1 Introduction

Preview

1.1 Background

1.2 What Is Digital Image Processing?

1.3 Background on MATLAB and the Image Processing Toolbox

1.4 Areas of Image Processing Covered in the Book

1.5 The Book Web Site

1.6 Notation

1.7 The MATLAB Working Environment

1.7.1 The MATLAB Desktop

1.7.2 Using the MATLAB Editor to Create M-Files

1.7.3 Getting Help

1.7.4 Saving and Retrieving a Work Session

1.8 How References Are Organized in the Book

Summary

2 Fundamentals

Preview

2.1 Digital Image Representation

2.1.1 Coordinate Conventions

2.1.2 Images as Matrices

2.2 Reading Images

2.3 Displaying Images

2.4 Writing Images

2.5 Data Classes

2.6 Image Types

2.6.1 Intensity Images

2.6.2 Binary Images

2.6.3 A Note on Terminology

2.7 Converting between Data Classes and Image Types

2.7.1 Converting between Data Classes

2.7.2 Converting between Image Classes and Types

2.8 Array Indexing

2.8.1 Vector Indexing

2.8.2 Matrix Indexing

2.8.3 Selecting Array Dimensions

2.9 Some Important Standard Arrays

2.10 Introduction to M-Function Programming

2.10.1 M-Files

2.10.2 Operators

2.10.3 Flow Control

2.10.4 Code Optimization

2.10.5 Interactive I/O

2.10.6 A Brief Introduction to Cell Arrays and Structures

Summary

3 Intensity Transformations and Spatial Filtering

Preview

3.1 Background

3.2 Intensity Transformation Functions

3.2.1 Function imadjust

3.2.2 Logarithmic and Contrast-Stretching Transformations

3.2.3 Some Utility M-Functions for Intensity Transformations

3.3 Histogram Processing and Function Plotting

3.3.1 Generating and Plotting Image Histograms

3.3.2 Histogram Equalization

3.3.3 Histogram Matching (Specification)

3.4 Spatial Filtering

3.4.1 Linear Spatial Filtering

3.4.2 Nonlinear Spatial Filtering

3.5 Image Processing Toolbox Standard Spatial Filters

3.5.1 Linear Spatial Filters

3.5.2 Nonlinear Spatial Filters

Summary

4 Frequency Domain, Processing

Preview

4.1 The 2-D Discrete Fourier Transform

4.2 Computing and Visualizing the 2-D DFT in MATLAB

4.3 Filtering in the Frequency Domain

4.3.1 Fundamental Concepts

4.3.2 Basic Steps in DFT Filtering

4.3.3 An M-function for Filtering in the Frequency Domain

4.4 Obtaining Frequency Domain Filters from Spatial Filters

4.5 Generating Filters Directly in the Frequency Domain

4.5.1 Creating Meshgrid Arrays for Use in Implementing Filters in the Frequency Domain

4.5.2 Lowpass Frequency Domain Filters

4.5.3 Wireframe and Surface Plotting

4.6 Sharpening Frequency Domain Filters

4.6.1 Basic Highpass Filtering

4.6.2 High-Frequency Emphasis Filtering

Summary

5 Image Restoration

Preview

5.1 A Model of the Image Degradation/Restoration Process

5.2 Noise Models

5.2.1 Adding Noise with Function imnoise

5.2.2 Generating Spatial Random Noise with a Specified Distribution

5.2.3 Periodic Noise

5.2.4 Estimating Noise Parameters

5.3 Restoration in the Presence of Noise Only—Spatial Filtering

5.3.1 Spatial Noise Filters

5.3.2 Adaptive Spatial Filters

5.4 Periodic Noise Reduction by Frequency Domain Filtering

5.5 Modeling the Degradation Function

5.6 Direct Inverse Filtering

5.7 Wiener Filtering

5.8 Constrained Least Squares (Regularized) Filtering

5.9 Iterative Nonlinear Restoration Using the Lucy-Richardson Algorithm

5.10 Blind Deconvolution

5.11 Geometric Transformations and Image Registration

5.11.1 Geometric Spatial Transformations

5.11.2 Applying Spatial Transformations to Images

5.11.3 Image Registration

Summary

6 Color Image Processing

Preview

6.1 Color Image Representation in MATLAB

6.1.1 RGB Images

6.1.2 Indexed Images

6.1.3 IPT Functions for Manipulating RGB and Indexed Images

6.2 Converting to Other Color Spaces

6.2.1 NTSC Color Space

6.2.2 The YCbCr Color Space

6.2.3 The HSV Color Space

6.2.4 The CMY and CMYK Color Spaces

6.2.5 The HSI Color Space

6.3 The Basics of Color Image Processing

6.4 Color Transformations

6.5 Spatial Filtering of Color Images

6.5.1 Color Image Smoothing

6.5.2 Color Image Sharpening

6.6 Working Directly in RGB Vector Space

6.6.1 Color Edge Detection Using the Gradient

6.6.2 Image Segmentation in RGB Vector Space

Summary

7 Wavelets

Preview

7.1 Background

7.2 The Fast Wavelet Transform

7.2.1 FWTs Using the Wavelet Toolbox

7.2.2 FWTs without the Wavelet Toolbox

7.3 Working with Wavelet Decomposition Structures

7.3.1 Editing Wavelet Decomposition Coefficients without the Wavelet Toolbox

7.3.2 Displaying Wavelet Decomposition Coefficients

7.4 The Inverse Fast Wavelet Transform

7.5 Wavelets in Image Processing

Summary

8 Image Compression

Preview

8.1 Background

8.2 Coding Redundancy

8.2.1 Huffman Codes

8.2.2 Huffman Encoding

8.2.3 Huffman Decoding

8.3 Interpixel Redundancy

8.4 Psychovisual Redundancy

8.5 JPEG Compression

8.5.1 JPEG

8.5.2 JPEG 2000

Summary

9 Morphological Image Processing

Preview

9.1 Preliminaries

9.1.1 Some Basic Concepts from Set Theory

9.1.2 Binary Images, Sets, and Logical Operators

9.2 Dilation and Erosion

9.2.1 Dilation

9.2.2 Structuring Element Decomposition

9.2.3 The strel Function

9.2.4 Erosion

9.3 Combining Dilation and Erosion

9.3.1 Opening and Closing

9.3.2 The Hit-or-Miss Transformation

9.3.3 Using Lookup Tables

9.3.4 Function bwmorph

9.4 Labeling Connected Components

9.5 Morphological Reconstruction

9.5.1 Opening by Reconstruction

9.5.2 Filling Holes

9.5.3 Clearing Border Objects

9.6 Gray-Scale Morphology

9.6.1 Dilation and Erosion

9.6.2 Opening and Closing

9.6.3 Reconstruction

Summary

10 Image Segmentation

Preview

10.1 Point, Line, and Edge Detection

10.1.1 Point Detection

10.1.2 Line Detection

10.1.3 Edge Detection Using Function edge

10.2 Line Detection Using the Hough Transform

10.2.1 Hough Transform Peak Detection

10.2.2 Hough Transform Line Detection and Linking

10.3 Thresholding

10.3.1 Global Thresholding

10.3.2 Local Thresholding

10.4 Region-Based Segmentation

10.4.1 Basic Formulation

10.4.2 Region Growing

10.4.3 Region Splitting and Merging

10.5 Segmentation Using the Watershed Transform

10.5.1 Watershed Segmentation Using the Distance Transform

10.5.2 Watershed Segmentation Using Gradients

10.5.3 Marker-Controlled Watershed Segmentation

Summary

11 Representation and Description

Preview

11.1 Background

11.1.1 Cell Arrays and Structures

11.1.2 Some Additional MATLAB and IPT Functions Used in This Chapter

11.1.3 Some Basic Utility M-Functions

11.2 Representation

11.2.1 Chain Codes

11.2.2 Polygonal Approximations Using Minimum-Perimeter Polygons

11.2.3 Signatures

11.2.4 Boundary Segments

11.2.5 Skeletons

11.3 Boundary Descriptors

11.3.1 Some Simple Descriptors

11.3.2 Shape Numbers

11.3.3 Fourier Descriptors

11.3.4 Statistical Moments

11.4 Regional Descriptors

11.4.1 Function regionprops

11.4.2 Texture

11.4.3 Moment Invariants

11.5 Using Principal Components for Description

Summary

12 Object Recognition

Preview

12.1 Background

12.2 Computing Distance Measures in MATLAB

12.3 Recognition Based on Decision-Theoretic Methods

12.3.1 Forming Pattern Vectors

12.3.2 Pattern Matching Using Minimum-Distance Classifiers

12.3.3 Matching by Correlation

12.3.4 Optimum Statistical Classifiers

12.3.5 Adaptive Learning Systems

12.4 Structural Recognition

12.4.1 Working with Strings in MATLAB

12.4.2 String Matching

Summary

Appendix A Function Summary

Appendix B ICE and MATLAB Graphical User Interfaces

Appendix C M-Functions

Bibliography

Index