Fundamentals of Image Data Mining: Analysis, Features, Classification and Retrieval (Texts in Computer Science)

✍ Scribed by Dengsheng Zhang

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 382
• Edition: 2nd ed. 2021
• Category: Library

✦ Synopsis

This unique and useful textbook presents a comprehensive review of the essentials of image data mining, and the latest cutting-edge techniques used in the field. The coverage spans all aspects of image analysis and understanding, offering deep insights into areas of feature extraction, machine learning, and image retrieval. The theoretical coverage is supported by practical mathematical models and algorithms, utilizing data from real-world examples and experiments. 

 

Topics and features: 

• Describes essential tools for image mining, covering Fourier transforms, Gabor filters, and contemporary wavelet transforms
• Develops many new exercises (most with MATLAB code and instructions)
• Includes review summaries at the end of each chapter
• Analyses state-of-the-art models, algorithms, and procedures for image mining
• Integrates new sections on pre-processing, discrete cosine transform, and statistical inference and testing
• Demonstrates how features like color, texture, and shape can be mined or extracted for image representation
• Applies powerful classification approaches: Bayesian classification, support vector machines, neural networks, and decision trees
• Implements imaging techniques for indexing, ranking, and presentation, as well as database visualization

 

This easy-to-follow, award-winning book illuminates how concepts from fundamental and advanced mathematics can be applied to solve a broad range of image data mining problems encountered by students and researchers of computer science. Students of mathematics and other scientific disciplines will also benefit from the applications and solutions described in the text, together with the hands-on exercises that enable the reader to gain first-hand experience of computing.

✦ Table of Contents

Preface
About This Book
Key Features
New Features of the Second Edition
Contents
About the Author
List of Figures
List of Tables
Preliminaries
1 Fourier Transform
1.1 Introduction
1.2 Fourier Series
1.2.1 Sinusoids
1.2.2 Fourier Series
1.2.3 Complex Fourier Series
1.3 Fourier Transform
1.4 Discrete Fourier Transform
1.4.1 DFT
1.4.2 Uncertainty Principle
1.4.3 Nyquist Theorem
1.5 2D Fourier Transform
1.6 Properties of 2D Fourier Transform
1.6.1 Separability
1.6.2 Translation
1.6.3 Rotation
1.6.4 Scaling
1.6.5 Convolution Theorem
1.7 Techniques of Computing FT Spectrum
1.8 Summary
1.9 Exercises
References
2 Windowed Fourier Transform
2.1 Introduction
2.2 Short-Time Fourier Transform
2.2.1 Spectrogram
2.3 Gabor Filters
2.3.1 Gabor Transform
2.3.2 Design of Gabor Filters
2.3.3 Spectra of Gabor Filters
2.4 Discrete Cosine Transform
2.4.1 1D DCT Model
2.4.2 DCT Bases
2.4.3 2D DCT
2.4.4 Computation of 2D DCT
2.5 Summary
2.6 Exercises
References
3 Wavelet Transform
3.1 Discrete Wavelet Transform
3.2 Multiresolution Analysis
3.3 Fast Wavelet Transform
3.3.1 DTW Decomposition Tree
3.3.2 1D Haar Wavelet Transform
3.3.3 2D Haar Wavelet Transform
3.3.4 Application of DWT on Image
3.4 Summary
3.5 Exercises
Image Representation and Feature Extraction
4 Color Feature Extraction
4.1 Introduction
4.2 Color Space
4.2.1 CIE XYZ, xyY Color Spaces
4.2.2 RGB Color Space
4.2.3 HSV, HSL and HSI Color Spaces
4.2.4 CIE LUV Color Space
4.2.5 Y′CbCr Color Space
4.3 Image Clustering and Segmentation
4.3.1 K-means Clustering
4.3.2 JSEG Segmentation
4.4 Color Feature Extraction
4.4.1 Color Histogram
4.4.1.1 Component Histogram
4.4.1.2 Indexed Color Histogram
4.4.1.3 Dominant Color Histogram
4.4.2 Color Structure Descriptor
4.4.3 Dominant Color Descriptor
4.4.4 Color Coherence Vector
4.4.5 Color Correlogram
4.4.6 Color Layout Descriptor
4.4.7 Scalable Color Descriptor
4.4.8 Color Moments
4.5 Image Enhancement
4.5.1 Noise Removal
4.5.2 Contrast Enhancement
4.6 Summary
4.7 Exercises
References
5 Texture Feature Extraction
5.1 Introduction
5.2 Spatial Texture Feature Extraction Methods
5.2.1 Tamura Textures
5.2.2 Gray-Level Co-Occurrence Matrices
5.2.3 Markov Random Field
5.2.4 Fractal Dimension
5.2.5 Discussions
5.3 Spectral Texture Feature Extraction Methods
5.3.1 DCT-Based Texture Feature
5.3.2 Texture Features Based on Gabor Filters
5.3.2.1 Gabor Filters
5.3.2.2 Gabor Spectrum
5.3.2.3 Texture Representation
5.3.2.4 Rotation Invariant Gabor Features
5.3.3 Texture Features Based on Wavelet Transform
5.3.3.1 Selection and Application of Wavelets
5.3.3.2 Contrast of DWT and Other Spectral Transforms
5.3.3.3 Multiresolution Analysis
5.3.4 Texture Features Based on Curvelet Transform
5.3.4.1 Curvelet Transform
5.3.4.2 Discrete Curvelet Transform
5.3.4.3 Curvelet Spectra
5.3.4.4 Curvelet Features
5.3.5 Discussions
5.4 Summary
5.5 Exercises
References
6 Shape Representation
6.1 Introduction
6.2 Perceptual Shape Descriptors
6.2.1 Circularity and Compactness
6.2.2 Eccentricity and Elongation
6.2.3 Convexity and Solidarity
6.2.4 Euler Number
6.2.5 Bending Energy
6.3 Contour-Based Shape Methods
6.3.1 Shape Signatures
6.3.1.1 Position Function
6.3.1.2 Centroid Distance
6.3.1.3 Angular Functions
6.3.1.4 Curvature Signature
6.3.1.5 Area Function
6.3.1.6 Discussions
6.3.2 Shape Context
6.3.3 Boundary Moments
6.3.4 Stochastic Method
6.3.5 Scale Space Method
6.3.5.1 Scale Space
6.3.5.2 Curvature Scale Space
6.3.6 Fourier Descriptor
6.3.7 Discussions
6.3.8 Syntactic Analysis
6.3.9 Polygon Decomposition
6.3.9.1 Chain Code Representation
6.3.9.2 Smooth Curve Decomposition
6.3.9.3 Discussions
6.4 Region-Based Shape Feature Extraction
6.4.1 Geometric Moments
6.4.2 Complex Moments
6.4.3 Generic Fourier Descriptor
6.4.4 Shape Matrix
6.4.5 Shape Profiles
6.4.5.1 Shape Projections
6.4.5.2 Radon Transform
6.4.6 Discussions
6.4.7 Convex Hull
6.4.8 Medial Axis
6.5 Summary
6.6 Exercises
References
Image Classification and Annotation
7 Bayesian Classification
7.1 Introduction
7.2 Naïve Bayesian Image Classification
7.2.1 NB Formulation
7.2.2 NB with Independent Features
7.2.3 NB with Bag of Features
7.3 Image Annotation with Word Co-occurrence
7.4 Image Annotation with Joint Probability
7.5 Cross-Media Relevance Model
7.6 Image Annotation with Parametric Model
7.7 Image Classification with Gaussian Process
7.8 Summary
7.9 Exercises
References
8 Support Vector Machine
8.1 Linear Classifier
8.1.1 A Theoretical Solution
8.1.2 An Optimal Solution
8.1.3 A Suboptimal Solution
8.2 K Nearest Neighbor Classification
8.3 Support Vector Machine
8.3.1 The Perceptron
8.3.2 SVM—The Primal Form
8.3.2.1 The Margin Between Two Classes
8.3.2.2 Margin Maximization
8.3.2.3 The Primal Form of SVM
8.3.3 The Dual Form of SVM
8.3.3.1 The Dual-Form Perceptron
8.3.4 Kernel-Based SVM
8.3.4.1 The Dual-Form SVM Versus NN Classifier
8.3.4.2 Kernel Definition
8.3.4.3 Building New Kernels
8.3.4.4 The Kernel Trick
8.3.5 The Pyramid Match Kernel
8.3.6 Discussions
8.4 Fusion of SVMs
8.4.1 Fusion of Binary SVMs
8.4.2 Multilevel Fusion of SVMs
8.4.3 Fusion of SVMs with Different Features
8.5 Summary
8.6 Exercises
References
9 Artificial Neural Network
9.1 Introduction
9.2 Artificial Neurons
9.2.1 An AND Neuron
9.2.2 An OR Neuron
9.3 Perceptron
9.4 Nonlinear Neural Network
9.5 Activation and Inhibition
9.5.1 Sigmoid Activation
9.5.2 Shunting Inhibition
9.6 The Backpropagation Neural Network
9.6.1 The BP Network and Error Function
9.6.2 Layer K Weight Estimation and Updating
9.6.3 Layer K−1 Weight Estimation and Updating
9.6.4 The BP Algorithm
9.7 Convolutional Neural Network
9.7.1 CNN Architecture
9.7.2 Input Layer
9.7.3 Convolution Layer 1 (C1)
9.7.3.1 2D Convolution
9.7.3.2 Stride and Padding
9.7.3.3 Bias
9.7.3.4 Volume Convolution in Layer C1
9.7.3.5 Depth of the Feature Map Volume
9.7.3.6 ReLU Activation
9.7.3.7 Batch Normalization
9.7.4 Pooling or Subsampling Layer 1 (S1)
9.7.5 Convolution Layer 2 (C2)
9.7.6 Hyperparameters
9.8 Implementation of CNN
9.8.1 CNN Architecture
9.8.2 Filters of the Convolution Layers
9.8.3 Filters of the Fully Connected Layers
9.8.4 Feature Maps of Convolution Layers
9.8.5 Matlab Implementation
9.9 Summary
9.10 Exercises
References
10 Image Annotation with Decision Tree
10.1 Introduction
10.2 ID3
10.2.1 ID3 Splitting Criterion
10.3 C4.5
10.3.1 C4.5 Splitting Criterion
10.4 CART
10.4.1 Classification Tree Splitting Criterion
10.4.2 Regression Tree Splitting Criterion
10.4.3 Application of Regression Tree
10.5 DT for Image Classification
10.5.1 Feature Discretization
10.5.2 Building the DT
10.5.3 Image Classification and Annotation with DT
10.6 Summary
10.7 Exercises
References
Image Retrieval and Presentation
11 Image Indexing
11.1 Numerical Indexing
11.1.1 List Indexing
11.1.2 Tree Indexing
11.2 Inverted File Indexing
11.2.1 Inverted File for Textual Documents Indexing
11.2.2 Inverted File for Image Indexing
11.2.2.1 Determine the Area Weight aw
11.2.2.2 Determine the Position Weight pw
11.2.2.3 Determine the Relationship Weight rw
11.2.2.4 Inverted File for Image Indexing
11.3 Summary
11.4 Exercises
References
12 Image Ranking
12.1 Introduction
12.2 Similarity Measures
12.2.1 Distance Metric
12.2.2 Minkowski-Form Distance
12.2.3 Mass-Based Distance
12.2.4 Cosine Distance
12.2.5 χ2 Statistic
12.2.6 Histogram Intersection
12.2.7 Quadratic Distance
12.2.8 Mahalanobis Distance
12.3 Performance Measures
12.3.1 Recall and Precision Pair (RPP)
12.3.2 F-measure
12.3.3 Percentage of Weighted Hits (PWH)
12.3.4 Percentage of Similarity Ranking (PSR)
12.3.5 Bullseye Accuracy
12.4 Hypothesis Testing
12.4.1 Introduction
12.4.2 Fundamental Theorems of Statistics
12.4.3 Properties of Normal Distribution
12.4.4 HT on a Single Population
12.4.5 Power of Test
12.4.6 HT on Difference of Means
12.4.7 Summary of HT
12.4.8 Margin of Error
12.5 Summary
12.6 Exercises
References
13 Image Presentation
13.1 Introduction
13.2 Caption Browsing
13.3 Category Browsing
13.3.1 Category Browsing on the Web
13.3.2 Hierarchical Category Browsing
13.4 Content Browsing
13.4.1 Content Browsing in 3D Space
13.4.2 Content Browsing with Fish Eye View
13.4.3 Force-Directed Content Browsing
13.5 Query by Example
13.6 Query by Keywords
13.7 Summary
13.8 Exercises
References
Appendix Deriving the Conditional Probability of a Gaussian Process
Index