A hands-on guide to image registration theory and methods--with examples of a wide range of real-world applicationsTheory and Applications of Image Registration offers comprehensive coverage of feature-based image registration methods. It provides in-depth exploration of an array of fundamental issues, including image orientation detection, similarity measures, feature extraction methods, and elastic transformation functions. Also covered are robust parameter estimation, validation methods, multi-temporal and multi-modality image registration, methods for determining the orientation of an image, methods for identifying locally unique neighborhoods in an image, methods for detecting lines in an image, methods for finding corresponding points and corresponding lines in images, registration of video images to create panoramas, and much more.Theory and Applications of Image Registration provides readers with a practical guide to the theory and underpinning principles. Throughout the book numerous real-world examples are given, illustrating how image registration can be applied to problems in various fields, including biomedicine, remote sensing, and computer vision. Also provided are software routines to help readers develop their image registration skills. Many of the algorithms described in the book have been implemented, and the software packages are made available to the readers of the book on a companion website. In addition, the book:* Explores the fundamentals of image registration and provides a comprehensive look at its multi-disciplinary applications* Reviews real-world applications of image registration in the fields of biomedical imaging, remote sensing, computer vision, and more* Discusses methods in the registration of long videos in target tracking and 3-D reconstruction* Addresses key research topics and explores potential solutions to a number of open problems in image registration* Includes a companion website featuring fully implemented algorithms and image registration software for hands-on learningTheory and Applications of Image Registration is a valuable resource for researchers and professionals working in industry and government agencies where image registration techniques are routinely employed. It is also an excellent supplementary text for graduate students in computer science, electrical engineering, software engineering, and medical physics.

Contributors xvAcknowledgments xviiAbout the Companion Website xix1 Introduction 11.1 Organization of the Book 31.2 Further Reading 5References 52 Image Orientation Detection 92.1 Introduction 92.2 Geometric Gradient and Geometric Smoothing 132.2.1 Calculating Geometric Gradients 152.3 Comparison of Geometric Gradients and Intensity Gradients 182.4 Finding the Rotational Difference between Two Images 212.5 Performance Evaluation 232.5.1 Reliability 232.5.2 Accuracy 312.5.3 Computational Complexity 322.6 Registering Images with a Known Rotational Difference 342.7 Discussion 362.8 Further Reading 37References 403 Feature Point Detection 433.1 Introduction 433.2 Variant Features 443.2.1 Central Moments 443.2.2 Uniqueness 483.3 Invariant Features 503.3.1 Rotation-Invariant Features 503.3.1.1 Laplacian of Gaussian (LoG) Detector 513.3.1.2 Entropy 533.3.1.3 InvariantMoments 553.3.2 SIFT: A Scale-and Rotation-Invariant Point Detector 583.3.3 Radiometric-Invariant Features 603.3.3.1 Harris Corner Detector 603.3.3.2 Hessian Corner Detector 633.4 Performance Evaluation 643.5 Further Reading 68References 684 FeatureLineDetection 754.1 Hough Transform Using Polar Equation of Lines 794.2 Hough Transform Using Slope and y-Intercept Equation of Lines 824.3 Line Detection Using Parametric Equation of Lines 864.4 Line Detection by Clustering 894.5 Line Detection by Contour Tracing 924.6 Line Detection by Curve Fitting 954.7 Line Detection by Region Subdivision 1014.8 Comparison of the Line Detection Algorithms 1064.8.1 Sensitivity to Noise 1064.8.2 Positional and Directional Errors 1064.8.3 Length Accuracy 1094.8.4 Speed 1094.8.5 Quality of Detected Lines 1094.9 Revisiting Image Dominant Orientation Detection 1174.10 Further Reading 121References 1255 Finding Homologous Points 1335.1 Introduction 1335.2 Point Pattern Matching 1345.2.1 Parameter Estimation by Clustering 1375.2.2 Parameter Estimation by RANSAC 1415.3 Point Descriptors 1465.3.1 Histogram-Based Descriptors 1475.3.2 SIFT Descriptor 1485.3.3 GLOH Descriptor 1515.3.4 Composite Descriptors 1525.3.4.1 Hu InvariantMoments 1525.3.4.2 Complex Moments 1525.3.4.3 Cornerness Measures 1535.3.4.4 Power Spectrum Features 1545.3.4.5 Differential Features 1555.3.4.6 Spatial Domain Features 1555.4 SimilarityMeasures 1605.4.1 Correlation Coefficient 1605.4.2 Minimum Ratio 1615.4.3 Spearman's 1615.4.4 Ordinal Measure 1625.4.5 Correlation Ratio 1625.4.6 Shannon Mutual Information 1645.4.7 Tsallis Mutual Information 1655.4.8 F-Information 1665.5 Distance Measures 1675.5.1 Sum of Absolute Differences 1675.5.2 Median of Absolute Differences 1675.5.3 Square Euclidean Distance 1685.5.4 Intensity-Ratio Variance 1685.5.5 Rank Distance 1695.5.6 Shannon Joint Entropy 1695.5.7 Exclusive F-Information 1705.6 TemplateMatching 1705.6.1 Coarse-to-Fine Matching 1715.6.2 MultistageMatching 1725.6.3 Rotationally InvariantMatching 1735.6.4 Gaussian-Weighted TemplateMatching 1745.6.5 Template Matching in Different Modality Rotated Images 1755.7 Robust Parameter Estimation 1785.7.1 Ordinary Least-Squares Estimator 1805.7.2 Weighted Least-Squares Estimator 1825.7.3 Least Median of Squares Estimator 1845.7.4 Least Trimmed Squares Estimator 1845.7.5 Rank Estimator 1855.8 Finding Optimal Transformation Parameters 1935.9 Performance Evaluation 1935.10 Further Reading 197References 2006 Finding Homologous Lines 2156.1 Introduction 2156.2 Determining Transformation Parameters from Line Parameters 2156.3 Finding Homologous Lines by Clustering 2216.3.1 Finding the Rotation Parameter 2226.3.2 Finding the Translation Parameters 2236.4 Finding Homologous Lines by RANSAC 2296.5 Line Grouping Using Local Image Information 2326.6 Line Grouping Using Vanishing Points 2356.6.1 Methods Searching the Image Space 2356.6.2 Methods Searching the Polar Space 2366.6.3 Methods Searching the Gaussian Sphere 2366.6.4 A Method Searching Both Image and Gaussian Sphere 2376.6.5 Measuring the Accuracy of Detected Vanishing Points 2446.6.6 Discussion 2476.7 Robust Parameter Estimation Using Homologous Lines 2536.8 Revisiting Image Dominant Orientation Detection 2556.9 Further Reading 256References 2577 Nonrigid Image Registration 2617.1 Introduction 2617.2 Finding Homologous Points 2627.2.1 Coarse-to-Fine Matching 2627.2.2 Correspondence by Template Matching 2697.3 Outlier Removal 2747.4 Elastic Transformation Models 2787.4.1 Surface Spline (SS) Interpolation 2807.4.2 Piecewise Linear (PWL) Interpolation 2827.4.3 Moving Least Squares (MLS) Approximation 2837.4.4 Weighted Linear (WL) Approximation 2857.4.5 Performance Evaluation 2877.4.6 Choosing the Right Transformation Model 2917.5 Further Reading 292References 2938 Volume Image Registration 2998.1 Introduction 2998.2 Feature Point Detection 3018.2.1 Central Moments 3018.2.2 Entropy 3028.2.3 LoG Operator 3028.2.4 First-Derivative Intensities 3038.2.5 Second-Derivative Intensities 3048.2.6 Speed-Up Considerations in Feature Point Detection 3058.2.7 Evaluation of Feature Point Detectors 3058.3 Finding Homologous Points 3078.3.1 Finding Initial Homologous Points Using Image Descriptors 3108.3.2 Finding Initial Homologous Points by Template Matching 3138.3.3 Finding Final Homologous Points from Coarse to Fine 3158.3.4 Finding the Final Homologous Points by Outlier Removal 3208.4 Transformation Models for Volume Image Registration 3218.4.1 Volume Spline 3238.4.2 Weighted Rigid Transformation 3258.4.3 Computing the Overall Transformation 3278.5 Performance Evaluation 3308.5.1 Accuracy 3308.5.2 Reliability 3338.5.3 Speed 3338.6 Further Reading 335References 3379 Validation Methods 3439.1 Introduction 3439.2 Validation Using Simulation Data 3449.3 Validation Using a Gold Standard 3459.4 Validation by an Expert Observer 3479.5 Validation Using a Consistency Measure 3489.6 Validation Using a Similarity/DistanceMeasure 3509.7 Further Reading 351References 35210 Video Image Registration 357EdgardoMolina,Wai Lun Khoo, Hao Tang, and Zhigang Zhu10.1 Introduction 35710.2 Motion Modeling 35810.2.1 The Motion Field of Rigid Objects 35810.2.2 Motion Models 36010.2.2.1 Pure Rotation and a 3-D Scene 36110.2.2.2 General Motion and a Planar Scene 36210.2.2.3 TranslationalMotion and a 3-D Scene 36310.3 Image Alignment 36510.3.1 Feature-Based Methods 36710.3.2 Mechanical-Based Methods 36910.4 Image Composition 37010.4.1 Compositing Surface 37010.4.2 ImageWarping 37110.4.3 Pixel Selection and Blending 37310.5 Application Examples 37410.5.1 Pushbroom Stereo Mosaics Under TranslationalMotion 37410.5.1.1 Parallel-Perspective Geometry and Panoramas 37410.5.1.2 Stereo and Multiview Panoramas 37610.5.1.3 Results 37810.5.2 Stereo Mosaics when Moving a Camera on a Circular Path 37810.5.2.1 Circular Geometry 37910.5.2.2 Stereo Geometry 37910.5.2.3 Geometry and ResultsWhen Using PRISM 38110.5.3 Multimodal Panoramic Registration of Video Images 38210.5.3.1 Concentric Geometry 38310.5.3.2 Multimodal Alignment 38510.5.3.3 Results 38710.5.4 Video Mosaics Under GeneralMotion 38710.5.4.1 Direct Layering Approach 38910.5.4.2 Multiple Runs and Results 39210.6 Further Reading 393References 39511 Multitemporal Image Registration 39711.1 Introduction 39711.2 Finding Transformation Parameters from Line Parameters 39811.3 Finding an Initial Set of Homologous Lines 39911.4 Maximizing the Number of Homologous Lines 40311.5 Examples of Multitemporal Image Registration 40611.6 Further Reading 413References 41512 Open Problems and Research Topics 41912.1 Finding Rotational Difference between Multimodality Images 41912.2 Designing a Robust Image Descriptor 42012.3 Finding Homologous Lines for Nonrigid Registration 42112.4 Nonrigid Registration Using Homologous Lines 42312.5 Transformation Models with Nonsymmetric Basis Functions 42312.6 Finding Homologous Points along Homologous Contours 42612.7 4-D Image Registration 429References 430Glossary 433Acronyms 437Symbols 439A Image Registration Software 441A.1 Chapter 2: Image Orientation Detection 441A.1.1 Introduction 441A.1.2 Operations 442A.2 Chapter 3: Feature Point Detection 444A.2.1 Introduction 444A.2.2 Operations 445A.3 Chapter 4: Feature Line Detection 448A.3.1 Introduction 448A.3.2 Operations 449A.4 Chapter 5: Finding Homologous Points 452A.4.1 Introduction 452A.4.2 Operations 452A.5 Chapter 6: Finding Homologous Lines 459A.5.1 Introduction 459A.5.2 Operations 460A.6 Chapter 7: Nonrigid Image Registration 469A.6.1 Introduction 469A.6.2 Operations 469A.7 Chapter 8: Volume Image Registration 479A.7.1 Introduction 479A.7.2 I/O File Formats 479A.7.3 Operations 480References 487Index 489