3.3;STAPLE;235
7.3.4;References;237
7.4;Biometric Technologies;238
7.4.1; Introduction;238
7.4.2; Multi-modal Biometrics;239
7.4.2.1;Fingerprints;239
7.4.2.2;Signatures;240
7.4.2.3; Faces;240
7.4.2.4; Iris and Retina;240
7.4.2.5; Gait Biometrics;240
7.4.2.6; Other Biometrics;240
7.4.3;Multi-biometrics;240
7.4.3.1;Multi-sensor System;241
7.4.3.2;Multi-algorithm System;241
7.4.3.3;Multi-instance System;242
7.4.3.4;Multi-sample System;242
7.4.4;Epilogue;242
7.4.5;References;243
8;Index;244