Digital Color Imaging (eBook)

Christine FERNANDEZ-MALOIGNE is Professor at Poitiers University, Manager of Xlim-SIC Laboratory, Chasseneuil, France. Frédérique ROBERT-INACIO is Researcher at IM2NP and lecturer at ISEN, Toulon, France. Ludovic MACAIRE is Full Professor at Lille 1 University,LAGIS Laboratory, Villeneuve d'Asq, France.

Foreword xi
Henri Maître

Chapter 1. Color Representation and Processing in Polar ColorSpaces 1
Jesús Angulo, Sébastien Lefèvre and OlivierLezoray

1.1. Introduction 1

1.2. The HSI triplet 3

1.3. Processing of hue: a variable on the unit circle 8

1.4. Color morphological filtering in the HSI space 15

1.5. Morphological color segmentation in the HSI space 24

1.6. Conclusion 35

1.7. Bibliography 36

Chapter 2. Adaptive Median Color Filtering 41
Frédérique Robert-Inacio and Eric Dinet

2.1. Introduction 41

2.2. Noise 42

2.3. Nonlinear filtering 47

2.4. Median filter: methods derived from vector methods 51

2.5. Adaptive filters 60

2.6. Performance comparison 66

2.7. Conclusion 71

2.8. Bibliography 72

Chapter 3. Anisotropic Diffusion PDEs for Regularization ofMultichannel Images: Formalisms and Applications 75
David Tschumperlé

3.1. Introduction 75

3.2. Preliminary concepts 80

3.3. Local geometry in multi-channel images 81

3.4. PDEs for multi-channel image smoothing: overview 87

3.5. Regularization and curvature preservation 102

3.6. Numerical implementation 109

3.7. Some applications 112

3.8. Conclusion 116

3.9. Bibliography 116

Chapter 4. Linear Prediction in Spaces with SeparateAchromatic and Chromatic Information 123
Olivier Alata, Imtnan Qazi, Jean-Christophe Burie and ChristineFernandez-Maloigne

4.1. Introduction 123

4.2. Complex vector 2D linear prediction 124

4.3. Spectral analysis in the IHLS and L*a*b* color spaces129

4.4. Application to segmentation of textured color images136

4.5. Conclusion 145

4.6. Bibliography 146

Chapter 5. Region Segmentation 149
Alain Clément, Laurent Busin, Olivier Lezoray and LudovicMacaire

5.1. Introduction 149

5.2. Compact histograms 150

5.3. Spatio-colorimetric classification 158

5.4. Segmentation by graph analysis 167

5.5. Evaluation of segmentation methods against a "groundtruth" 181

5.6. Conclusion 186

5.7. Bibliography 187

Chapter 6. Color Texture Attributes 193
Nicolas Vandenbroucke, Olivier Alata, Christèle Lecomte, AlicePorebski and Imtnan Qazi

6.1. Introduction 193

6.2. Statistical features 201

6.3. Spatio-frequential features 213

6.4. Stochastic modeling 217

6.5. Color texture classification 223

6.6. Conclusion 232

6.7. Bibliography 233

Chapter 7. Photometric Color Invariants for ObjectRecognition 241
Damien Muselet

7.1. Introduction 241

7.2. Basic assumptions 246

7.3. Color invariant characteristics 255

7.4. Conclusion 280

7.5. Bibliography 280

Chapter 8. Color Key Point Detectors and Local ColorDescriptors 285
Damien Muselet and Xiaohu Song

8.1. Introduction 285

8.2. Color key point and region detectors 286

8.3. Local color descriptors 299

8.4. Conclusion 308

8.5. Bibliography 309

Chapter 9. Motion Estimation in Color Image Sequences317
Bertrand Augereau and Jenny Benois-Pineau

9.1. Introduction 317

9.2. Extension of classical motion estimation techniques tocolor image spaces 318

9.3. Apparent motion and vector images 324

9.4. Conclusion 334

9.5. Bibliography 336

Appendix A. Appendix to Chapter 7: Summary of Hypotheses andColor Characteristic Invariances 339

A.1.Bibliography 344

List of Authors 345

Index 349