The essential guide to image processing / 2nd ed.

Front Cover 1
The Essential Guide to Image Processing 4
Copyright Page 5
Table of Contents 6
Preface 20
About the Author 22
Chapter 1. Introduction to Digital Image Processing 24
1.1 Types of Images 25
1.2 Scale of Images 27
1.3 Dimension of Images 29
1.4 Digitization of Images 29
1.5 Sampled Images 30
1.6 Quantized Images 32
1.7 Color Images 36
1.8 Size of Image Data 38
1.9 Objectives of this Guide 40
1.10 Organization of the Guide 42
Reference 44
Chapter 2. The SIVA Image Processing Demos 46
2.1 Introduction 46
2.2 LabVIEW for Image Processing 47
2.2.1 The LabVIEW Development Environment 48
2.2.2 Image Processing and Machine Vision in LabVIEW 50
2.3 Examples from the SIVA Image Processing Demos 54
2.4 Conclusions 61
References 64
Chapter 3. Basic Gray Level Image Processing 66
3.1 Introduction 66
3.2 Notation 67
3.3 Image Histogram 67
3.4 Linear Point Operations on Images 70
3.4.1 Additive Image Offset 71
3.4.2 Multiplicative Image Scaling 73
3.4.3 Image Negative 74
3.4.4 Full-Scale Histogram Stretch 76
3.5 Nonlinear Point Operations on Images 78
3.5.1 Logarithmic Point Operations 78
3.5.2 Histogram Equalization 79
3.5.3 Histogram Shaping 82
3.6 Arithmetic Operations Between Images 83
3.6.1 Image Averaging for Noise Reduction 84
3.6.2 Image Differencing for Change Detection 86
3.7 Geometric Image Operations 88
3.7.1 Nearest Neighbor Interpolation 88
3.7.2 Bilinear Interpolation 89
3.7.3 Image Translation 89
3.7.4 Image Rotation 90
3.7.5 Image Zoom 90
Chapter 4. Basic Binary Image Processing 92
4.1 Introduction 92
4.2 Image Thresholding 94
4.3 Region Labeling 100
4.3.1 Region Labeling Algorithm 100
4.3.2 Region Counting Algorithm 101
4.3.3 Minor Region Removal Algorithm 101
4.4 Binary Image Morphology 102
4.4.1 Logical Operations 102
4.4.2 Windows 103
4.4.3 Morphological Filters 105
4.4.4 Morphological Boundary Detection 113
4.5 Binary Image Representation and Compression 115
4.5.1 Run-Length Coding 116
4.5.2 Chain Coding 117
Chapter 5. Basic Tools for Image Fourier Analysis 120
5.1 Introduction 120
5.2 Discrete-Space Sinusoids 120
5.3 Discrete-Space Fourier Transform 123
5.3.1 Linearity of DSFT 124
5.3.2 Inversion of DSFT 124
5.3.3 Magnitude and Phase of DSFT 124
5.3.4 Symmetry of DSFT 125
5.3.5 Translation of DSFT 125
5.3.6 Convolution and the DSFT 126
5.4 2D Discrete Fourier Transform (DFT) 126
5.4.1 Linearity and Invertibility of DFT 128
5.4.2 Symmetry of DFT 129
5.4.3 Periodicity of DFT 129
5.4.4 Image Periodicity Implied by DFT 129
5.4.5 Cyclic Convolution Property of the DFT 130
5.4.6 Linear Convolution Using the DFT 133
5.4.7 Computation of the DFT 135
5.4.8 Displaying the DFT 135
5.5 Understanding Image Frequencies and the DFT 138
5.5.1 Frequency Granularity 138
5.5.2 Frequency Orientation 141
5.6 Related Topics in this Guide 144
Chapter 6. Multiscale Image Decompositions and Wavelets 146
6.1 Overview 146
6.2 Pyramid Representations 149
6.2.1 Decimation and Interpolation 149
6.2.2 Gaussian Pyramid 150
6.2.3 Laplacian Pyramid 151
6.3 Wavelet Representations 152
6.3.1 Filter Banks 152
6.3.2 Wavelet Decomposition 153
6.3.3 Discrete Wavelet Bases 156
6.3.4 Continuous Wavelet Bases 158
6.3.5 More on Wavelet Image Representations 159
6.3.6 Relation to Human Visual System 160
6.3.7 Applications 161
6.4 Other Multiscale Decompositions 161
6.4.1 Undecimated Wavelet Transform 161
6.4.2 Wavelet Packets 161
6.4.3 Geometric Wavelets 162
6.5 Conclusion 163
References 163
Chapter 7. Image Noise Models 166
7.1 Summary 166
7.2 Preliminaries 166
7.2.1 What is Noise? 166
7.2.2 Notions of Probability 167
7.3 Elements of Estimation Theory 169
7.4 Types of Noise and Where They Might Occur 172
7.4.1 Gaussian Noise 172
7.4.2 Heavy Tailed Noise 173
7.4.3 Salt and Pepper Noise 177
7.4.4 Quantization and Uniform Noise 178
7.4.5 Photon Counting Noise 179
7.4.6 Photographic Grain Noise 182
7.5 CCD Imaging 183
7.6 Speckle 184
7.6.1 Speckle in Coherent Light Imaging 184
7.6.2 Atmospheric Speckle 188
7.7 Conclusions 189
References 190
Chapter 8. Color and Multispectral Image Representation and Display 192
8.1 Introduction 192
8.2 Preliminary Notes on Display of Images 193
8.3 Notation and Prerequisite Knowledge 197
8.3.1 Practical Sampling 197
8.3.2 One-Dimensional Discrete System Representation 198
8.3.3 Multidimensional System Representation 200
8.4 Analog Images as Physical Functions 202
8.5 Colorimetry 203
8.5.1 Color Sampling 205
8.5.2 Discrete Representation of Color-Matching 206
8.5.3 Properties of Color-Matching Functions 207
8.5.4 Notes on Sampling for Color Aliasing 212
8.5.5 A Note on the Nonlinearity of the Eye 214
8.5.6 Uniform Color Spaces 214
8.6 Sampling of Color Signals and Sensors 216
8.7 Color I/O Device Calibration 219
8.7.1 Calibration Definitions and Terminology 219
8.7.2 CRT Calibration 220
8.7.3 Scanners and Cameras 221
8.7.4 Printers 222
8.7.5 Calibration Example 223
8.8 Summary and Future Outlook 225
References 225
Chapter 9. Capturing Visual Image Properties with Probabilistic Models 228
9.1 The Gaussian Model 230
9.2 The Wavelet Marginal Model 234
9.3 Wavelet Local Contextual Models 238
9.4 Discussion 243
References 243
Chapter 10. Basic Linear Filtering with Application to Image Enhancement 248
10.1 Introduction 248
10.2 Impulse Response, Linear Convolution, and Frequency Response 250
10.3 Linear Image Enhancement 253
10.3.1 Moving Average Filter 254
10.3.2 Ideal Lowpass Filter 256
10.3.3 Gaussian Filter 259
10.4 Discussion 262
References 262
Chapter 11. Multiscale Denoising of Photographic Images 264
11.1 Introduction 264
11.2 Distinguishing Images from Noise in Multiscale Representations 265
11.3 Subband Denoising\u2014A Global Approach 267
11.3.1 Band Thresholding 267
11.3.2 Band Weighting 270
11.4 Subband Coefficient Denoising\u2014A Pointwise Approach 272
11.4.1 Coefficient Thresholding 273
11.4.2 Coefficient Weighting 275
11.5 Subband Neighborhood Denoising\u2014Striking a Balance 276
11.5.1 Neighborhood Thresholding 276
11.5.2 Neighborhood Weighting 279
11.6 Statistical Modeling for Optimal Denoising 280
11.6.1 The Bayesian View 281
11.6.2 Empirical Bayesian Methods 281
11.7 Conclusions 282
References 283
Chapter 12. Nonlinear Filtering for Image Analysis and Enhancement 286
12.1 Introduction 286
12.2 Weighted Median Smoothers and Filters 287
12.2.1 Running Median Smoothers 287
12.2.2 Weighted Median Smoothers 290
12.2.3 Weighted Median Filters 298
12.3 Image Noise Cleaning 300
12.4 Image Zooming 305
12.5 Image Sharpening 307
12.6 Conclusion 312
References 313
Chapter 13. Morphological Filtering 316
13.1 Introduction 316
13.2 Morphological Image Operators 317
13.2.1 Morphological Filters for Binary Images 317
13.2.2 Morphological Filters for Gray-level Images 318
13.2.3 Universality of Morphological Operators 319
13.2.4 Median, Rank, and Stack Filters 320
13.2.5 Algebraic Generalizations of Morphological Operators 321
13.3 Morphological Filters for Image Enhancement 322
13.3.1 Noise Suppresion and Image Smoothing 323
13.3.2 Connected Filters for Smoothing and Simplification 324
13.3.3 Contrast Enhancement 328
13.4 Morphological Operators for Template Matching 330
13.4.1 Morphological Correlation 330
13.4.2 Binary Object Detection and Rank Filtering 331
13.4.3 Hit-Miss Filter 332
13.5 Morphological Operators for Feature Detection 332
13.5.1 Edge Detection 332
13.5.2 Peak/Valley Blob Detection 337
13.6 Design Approaches for Morphological Filters 340
13.7 Conclusions 342
References 342
Chapter 14. Basic Methods for Image Restoration and Identification 346
14.1 Introduction 346
14.2 Blur Models 349
14.2.1 No Blur 350
14.2.2 Linear Motion Blur 350
14.2.3 Uniform Out-of-Focus Blur 352
14.2.4 Atmospheric Turbulence Blur 353
14.3 Image Restoration Algorithms 353
14.3.1 Inverse Filter 355
14.3.2 Least-Squares Filters 355
14.3.3 Iterative Filters 361
14.3.4 Boundary Value Problem 365
14.4 Blur Identification Algorithms 366
14.4.1 Spectral Blur Estimation 367
14.4.2 Maximum Likelihood Blur Estimation 368
References 370
Chapter 15. Iterative Image Restoration 372
15.1 Introduction 372
15.2 Iterative Recovery Algorithms 373
15.3 Spatially Invariant Degradation 374
15.3.1 Degradation Model 374
15.3.2 Basic Iterative Restoration Algorithm 374
15.3.3 Convergence 375
15.3.4 Reblurring 377
15.3.5 Experimental Results 378
15.4 Matrix-Vector Formulation 384
15.4.1 Basic Iteration 385
15.4.2 Least-Squares Iteration 385
15.4.3 Constrained Least-Squares Iteration 386
15.4.4 Spatially Adaptive Iteration 390
15.5 Use of Constraints 391
15.5.1 Experimental Results 393
15.6 Additional Considerations 394
15.6.1 Other Forms of the Iterative Algorithm 394
15.6.2 Hierarchical Bayesian Image Restoration 395
15.6.3 Blind Deconvolution 397
15.6.4 Additional Applications 398
15.7 Discussion 404
References 404
Chapter 16. Lossless Image Compression 408
16.1 Introduction 408
16.2 Basics of Lossless Image Coding 409
16.3 Lossless Symbol Coding 413
16.3.1 Basic Concepts from Information Theory 414
16.3.2 Context-Based Entropy Coding 417
16.3.3 Huffman Coding 418
16.3.4 Arithmetic Coding 422
16.3.5 Lempel-Ziv Coding 427
16.3.6 Elias and Exponential-Golomb Codes 429
16.4 Lossless Coding Standards 433
16.4.1 The JBIG and JBIG2 Standards 433
16.4.2 The Lossless JPEG Standard 434
16.4.3 The JPEG2000 Standard 435
16.5 Other Developments in Lossless Coding 436
16.5.1 CALIC 436
16.5.2 Perceptually Lossless Image Coding 438
References 440
Chapter 17. JPEG and JPEG2000 444
17.1 Introduction 444
17.2 Lossy JPEG Codec Structure 446
17.2.1 Encoder Structure 446
17.2.2 Decoder Structure 448
17.3 Discrete Cosine Transform 448
17.4 Quantization 449
17.4.1 DCT Coefficient Quantization Procedure 449
17.4.2 Quantization Table Design 452
17.5 Coefficient-to-Symbol Mapping and Coding 455
17.5.1 DC Coefficient Symbols 455
17.5.2 Mapping AC Coefficient to Symbols 456
17.5.3 Entropy Coding 457
17.6 Image Data Format and Components 458
17.7 Alternative Modes of Operation 459
17.7.1 Progressive Mode 460
17.7.2 Hierarchical Mode 462
17.8 JPEG Part 3 464
17.8.1 Variable Quantization 464
17.8.2 Tiling 466
17.9 The JPEG2000 Standard 468
17.10 JPEG2000 Part 1: Coding Architecture 469
17.10.1 Preprocessing: Tiling, Level Offset, and Color Transforms 470
17.10.2 Discrete Wavelet Transform (DWT) 471
17.10.3 Quantization and Inverse Quantization 472
17.10.4 Precincts and Code-blocks 473
17.10.5 Entropy Coding 473
17.10.6 Bitstream Organization 477
17.10.7 Additional Features 478
17.11 Performance and Extensions 480
17.11.1 Comparison of Performance 480
17.11.2 Part 2 Extensions 480
17.12 Additional Information 481
17.12.1 Useful Information and Links for the JPEG Standard 481
17.12.2 Useful Information and Links for the JPEG2000 Standard 481
References 482
Chapter 18. Wavelet Image Compression 486
18.1 What Are Wavelets: Why Are They Good for Image Coding? 486
18.2 The Compression Problem 492
18.3 The Transform Coding Paradigm 494
18.3.1 Transform Structure 497
18.3.2 Quantization 497
18.3.3 Entropy Coding 498
18.4 Subband Coding: The Early Days 499
18.5 New and More Efficient Class of Wavelet Coders 501
18.5.1 Zerotree-Based Framework and EZW Coding 501
18.5.2 Advanced Wavelet Coders: High-Level Characterization 508
18.6 Adaptive Wavelet Transforms: Wavelet Packets 509
18.7 JPEG2000 and Related Developments 513
18.8 Conclusion 514
References 515
Chapter 19. Gradient and Laplacian Edge Detection 518
19.1 Introduction 518
19.2 Gradient-Based Methods 521
19.2.1 Continuous Gradient 521
19.2.2 Discrete Gradient Operators 527
19.3 Laplacian-Based Methods 532
19.3.1 Continuous Laplacian 532
19.3.2 Discrete Laplacian Operators 533
19.3.3 The Laplacian of Gaussian (Marr-Hildreth Operator) 535
19.3.4 Difference of Gaussian 538
19.4 Canny\u2019s Method 539
19.5 Approaches for Color and Multispectral Images 541
19.6 Summary 545
References 545
Chapter 20. Diffusion Partial Differential Equations for Edge Detection 548
20.1 Introduction and Motivation 548
20.1.1 Partial Differential Equations in Image and Video Processing 548
20.1.2 Edges and Anisotropic Diffusion 548
20.2 Background on Diffusion 549
20.2.1 Scale Space and Isotropic Diffusion 549
20.3 Anisotropic Diffusion Techniques 551
20.3.1 The Diffusion Coefficient 551
20.3.2 The Diffusion PDE 556
20.3.3 Variational Formulation 558
20.3.4 Multiresolution Diffusion 560
20.3.5 Multispectral Anisotropic Diffusion 561
20.3.6 Speckle Reducing Anisotropic Diffusion 562
20.4 Application of Anisotropic Diffusion to Edge Detection 565
20.4.1 Edge Detection by Thresholding 565
20.4.2 Edge Detection from Image Features 566
20.4.3 Quantitative Evaluation of Edge Detection by Anisotropic Diffusion 566
20.5 Using Vector Diffusion and Parametric Active Contours to Locate Edges 569
20.5.1 Parametric Active Contours 569
20.5.2 Gradient Vector Flow 570
20.5.3 Motion Gradient Vector Flow 571
20.6 Conclusions 572
References 573
Chapter 21. Image Quality Assessment 576
21.1 Introduction 576
21.2 Human Vision Modeling Based Metrics 578
21.2.1 Building Blocks 579
21.2.2 HVS-Based Models 584
21.3 Structural Approaches 594
21.3.1 The Structural Similarity Index 594
21.3.2 Image Quality Assessment Using SSIM 596
21.3.3 Relation to HVS-Based Models 599
21.4 Information Theoretic Approaches 601
21.4.1 Information Theoretic Metrics 601
21.4.2 Image Quality Assessment Using Information Theoretic Metrics 606
21.4.3 Relation to HVS-Based Metrics and Structural Similarity 608
21.5 Performance of Image Quality Metrics 610
21.6 Conclusion 611
References 612
Chapter 22. Image Watermarking: Techniques and Applications 620
22.1 Introduction 620
22.2 Applications of Watermarking Techniques 621
22.3 Classification of Watermarking Algorithms 624
22.4 Watermark Embedding, Detection, and Decoding 627
22.5 Copyright Protection Watermarking 629
22.5.1 Requirements and Metrics 629
22.5.2 Attacks Against Copyright Protection Watermarking Systems 634
22.5.3 Benchmarking of Copyright Protection Image Watermarking Algorithms 636
22.5.4 Spread Spectrum Watermarking 638
22.5.5 Watermarking with Side Information 650
22.6 Image Content Integrity and Authentication Watermarking 659
References 664
Chapter 23. Fingerprint Recognition 672
23.1 Introduction 672
23.2 Emerging Applications 672
23.3 Fingerprint as a Biometric 673
23.4 History of Fingerprints 674
23.5 System Architecture 675
23.6 Fingerprint Sensing 676
23.7 Fingerprint Features 678
23.8 Feature Extraction 680
23.9 Fingerprint Enhancement 683
23.10 Fingerprint Classification 687
23.11 Fingerprint Matching 691
23.12 Summary and Future Prospects 696
References 697
Chapter 24. Unconstrained Face Recognition from a Single Image 700
24.1 Introduction 700
24.1.1 Biometric Perspective 700
24.1.2 Experimental Perspective 701
24.1.3 Theoretical Perspective 702
24.1.4 Unconstrained Face Recognition 707
24.2 Linear Lambertian Object: Face Recognition Under Illumination Variation 707
24.2.1 Linear Lambertian Objects 707
24.2.2 Literature Review and Proposed Approach 709
24.2.3 The Importance of the Attached Shadow 710
24.2.4 Recognition in the Presence of a Single Light Source 711
24.2.5 Recognition in the Presence of Multiple Light Sources 714
24.3 Illuminating Light Field: Face Recognition Under Illumination and Pose Variations 716
24.3.1 Literature Review 716
24.3.2 Illumination- and Pose-Invariant Identity Signature 717
24.3.3 Implementations and Experiments 720
24.4 Face Modeling and Verification Across Age Progression 722
24.4.1 Shape Transformation Model for Young Individuals [60] 725
24.4.2 Shape Transformation Model for Adults [61] 727
24.4.3 Texture Transformation Model 729
24.5 Conclusions 731
References 732
Chapter 25. How Iris Recognition Works 738
25.1 Introduction 738
25.2 Localizing the Iris and its Boundaries 739
25.3 Iris Feature Encoding by 2D Gabor Wavelet Demodulation 743
25.4 The Test of Statistical Independence: Combinatorics of Phase Sequences 745
25.5 Recognizing Irises Regardless of Size, Position, and Orientation 749
25.6 Uniqueness of Failing the Test of Statistical Independence 751
25.7 Decision Environment for Iris Recognition 754
25.8 Speed Performance Summary 757
25.9 Appendix: 2D Focus Assessment at the Video Frame Rate 758
References 761
Chapter 26. Computed Tomography 764
26.1 Introduction 764
26.2 Background 765
26.2.1 X-ray Computed Tomography 765
26.2.2 Nuclear Imaging Using PET and SPECT 767
26.2.3 Mathematical Preliminaries 770
26.2.4 Examples 770
26.3 2D Image Reconstruction 771
26.3.1 Fourier Space and Filtered Backprojection Methods for Parallel-Beam Projections 771
26.3.2 Fan-Beam Filtered Backprojection 774
26.3.3 Region of Interest Reconstruction 776
26.4 Extending 2D Methods into 3D 778
26.4.1 Extracting 2D Data from 3D 778
26.4.2 Spiral CT 779
26.4.3 Rebinning Methods in 3D PET 781
26.5 3D Image Reconstruction 782
26.5.1 Fully 3D Reconstruction with Missing Data 782
26.5.2 Cone-Beam Tomography 784
26.5.3 Helical Multi-Slice CT Imaging 787
26.6 Iterative Reconstruction Methods 790
26.6.1 Finite Dimensional Formulations and Algebraic Reconstruction Technique 790
26.6.2 Statistical Formulations 792
26.6.3 Maximum Likelihood Methods 793
26.6.4 Bayesian Reconstruction Methods 795
26.7 Summary 796
References 797
Chapter 27. Computer-Assisted Microscopy 800
27.1 Introduction 800
27.2 Computer-Assisted Microscopy Systems 802
27.2.1 Hardware 803
27.2.2 Image Capture 808
27.2.3 Imaging Software 808
27.3 Software for Hardware Control 809
27.3.1 Automated Slide Scanning 810
27.3.2 Autofocusing 810
27.3.3 Image Capture 813
27.4 Image Processing and Analysis Software 814
27.4.1 Correction of Instrumentation-based Errors 814
27.4.2 Background Shading Correction 814
27.4.3 Color Compensation 817
27.4.4 Image Enhancement 818
27.4.5 Segmentation for Object Identification 819
27.4.6 Object Measurement 821
27.4.7 The User Interface 822
27.5 A Computerized Microscopy System for Clinical Cytogenetics 822
27.5.1 Hardware 822
27.5.2 Software 823
27.6 Applications in Clinical Cytogenetics 825
27.6.1 Fetal Cell Screening in Maternal Blood 825
27.6.2 Subtelomeric FISH for Detection of Cryptic Translocations 827
27.6.3 Detection of Gene Duplications 831
27.6.4 Four-Color FISH for Aneuploidy Screening 840
27.6.5 Thick Specimen Imaging 843
27.7 Commercially Available Systems 848
27.8 Conclusions 849
References 851
Chapter 28. Towards Video Processing 856
Index 858