Davies / Farrell / Forrest Machine Vision

Theory, Algorithms, Practicalities
1. Auflage 2014
ISBN: 978-1-4832-7561-1
Verlag: Elsevier Science & Techn.
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 572 Seiten, Web PDF

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.

Roy Davies is Emeritus Professor of Machine Vision at Royal Holloway, University of London. He has worked on many aspects of vision, from feature detection to robust, real-time implementations of practical vision tasks. His interests include 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 holds a DSc at the University of London, and has been awarded Distinguished Fellow of the British Machine Vision Association, and Fellow of the International Association of Pattern Recognition.

Autoren/Hrsg.

Davies, E. R.

Farrell, P. G.

Forrest, J. R.

Inhaltsverzeichnis

1;Front Cover
;1
2;Machine Vision: Theory, Algorithms, Practicalities;4
3;Copyright Page;5
4;Preface;6
5;Acknowledgements;10
6;Table of Contents;12
7;Chapter 1. Vision, the Challenge;26
7.1;1.1 Introduction—man and his senses;26
7.2;1.2 The nature of vision;28
7.3;1.3 Automated visual inspection;36
7.4;1.4 What this book is about;38
7.5;1.5 The following chapters;39
7.6;1.6 Bibliographical notes;40
8;Part-1 Low-Level Processing;42
9;Chapter 2. Images and Imaging Operations;44
9.1;2.1 Introduction;44
9.2;2.2 Image processing operations;47
9.3;2.3 Convolutions and point spread functions;58
9.4;2.4 Sequential versus parallel operations;61
9.5;2.5 Concluding remarks;63
9.6;2.6 Bibliographical and historical notes;63
9.7;2.7 Problems;63
10;Chapter 3. Basic Image Filtering Operations;65
10.1;3.1 Introduction;65
10.2;3.2 Noise suppression by Gaussian smoothing;67
10.3;3.3 Median filtering;69
10.4;3.4 Mode filtering;72
10.5;3.5 Bias generated by noise suppression filters;79
10.6;3.6 Reducing computational load;90
10.7;3.7 The role of filters in industrial applications of vision;95
10.8;3.8 Sharp-unsharp masking;96
10.9;3.9 Concluding remarks;97
10.10;3.10 Bibliographical and historical notes;99
10.11;3.11 Problems;100
11;Chapter 4. Thresholding Techniques;102
11.1;4.1 Introduction;102
11.2;4.2 Region-growing methods;103
11.3;4.3 Thresholding;104
11.4;4.4 Adaptive thresholding;116
11.5;4.5 Concluding remarks;121
11.6;4.6 Bibliographical and historical notes;123
11.7;4.7 Problems;124
12;Chapter 5. Locating Objects via Their Edges;125
12.1;5.1 Introduction;125
12.2;5.2 Basic theory of edge detection;126
12.3;5.3 The template matching approach;127
12.4;5.4 Theory of 3 x 3 template operators;129
12.5;5.5 Summary—design constraints and conclusions;135
12.6;5.6 The design of differential gradient operators;136
12.7;5.7 The concept of a circular operator;138
12.8;5.8 Detailed implementation of circular operators;139
12.9;5.9 Structured bands of pixels in neighbourhoods of various sizes;141
12.10;5.10 The systematic design of differential edge operators;145
12.11;5.11 Problems with the above approach—some alternative
schemes;146
12.12;5.12 Concluding remarks;150
12.13;5.13 Bibliographical and historical notes;151
12.14;5.14 Problems;152
13;Chapter 6. Binary Shape Analysis;153
13.1;6.1 Introduction;153
13.2;6.2 Connectedness in binary images;154
13.3;6.3 Object labelling and counting;155
13.4;6.4 Metric properties in digital images;159
13.5;6.5 Size filtering;161
13.6;6.6 The convex hull and its computation;164
13.7;6.7 Distance functions and their uses;169
13.8;6.8 Skeletons and thinning;174
13.9;6.9 Some simple measures for shape recognition;187
13.10;6.10 Shape description by moments;188
13.11;6.11 Boundary tracking procedures;188
13.12;6.12 Concluding remarks;190
13.13;6.13 Bibliographical and historical notes;191
13.14;6.14 Problems;192
14;Chapter 7. Boundary Pattern Analysis;193
14.1;7.1 Introduction;193
14.2;7.2 Boundary tracking procedures;195
14.3;7.3 Template matching—a reminder;196
14.4;7.4 Centroidal profiles;196
14.5;7.5 Problems with the centroidal profile approach;198
14.6;7.6 The (s,.)
plot;203
14.7;7.7 Tackling the problems of occlusion;205
14.8;7.8 Chain code;208
14.9;7.9 The [r,s) plot;209
14.10;7.10 Accuracy of boundary length measures;209
14.11;7.11 Concluding remarks;211
14.12;7.12 Bibliographical and historical notes;212
14.13;7.13 Problems;213
15;Part-2 Intermediate-Level Processing;214
16;Chapter 8. Line Detection;216
16.1;8.1 Introduction;216
16.2;8.2 Application of the Hough transform to line detection;216
16.3;8.3 The foot-of-normal method;221
16.4;8.4 Longitudinal line localization;228
16.5;8.5 Final line fitting;228
16.6;8.6 Concluding remarks;229
16.7;8.7 Bibliographical and historical notes;230
16.8;8.8 Problem;231
17;Chapter 9. Circle Detection;232
17.1;9.1 Introduction;232
17.2;9.2 Hough-based schemes for circular object detection;233
17.3;9.3 The problem of unknown circle radius;237
17.4;9.4 The problem of accurate centre location;242
17.5;9.5 Overcoming the speed problem;253
17.6;9.6 Concluding remarks;263
17.7;9.7 Bibliographical and historical notes;264
17.8;9.8 Problem;264
18;Chapter 10. The Hough Transform and Its Nature;265
18.1;10.1 Introduction;265
18.2;10.2 The generalized Hough transform;265
18.3;10.3 Setting up the generalized Hough transform—some relevant
questions;267
18.4;10.4 Spatial matched filtering in images;268
18.5;10.5 From spatial matched filters to generalized Hough transforms;269
18.6;10.6 Gradient weighting versus uniform weighting;270
18.7;10.7 Summary;275
18.8;10.8 Applying the generalized Hough transform to line detection;276
18.9;10.9 An instructive example;278
18.10;10.10 Tradeoffs to reduce computational load;279
18.11;10.11 The effects of occlusions for objects with straight edges;279
18.12;10.12 Fast implementations of the Hough transform;282
18.13;10.13 The approach of Gerig and Klein;286
18.14;10.14 Concluding remarks;287
18.15;10.15 Bibliographical and historical notes;288
19;Chapter 11. Ellipse Detection;290
19.1;11.1 Introduction;290
19.2;11.2 The diameter bisection method;290
19.3;11.3 The chord-tangent method;293
19.4;11.4 Finding the remaining ellipse parameters;294
19.5;11.5 Reducing computational load for the generalized Hough
transform method;296
19.6;11.6 Comparing the various methods;304
19.7;11.7 Concluding remarks;306
19.8;11.8 Bibliographical and historical notes;308
19.9;11.9 Problems;308
20;Chapter 12. Polygon Detection;309
20.1;12.1 Introduction;309
20.2;12.2 The generalized Hough transform;309
20.3;12.3 Application to the detection of regular polygons;310
20.4;12.4 The case of an arbitrary triangle;312
20.5;12.5 The case of an arbitrary rectangle;313
20.6;12.6 Lower bounds on the numbers of parameter planes;315
20.7;12.7 An extension of the triangle result;318
20.8;12.8 Discussion;320
20.9;12.9 Determining orientation;323
20.10;12.10 Concluding remarks;324
20.11;12.11 Bibliographical and historical notes;324
20.12;12.12 Problems;325
21;Chapter 13. Hole Detection;326
21.1;13.1 Introduction;326
21.2;13.2 The template matching approach;326
21.3;13.3 The lateral histogram technique;328
21.4;13.4 The removal of ambiguities in the lateral histogram technique;329
21.5;13.5 Application of the lateral histogram technique for object
location;333
21.6;13.6 A strategy based on applying the histograms in turn;342
21.7;13.7 Appraisal of the hole detection problem;344
21.8;13.8 Concluding remarks;346
21.9;13.9 Bibliographical and historical notes;347
21.10;13.10 Problems;347
22;Chapater 14. Corner Location;348
22.1;14.1 Introduction;348
22.2;14.2 Template matching;348
22.3;14.3 Second-order derivative schemes;349
22.4;14.4 A median-based corner detector;352
22.5;14.5 The Hough transform approach to corner detection;358
22.6;14.6 The lateral histogram approach to corner detection;361
22.7;14.7 Corner orientation;363
22.8;14.8 Concluding remarks;364
22.9;14.9 Bibliographical and historical notes;365
22.10;14.10 Problems;365
23;Part-3 Application-Level Processing;368
24;Chapter 15. Abstract Pattern Matching Techniques;370
24.1;15.1 Introduction;370
24.2;15.2 A graph-theoretic approach to object location;371
24.3;15.3 Possibilities for saving computation;378
24.4;15.4 Using the generalized Hough transform for feature collation;382
24.5;15.5 Generalizing the maximal clique and other approaches;386
24.6;15.6 Relational descriptors;386
24.7;15.7 Search;390
24.8;15.8 Concluding remarks;391
24.9;15.9 Bibliographical and historical notes;392
24.10;15.10 Problems;393
25;Chapter 16. The Three-Dimensional World;394
25.1;16.1 Introduction;394
25.2;16.2 Three-dimensional vision—the variety of methods;395
25.3;16.3 Projection schemes for three-dimensional vision;397
25.4;16.4 Shape from shading;402
25.5;16.5 Photometric stereo;407
25.6;16.6 The assumption of surface smoothness;410
25.7;16.7 Shape from texture;411
25.8;16.8 Use of structured lighting;412
25.9;16.9 Three-dimensional object recognition schemes;414
25.10;16.10 The method of Ballard and Sabbah;416
25.11;16.11 The method of Silberberg et al.;419
25.12;16.12 Horaud's junction orientation
technique;420
25.13;16.13 The 3DPO system of Bolles and Horaud;425
25.14;16.14 The IVISM system;427
25.15;16.15 Lowe's approach;428
25.16;16.16 Motion and optical flow;430
25.17;16.17 Concluding remarks;431
25.18;16.18 Bibliographical and historical notes;432
25.19;16.19 Problems;434
26;Chapter 17. Automated Visual Inspection;436
26.1;17.1 Introduction;436
26.2;17.2 The process of inspection;437
26.3;17.3 Review of the types of object to be inspected;438
26.4;17.4 Summary—the main categories of inspection;441
26.5;17.5 Shape deviations relative to a standard template;443
26.6;17.6 Inspection of circular products;444
26.7;17.7 Inspection of printed circuits;452
26.8;17.8 Steel strip and wood inspection;454
26.9;17.9 Bringing inspection to the factory;455
26.10;17.10 Concluding remarks;456
26.11;17.11 Bibliographical and historical notes;457
27;Chapter 18. Statistical Pattern Recognition;460
27.1;18.1 Introduction;460
27.2;18.2 The nearest neighbour algorithm;461
27.3;18.3 Bayes' decision theory;464
27.4;18.4 Relation of the nearest neighbour and Bayes' approaches;466
27.5;18.5 The optimum number of features;470
27.6;18.6 Cost functions and error-reject tradeoff;471
27.7;18.7 The relevance of probability in image analysis;473
27.8;18.8 Concluding remarks;474
27.9;18.9 Bibliographical and historical notes;475
27.10;18.10 Problems;476
28;Chapter 19. Image Acquisition;477
28.1;19.1 Introduction;477
28.2;19.2 Illumination schemes;478
28.3;19.3 Cameras and digitization;486
28.4;19.4 The sampling theorem;490
28.5;19.5 Concluding remarks;494
28.6;19.6 Bibliographical and historical notes;494
29;Chapter 20. The Need for Speed: Real-Time Electronic
Hardware Systems;496
29.1;20.1 Introduction;496
29.2;20.2 Parallel processing;497
29.3;20.3 SIMD systems;499
29.4;20.4 The gain in speed attainable with N processors;501
29.5;20.5 Flynn's classification;502
29.6;20.6 Optimal implementation of an image analysis algorithm;504
29.7;20.7 Board-level processing systems;508
29.8;20.8 VLSI;509
29.9;20.9 Concluding remarks;510
29.10;20.10 Bibliographical and historical notes;511
30;Part-4 Perspectives on Vision;514
31;Chapter 21. Machine Vision, Art or Science?;516
31.1;21.1 Introduction;516
31.2;21.2 Parameters of importance in machine vision;517
31.3;21.3 Tradeoffs;519
31.4;21.4 Future directions;522
31.5;21.5 Hardware, algorithms and processes;524
31.6;21.6 A retrospective view;524
31.7;21.7 Just a glimpse of vision?;526
31.8;21.8 Bibliographical and historical notes;526
32;Appendix:
Programming Notation;528
32.1;A.1 Introduction;528
32.2;A.2 The Pascal language;529
32.3;A.3 Special syntax embedded in Pascal;535
32.4;A.4 On the validity of the "repeat until finished" construct;539
33;References;540
34;Subject Index;556
35;Author Index;567