Machine Vision
Theory, Algorithms, Practicalities
- 1st Edition - September 23, 2014
- Latest edition
- Author: E. R. Davies
- Editors: P. G. Farrell, J. R. Forrest
- Language: English
- Paperback ISBN:9 7 8 - 1 - 4 8 3 2 - 3 7 9 5 - 4
- eBook ISBN:9 7 8 - 1 - 4 8 3 2 - 7 5 6 1 - 1
Machine Vision: Theory, Algorithms, Practicalities covers the limitations, constraints, and tradeoffs of vision algorithms. This book is organized into four parts encompassing 21… Read more
Machine Vision: Theory, Algorithms, Practicalities covers the limitations, constraints, and tradeoffs of vision algorithms. This book is organized into four parts encompassing 21 chapters that tackle general topics, such as noise suppression, edge detection, principles of illumination, feature recognition, Bayes’ theory, and Hough transforms. Part 1 provides research ideas on imaging and image filtering operations, thresholding techniques, edge detection, and binary shape and boundary pattern analyses. Part 2 deals with the area of intermediate-level vision, the nature of the Hough transform, shape detection, and corner location. Part 3 demonstrates some of the practical applications of the basic work previously covered in the book. This part also discusses some of the principles underlying implementation, including on lighting and hardware systems. Part 4 highlights the limitations and constraints of vision algorithms and their corresponding solutions. This book will prove useful to students with undergraduate course on vision for electronic engineering or computer science.
Preface
Acknowledgements
Glossary of Acronyms and Abbreviations
1 Vision, the Challenge
1.1 Introduction—Man and his Senses
1.2 The Nature of Vision
1.3 Automated Visual Inspection
1.4 What this Book is About
1.5 The Following Chapters
1.6 Bibliographical Notes
Part 1 Low-Level Processing
2 Images and Imaging Operations
2.1 Introduction
2.2 Image Processing Operations
2.3 Convolutions and Point Spread Functions
2.4 Sequential Versus Parallel Operations
2.5 Concluding Remarks
2.6 Bibliographical and Historical Notes
2.7 Problems
3 Basic Image Filtering Operations
3.1 Introduction
3.2 Noise Suppression by Gaussian Smoothing
3.3 Median Filtering
3.4 Mode Filtering
3.5 Bias Generated by Noise Suppression Filters
3.6 Reducing Computational Load
3.7 The Role of Filters in Industrial Applications of Vision
3.8 Sharp-Unsharp Masking
3.9 Concluding Remarks
3.10 Bibliographical and Historical Notes
3.11 Problems
4 Thresholding Techniques
4.1 Introduction
4.2 Region-Growing Methods
4.3 Thresholding
4.4 Adaptive Thresholding
4.5 Concluding Remarks
4.6 Bibliographical and Historical Notes
4.7 Problems
5 Locating Objects via Their Edges
5.1 Introduction
5.2 Basic Theory of Edge Detection
5.3 The Template Matching Approach
5.4 Theory of 3 x 3 Template Operators
5.5 Summary—Design Constraints and Conclusions
5.6 The Design of Differential Gradient Operators
5.7 The Concept of a Circular Operator
5.8 Detailed Implementation of Circular Operators
5.9 Structured Bands of Pixels in Neighbourhoods of Various Sizes
5.10 The Systematic Design of Differential Edge Operators
5.11 Problems with the Above Approach—Some Alternative Schemes
5.12 Concluding Remarks
5.13 Bibliographical and Historical Notes
5.14 Problems
6 Binary Shape Analysis
6.1 Introduction
6.2 Connectedness in Binary Images
6.3 Object Labelling and Counting
6.4 Metric Properties in Digital Images
6.5 Size Filtering
6.6 The Convex Hull and its Computation
6.7 Distance Functions and their Uses
6.8 Skeletons and Thinning
6.9 Some Simple Measures for Shape Recognition
6.10 Shape Description by Moments
6.11 Boundary Tracking Procedures
6.12 Concluding Remarks
6.13 Bibliographical and Historical Notes
6.14 Problems
7 Boundary Pattern Analysis
7.1 Introduction
7.2 Boundary Tracking Procedures
7.3 Template Matching—a Reminder
7.4 Centroidal Profiles
7.5 Problems with the Centroidal Profile Approach
7.6 The (s,ψ) Plot
7.7 Tackling the Problems of Occlusion
7.8 Chain Code
7.9 The (r,s) Plot
7.10 Accuracy of Boundary Length Measures
7.11 Concluding Remarks
7.12 Bibliographical and Historical Notes
7.13 Problems
Part 2 Intermediate-Level Processing
8 Line Detection
8.1 Introduction
8.2 Application of the Hough Transform to Line Detection
8.3 The Foot-of-Normal Method
8.4 Longitudinal Line Localization
8.5 Final Line Fitting
8.6 Concluding Remarks
8.7 Bibliographical and Historical Notes
8.8 Problem
9 Circle Detection
9.1 Introduction
9.2 Hough-Based Schemes for Circular Object Detection
9.3 The Problem of Unknown Circle Radius
9.4 The Problem of Accurate Centre Location
9.5 Overcoming the Speed Problem
9.6 Concluding Remarks
9.7 Bibliographical and Historical Notes
9.8 Problem
10 The Hough Transform and Its Nature
10.1 Introduction
10.2 The Generalized Hough Transform
10.3 Setting up the Generalized Hough Transform—Some Relevant Questions
10.4 Spatial Matched Filtering in Images
10.5 From Spatial Matched Filters to Generalized Hough Transforms
10.6 Gradient Weighting Versus Uniform Weighting
10.7 Summary
10.8 Applying the Generalized Hough Transform to Line Detection
10.9 An Instructive Example
10.10 Tradeoffs to Reduce Computational Load
10.11 The Effects of Occlusions for Objects with Straight Edges
10.12 Fast Implementations of the Hough Transform
10.13 The Approach of Gerig and Klein
10.14 Concluding Remarks
10.15 Bibliographical and Historical Notes
11 Ellipse Detection
11.1 Introduction
11.2 The Diameter Bisection Method
11.3 The Chord-Tangent Method
11.4 Finding the Remaining Ellipse Parameters
11.5 Reducing Computational Load for the Generalized Hough Transform Method
11.6 Comparing the Various Methods
11.7 Concluding Remarks
11.8 Bibliographical and Historical Notes
11.9 Problems
12 Polygon Detection
12.1 Introduction
12.2 The Generalized Hough Transform
12.3 Application to the Detection of Regular Polygons
12.4 The Case of an Arbitrary Triangle
12.5 The Case of an Arbitrary Rectangle
12.6 Lower Bounds on the Numbers of Parameter Planes
12.7 An Extension of the Triangle Result
12.8 Discussion
12.9 Determining Orientation
12.10 Concluding Remarks
12.11 Bibliographical and Historical Notes
12.12 Problems
13 Hole Detection
13.1 Introduction
13.2 The Template Matching Approach
13.3 The Lateral Histogram Technique
13.4 The Removal of Ambiguities in the Lateral Histogram Technique
13.5 Application of the Lateral Histogram Technique for Object Location
13.6 A Strategy Based on Applying the Histograms in Turn
13.7 Appraisal of the Hole Detection Problem
13.8 Concluding Remarks
13.9 Bibliographical and Historical Notes
13.10 Problems
14 Corner Detection
14.1 Introduction
14.2 Template Matching
14.3 Second-Order Derivative Schemes
14.4 A Median-Based Corner Detector
14.5 The Hough Transform Approach to Corner Detection
14.6 The Lateral Histogram Approach to Corner Detection
14.7 Corner Orientation
14.8 Concluding Remarks
14.9 Bibliographical and Historical Notes
14.10 Problems
Part 3 Application-Level Processing
15 Abstract Pattern Matching Techniques
15.1 Introduction
15.2 A Graph-Theoretic Approach to Object Location
15.3 Possibilities for Saving Computation
15.4 Using the Generalized Hough Transform for Feature Collation
15.5 Generalizing the Maximal Clique and Other Approaches
15.6 Relational Descriptors
15.7 Search
15.8 Concluding Remarks
15.9 Bibliographical and Historical Notes
15.10 Problems
16 The Three-Dimensional World
16.1 Introduction
16.2 Three-Dimensional Vision—the Variety of Methods
16.3 Projection Schemes for Three-Dimensional Vision
16.4 Shape from Shading
16.5 Photometric Stereo
16.6 The Assumption of Surface Smoothness
16.7 Shape from Texture
16.8 Use of Structured Lighting
16.9 Three-Dimensional Object Recognition Schemes
16.10 The Method of Ballard and Sabbah
16.11 The Method of Silberberg et al.
16.12 Horaud's Junction Orientation Technique
16.13 The 3DPO System of Bolles and Horaud
16.14 The IVISM System
16.15 Lowe's Approach
16.16 Motion and Optical Flow
16.17 Concluding Remarks
16.18 Bibliographical and Historical Notes
16.19 Problems
17 Automated Visual Inspection
17.1 Introduction
17.2 The Process of Inspection
17.3 Review of the Types of Object to be Inspected
17.4 Summary-the Main Categories of Inspection
17.5 Shape Deviations Relative to a Standard Template
17.6 Inspection of Circular Products
17.7 Inspection of Printed Circuits
17.8 Steel Strip and Wood Inspection
17.9 Bringing Inspection to the Factory
17.10 Concluding Remarks
17.11 Bibliographical and Historical Notes
18 Statistical Pattern Recognition
18.1 Introduction
18.2 The Nearest Neighbour Algorithm
18.3 Bayes' Decision Theory
18.4 Relation of the Nearest Neighbour and Bayes' Approaches
18.5 The Optimum Number of Features
18.6 Cost Functions and Error-Reject Tradeoff
18.7 The Relevance of Probability in Image Analysis
18.8 Concluding Remarks
18.9 Bibliographical and Historical Notes
18.10 Problems
19 Image Acquisition
19.1 Introduction
19.2 Illumination Schemes
19.3 Cameras and Digitization
19.4 The Sampling Theorem
19.5 Concluding Remarks
19.6 Bibliographical and Historical Notes
20 The Need for Speed: Real-Time Electronic Hardware Systems
20.1 Introduction
20.2 Parallel Processing
20.3 SIMD Systems
20.4 The Gain in Speed Attainable with N Processors
20.5 Flynn's Classification
20.6 Optimal Implementation of an Image Analysis Algorithm
20.7 Board-Level Processing Systems
20.8 VLSI
20.9 Concluding Remarks
20.10 Bibliographical and Historical Notes
Part 4 Perspectives on Vision
21 Machine Vision, Art or Science?
21.1 Introduction
21.2 Parameters of Importance in Machine Vision
21.3 Tradeoffs
21.4 Future Directions
21.5 Hardware, Algorithms and Processes
21.6 A Retrospective View
21.7 Just a Glimpse of Vision?
21.8 Bibliographical and Historical Notes
Appendix
Programming Notation
A.1 Introduction
A.2 The Pascal Language
A.3 Special Syntax Embedded in Pascal
A.4 On the Validity of the "Repeat until Finished" Construct
References
Subject Index
Author Index
- Edition: 1
- Latest edition
- Published: September 23, 2014
- Language: English
ED
E. R. Davies
Roy Davies was Emeritus Professor of Machine Vision at Royal Holloway, University of London. He worked on many aspects of vision, from feature detection to robust, real-time implementations of practical vision tasks. His interests included automated visual inspection, surveillance, vehicle guidance, crime detection and neural networks. He has published more than 200 papers, and three books. Machine Vision: Theory, Algorithms, Practicalities (1990) has been widely used internationally for more than 25 years, and is now out in this much enhanced fifth edition. Roy held a DSc at the University of London and was awarded Distinguished Fellow of the British Machine Vision Association, and Fellow of the International Association of Pattern Recognition.
Affiliations and expertise
Emeritus Professor of Machine Vision, Royal Holloway, University of London, UK (deceased)Read Machine Vision on ScienceDirect