Digital Image Processing

• Adopts algorithmic approach to illustrate image processing.
• Provides simple explanations to topics such as shannon-fano coding, morphological gradient, polygonal approximations, shape number, component labelling.
• Includes solved numerical examples interspersed throughout the book.
• Appendices discussing the basics of MATLAB programming, ImageJ, and information on other public domain image processing software.
• Includes a laboratory manual with simple examples illustrated through MATLAB and guided exercises through ImageJ, a GUI based public domain software.

Readership: Primary: Core / elective for (CSE, IT, ECE, EEE disciplines). Secondary: Engineering Diploma

Sridhar

Dr S Sridhar is Associate Professor, Department of Information Science& Technology, College of Engineering Guindy, Anna University, Chennai. He has more than 20 years of active teaching and research experience. He has served as a project trainee at Council for Scientific and Industrial Research (CSIR), Chennai and Indian Space Research Organisation (ISRO) for a short stint. A PhD from Anna University, his doctoral specialisation was in medical imaging. He has published a lot of research articles related to medical imaging in leading national and international journals.

1: Introduction to Image Processing
1.1 Overview of Image Processing
1.2 Nature of Digital Image Processing
1.3 Image Processing and Related Fields
1.3.1 Image Processing and Computer Graphics
1.3.2 Image Processing and Signal Processing
1.3.3 Image Processing and Machine Vision
1.3.4 Image Processing and Video Processing
1.3.5 Image Processing and Optics
1.3.6 Image Processing and Statistics
1.4 Digital Image Representation
1.5 Types of Images
1.5.1 Types of Images Based on Attributes
1.5.2 Types of Images Based on Colour
1.5.2.1 Grey scale images
1.5.2.2 Binary images
1.5.2.3 Colour images
1.5.2.4 Pseudocolour images
1.5.3 Types of Images Based on Dimensions
1.5.4 Types of Images Based on Data Types
1.6 Digital Image Processing Operations
1.7 Fundamental Steps in Image Processing
1.7.1 Image Enhancement
1.7.2 Image Restoration
1.7.3 Image Compression
1.7.4 Image Analysis
1.7.5 Image Synthesis
1.8 Image Processing Applications
1.8.1 Biometrics
1.8.2 Medical Imaging
1.8.3 Factory Automation
1.8.4 Remote Sensing
1.8.5 Document Image Processing
1.8.6 Defense/Military Applications
1.8.7 Photography
1.8.8 Entertainment
1.9 Digital Imaging System
1.9.1 Image Sensors
1.9.2 Image Storage
1.9.3 Image Processor
1.9.4 Output Devices
1.9.5 Networking Components
1.9.6 Image Processing Software
2: Digital Imaging System
2.1 Physical Aspects of Image Acquisition
2.1.1 Nature of Light
2.1.2 Lighting System Design
2.1.3 Simple Image Formation Process
2.2 Biological Aspects of Image Acquisition
2.2.1 Human Visual System
2.2.2 Properties of Human Visual System
2.2.2.1 Brightness adaptation
2.2.2.2 Intensity and brightness
2.2.2.3 Simultaneous contrast
2.2.2.4 Mach bands
2.2.2.5 Frequency response
2.3 Review of Digital Camera
2.4 Sampling and Quantization
2.5 Image Quality
2.5.1 Optical Resolution
2.5.2 Image Display Devices and Device Resolution
2.5.3 Digital Halftone Process
2.5.3.1 Random dithering
2.5.3.2 Ordered dithering
2.5.3.3 Non-periodic dithering
2.6 Image Storage and File Formats
2.6.1 Types of File Formats
2.6.2 Structure of File Format
3: Digital Image Processing Operations
3.1 Basic Relationships and Distance Metrics
3.1.1 Image Coordinate System
3.1.2 Image Topology
3.1.3 Connectivity
3.1.4 Relations
3.1.5 Distance Measures
3.1.6 Some Important Image Characteristics
3.2 Image Processing Operations
3.2.1 Arithmetic Operations
3.2.1.1 Image addition
3.2.1.2 Image subtraction
3.2.1.3 Image multiplication
3.2.1.4 Image division
3.2.1.5 Applications of arithmetic operations
3.2.2 Logical Operations
3.2.2.1 AND/NAND
3.2.2.2 OR/NOR
3.2.2.3 XOR/XNOR T
3.2.2.4 Invert/Logical NOT
3.2.3 Geometrical Operations
3.2.3.1 Translation
3.2.3.2 Scaling
3.2.3.3 Zooming ion
3.2.3.4 Linear interpolation
3.2.3.5 Mirror or reflection operation
3.2.3.6 Shearing
3.2.3.7 Rotation
3.2.3.8 Affine transform
3.2.3.9 Inverse transformation
3.2.3.10 3D Transforms hniques
3.2.4 Image Interpolation Techniques
3.2.4.1 Downsampling
3.2.4.2 Upsampling
3.2.5 Set Operations s
3.2.6 Statistical Operations perations
3.2.7 Convolution and Correlation Operations ions 3.3 Data Structures and Image Processing Applications Development
3.3.1 Matrix
3.3.2 Chain Code
3.3.3 Region Adjacency Graph
3.3.4 Relational Structures
3.3.5 Hierarchical Data Structures
3.3.5.1 Pyramids
3.3.5.2 Quadtrees
3.3.6 Application Development
4: Digital Image Transforms
4.1 Need for Image Transforms
4.1.1 Introduction to Fourier Transform
4.1.2 Discrete Fourier Transform
4.1.3 Fast Fourier Transform
4.2 Properties of Fourier Transform
4.2.1 Sampling Theorem
4.2.2 Parseval's Theorem
4.3 Discrete Cosine Transform
4.4 Discrete Sine Transform
4.5 Walsh Transform
4.6 Hadamard Transform
4.7 Haar Transform
4.8 Slant Transform
4.9 SVD and KL Transforms
4.9.1 Singular-value Decomposition Transform
4.9.2 Karhunen-Loeve Transform or Hotelling Transform
5: Image Enhancement and Restoration
5.1 Image Quality and Need for Image Enhancement
5.1.1 Image Quality Factors
5.1.2 Image Quality Assessment Tool
5.1.3 Image Quality Metrics
5.2 Image Enhancement Point Operations
5.2.1 Linear and Non-Linear Functions
5.2.1.1 Inversion (digital negative operation)
5.2.1.2 What is a non-linear operator?
5.2.2 Piecewise Linear Functions
5.2.2.2 Intensity slicing
5.2.2.3 Bit-plane slicing
5.2.3 Histogram-Based Techniques
5.2.3.1 Histogram stretching
5.2.3.2 Histogram sliding
5.2.3.3 Histogram equalization
5.2.3.4 Histogram specification
5.2.3.5 Local and adaptive contrast enhancement
5.3 Spatial Filtering Concepts
5.3.1 Image Smoothing Spatial Filters
5.3.1.1 How to design a discrete Gaussian mask?
5.3.1.2 Non-linear filters
5.3.1.3 Directional smoothing
5.3.2 Image Sharpening Spatial Filters
5.3.2.1 High-boost filter
5.4 Frequency Domain Filtering
5.4.1 Image Smoothing in Frequency Domain
5.4.2 Image Sharpening in Frequency Domain
5.4.2.1 Band-pass Filtering
5.5 Image Degradation (Restoration) Model
5.6 Categories of Image Degradations
5.6.1 Noise Modelling
5.6.1.1 Noise categories based on distribution
5.6.1.2 Noise categories based on correlation
5.6.1.3 Noise categories based on nature
5.6.1.4 Noise categories based on source
5.6.2 Blur and Distortions
5.7 Image Restoration in the Presence of Noise Only
5.7.1 Mean Filters
5.7.1.1 Arithmetic mean filter
5.7.1.2 Contra-harmonic mean filter
5.7.1.3 Geometric mean filter
5.7.1.4 Harmonic mean filter
5.7.1.5 Yp mean filter
5.7.2 Order-Statistics Filters
5.7.2.1 Median filter
5.7.2.2 Maximum filter
5.7.2.3 Minimum filter
5.7.2.4 Midpoint filter
5.7.2.5 Alpha-trimmed mean filter
5.8 Image Restoration Techniques
5.8.1 Constrained Method
5.8.2 Unconstrained Method
5.8.2.1 Wiener filter
5.8.2.2 Constrained least square filter
5.8.2.3 Pseudo-inverse filter
5.8.3 Interactive Image Restoration
5.8.4 Blind Image Restoration
5.9 Geometrical Transforms for Image Restoration
6: Image Compression
6.1 Image Compression Model
6.2 Compression Algorithm and its Types
6.2.1 Entropy Coding
6.2.2 Predictive Coding
6.2.3 Transform Coding
6.2.4 Layered Coding
6.3 Types of Redundancy
6.3.1 Coding Redundancy
6.3.2 Inter-pixel Redundancy
6.3.3 Psychovisual Redundancy
6.3.4 Chromatic Redundancy
6.4 Lossless Compression Algorithms
6.4.1 Run-Length Coding
6.4.2 Huffman Coding
6.4.2.1 Canonical Huffman code
6.4.2.2 Huffman decoder
6.4.2.3 Characteristics of Huffman coding
6.4.4 Bit-Plane Coding
6.4.5 Arithmetic Coding
6.4.6 Dictionary-Based Coding
6.4.6.1 Encoding
6.4.6.2 Decoding
6.4.7 Lossless Predictive Coding
6.5 Lossy Compression Algorithms
6.5.1. Lossy Predictive Coding
6.5.2 Vector Quantization
6.5.2.1 Codebook design
6.5.2.2 Generalized Lloyd algorithm
6.5.3 Block Transform Coding
6.5.3.1 Sub-image selection
6.5.3.2 Transform selection
6.5.3.3 Bit allocation
6.5.3.4 Zonal coding
6.5.3.5 Threshold mark
6.6 Image and Video Compression Standards
6.6.1 JPEG
6.6.1.1 Sequential DCT-based mode (baseline algorithm)
6.6.1.2 Lossless mode
6.6.1.3 Progressive encoding
6.6.1.4 Hierarchical mode
6.6.2 Video Compression-MPEG
6.6.2.1 Macroblock formation
6.6.2.2 Frame formation
6.6.2.3 Group of pictures
6.6.2.4 Motion estimation
6.6.2.5 Audio compression
6.6.3 MPEG Variations
7: Image Segmentation
7.1 Introduction
7.2 Classification of Image Segmentation Algorithms
7.3 Detection of Discontinuities and Line-Detection Approaches
7.4 Edge Detection
7.4.1 Stages in Edge Detection
7.4.1.1 Filtering
7.4.1.2 Differentiation
7.4.1.3 Localization
7.4.2 Types of Edge Detectors
7.4.3 First-Order Edge Detection Operators
7.4.3.1 Roberts operator
7.4.3.2 Prewitt operator
7.4.3.3 Sobel operator
7.4.3.4 Template matching masks
7.4.4 Second-Order Derivative Filters
7.4.4.1 Laplacian of Gaussian (Marr-Hildrith) operator
7.4.4.2 Combined detection
7.4.4.3 Difference of Gaussian filter
7.4.4.4 Canny edge detection
7.4.4.5 Pattern fit algorithm
7.4.5 Edge Operator Performance
7.4.6 Edge-linking Algorithms
7.4.6.1 Edge relaxation
7.4.6.2 Graph theoretic algorithms
7.5 Hough Transforms and Shape Detection
7.6 Corner Detection
7.7 Principle of Thresholding
7.7.1 Global Thresholding Algorithms
7.7.2 Multiple Thresholding gorithms
7.7.3 Adaptive Thresholding Algorithm
7.7.4 Optimal Thresholding Algorithms
7.7.4.1 Parametric methods Algorithms
7.7.4.2 Non-parametric methods
7.8 Principle of Region Growing
7.8.1 Region-growing Algorithmg
7.8.2 Split-and-merge Algorithm
7.8.3 Split-and-merge Algorithm using Pyramid Quadtree
7.9 Dynamic Segmentation Approaches g Pyramid Quadtree
7.9.1 Use of Motion in Segmentation
7.9.2 Hybrid Edge/Region Approaches
7.10 Validation of Segmentation Algorithms
8: Colour Image Processing Algorithms
8.1 Colour Fundamentals
8.2 Devices for Colour Imaging
8.2.1 Types of Cameras
8.2.2 Colour Monitors
8.3 Colour Image Storage and Processing
8.4 Colour Models
8.4.1 RGB Colour Model
8.4.2 HSI Colour Model
8.4.3 HSV Colour Model
8.4.4 HLS Colour Model
8.4.5 TV Colour Models
8.4.6 Printing Colour Models
8.5 Colour Quantization
8.5.1 Popularity (or Populosity) Algorithm
8.5.2 Median-cut Algorithm
8.5.3 Octree-based Algorithm
8.6 Pseudocolour Image Processing
8.7 Full Colour Processing
8.7.1 Colour Transformations
8.7.1.1 Intensity modifications
8.7.1.2 Colour negatives
8.7.1.3 Colour slicing
8.7.1.4 Tonal and colour correction
8.7.1.5 Histogram processing
8.7.2 Image Filters for Colour Images
8.7.3 Colour Image Segmentation
8.7.3.1 Thresholding
8.7.3.2 k-means clustering technique
8.7.3.3 RGB colour space segmentation
8.7.4 Colour Features
9: Image Morphology
9.1 Need for Morphological Processing
9.2 Morphological Operators
9.2.1 Algorithm for dilation and erosion
9.2.2 Opening and closing operationosion
9.3 Hit or Miss Transform
9.4 Basic Morphological Algorithmsn
9.4.1 Boundary extraction
9.4.2 Noise removal
9.4.3 Thinning
9.4.4 Thickening
9.4.5 Convex hull
9.4.6 Skeletonization
9.4.7 Medial axis transform and distance transform
9.4.8 Region filling
9.4.9 Extraction of connected component
9.4.10 Pruning
9.5 Gray Scale Morphology
9.5.1 Morphological gradient
9.5.2 Top-hat and well transformations
9.5.3 Morphological reconstruction
9.5.4 Watershed algorithm
10: Image Features Representation and Description
10.1 Boundary and Region Representation
10.1.1 Chain code
10.1.2 Polygonal approximations
10.1.3 Signatures
10.1.4 Bending Energy
10.1.5 Statistical moments
10.1.6 Region Representation
10.2 Boundary Descriptions
10.2.1 Simple Descriptors
10.2.2 Shape number
10.2.3 Fourier Descriptors
10.2.4 Run code
10.2.5 Projections
10.2.6 Concavity Tree
10.3 Component Labeling
10.3.1 Recursive Algorithm
10.3.2 Sequential Algorithm
10.4 Basics of Regional Descriptions
10.4.1 Histogram (Brightness) Features
10.4.2 Shape Features
10.4.3 Spatial moments
10.4.4 Central and invariant moments
10.4.4 Central and invariant moments
10.4.5 Topological Features
10.4.6 Transform Features
10.4.7 Texture Features
10.4.8 Syntactic and Structural features
10.5 Feature Selection Techniques
11: Object Recognition
11.1 Patterns and Pattern Classes
11.2 Template Matching
11.3 Introduction to Classification
11.4 Decision- Theoretic Methods
11.4.1 Linear Discriminant Analysis
11.4.2 Bayesian Classifiers
11.4.3 Non Parametric Statistical Methods
11.4.4 Regression Methods
11.5 Structural and Syntactic Classifier Algorithms
11.5.1 Grammar Oriented Recognition
11.5.2 Shape Matching Algorithms
11.5.3 String Matching Algorithms
11.5.4 Rule Based Algorithms
11.5.5 Graph oriented approaches
11.6 Evaluation of Classifier Algorithms
11.7 Biometrics Case studies
11.7.1 Face Recognition
11.7.2 Iris recognition
11.7.3 Fingerprint Recognition
11.7.4 Signature Verification
11.8 Clustering Techniques
11.8.1 Similarity Measures
11.8.2 Hierarchical Methods
11.8.3 k-means Algorithm
11.8.4 Cluster Evaluation Methods
12: Related Topics
12.1 Soft computing and Image Processing
12.1.1 Fuzzy Logic
12.1.2 Genetic Algorithms
12.1.3 Artificial Neural Networks
12.2 Multiresolution Analysis and Wavelet Transforms
12.2.1 Wavelet Transforms
12.3 Image Synthesis
12.3.1 Image Registration Techniques
12.3.2 Image Fusion Algorithms
12.3.3 Image Visualization
12.3.4 Image Understanding and Stereo Imaging
12.4 Digital Watermarking
12.5 Image Mining and Content Based Retrieval Systems
12.5.1 Data Mining for Image Data
12.5.2 Content - Based Image Retrieval Systems
Appendix A - A Brief Introduction to MATLAB programming
Appendix B - ImageJ and other open source alternatives
Appendix C - Laboratory exercises