Color Constancy

MARC EBNER, Lecturer (Privatdozent), Universität Würzburg, Lehrstuhl für Informatik, Am Hubland, 97074 Würzburg, Germany MARC EBNER, is currently a lecturer at the Department of Computer Science, Programming Languages and Programming Methodology, University of Würzburg, Germany. He has been at the university since 1999, recently having completed his habilitation dissertation, on which this book is based. He teaches courses on computer graphics and virtual reality and his research interests are in colour constancy, computer vision, self-reproducing programs, neutral networks, and evolutionary algorithms. Previous to this post, he has gained qualifications from Stuttgart University, New York University and Tubingen University. To date, he has authored 8 published journal articles, 29 refereed conference papers.

"I think Ebner's Color Constancy is an excellent summary of current algorithms, and provides empirical tests in a womanlike manner … As an inclusive time capsule, it has no parallel, and therefore is a valuable contribution to the field." (Color Research and Application, June 2008)

• Series Preface xiPreface xiii1 Introduction 11.1 What is Color Constancy? 11.2 Classic Experiments 31.3 Overview 72 The Visual System 92.1 Eye and Retina 92.2 Visual Cortex 162.3 On the Function of the Color Opponent Cells 302.4 Lightness 312.5 Color Perception Correlates with Integrated Reflectances 322.6 Involvement of the Visual Cortex in Color Constancy 353 Theory of Color Image Formation 393.1 Analog Photography 413.2 Digital Photography 463.3 Theory of Radiometry 473.4 Reflectance Models 523.5 Illuminants 563.6 Sensor Response 603.7 Finite Set of Basis Functions 634 Color Reproduction 674.1 Additive and Subtractive Color Generation 684.2 Color Gamut 694.3 Computing Primary Intensities 694.4 CIE XYZ Color Space 704.5 Gamma Correction 794.6 Von Kries Coefficients and Sensor Sharpening 835 Color Spaces 875.1 RGB Color Space 875.2 sRGB 875.3 CIE L∗u∗v∗Color Space 895.4 CIE L∗a∗b∗Color Space 925.5 CMY Color Space 935.6 HSI Color Space 935.7 HSV Color Space 965.8 Analog and Digital Video Color Spaces 996 Algorithms for Color Constancy under Uniform Illumination 1036.1 White Patch Retinex 1046.2 The Gray World Assumption 1066.3 Variant of Horn’s Algorithm 1136.4 Gamut-constraint Methods 1156.5 Color in Perspective 1216.6 Color Cluster Rotation 1286.7 Comprehensive Color Normalization 1296.8 Color Constancy Using a Dichromatic Reflection Model 1347 Algorithms for Color Constancy under Nonuniform Illumination 1437.1 The Retinex Theory of Color Vision 1437.2 Computation of Lightness and Color 1547.3 Hardware Implementation of Land’s Retinex Theory 1667.4 Color Correction on Multiple Scales 1697.5 Homomorphic Filtering 1707.6 Intrinsic Images 1757.7 Reflectance Images from Image Sequences 1887.8 Additional Algorithms 1908 Learning Color Constancy 1938.1 Learning a Linear Filter 1938.2 Learning Color Constancy Using Neural Networks 1948.3 Evolving Color Constancy 1988.4 Analysis of Chromatic Signals 2048.5 Neural Architecture based on Double Opponent Cells 2058.6 Neural Architecture Using Energy Minimization 2099 Shadow Removal and Brightening 2139.1 Shadow Removal Using Intrinsic Images 2139.2 Shadow Brightening 21510 Estimating the Illuminant Locally 21910.1 Local Space Average Color 21910.2 Computing Local Space Average Color on a Grid of Processing Elements 22110.3 Implementation Using a Resistive Grid 23010.4 Experimental Results 23711 Using Local Space Average Color for Color Constancy 23911.1 Scaling Input Values 23911.2 Color Shifts 24111.3 Normalized Color Shifts 24611.4 Adjusting Saturation 24911.5 Combining White Patch Retinex and the Gray World Assumption 25112 Computing Anisotropic Local Space Average Color 25512.1 Nonlinear Change of the Illuminant 25512.2 The Line of Constant Illumination 25712.3 Interpolation Methods 25912.4 Evaluation of Interpolation Methods 26212.5 Curved Line of Constant Illumination 26512.6 Experimental Results 26713 Evaluation of Algorithms 27513.1 Histogram-based Object Recognition 27513.2 Object Recognition under Changing Illumination 27913.3 Evaluation on Object Recognition Tasks 28213.4 Computation of Color Constant Descriptors 29013.5 Comparison to Ground Truth Data 29914 Agreement with Data from Experimental Psychology 30314.1 Perceived Color of Gray Samples When Viewed under Colored Light 30314.2 Theoretical Analysis of Color Constancy Algorithms 30514.3 Theoretical Analysis of Algorithms Based on Local Space Average Color 31214.4 Performance of Algorithms on Simulated Stimuli 31614.5 Detailed Analysis of Color Shifts 31914.6 Theoretical Models for Color Conversion 32014.7 Human Color Constancy 32415 Conclusion 327Appendix A Dirac Delta Function 329Appendix B Units of Radiometry and Photometry 331Appendix C Sample Output from Algorithms 333Appendix D Image Sets 339Appendix E Program Code 349Appendix F Parameter Settings 363Bibliography 369List of Symbols 381Index 385Permissions 391