Rosenfeld Image Modeling

1. Auflage 2014
ISBN: 978-1-4832-7560-4
Verlag: Elsevier Science & Techn.
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 460 Seiten, Web PDF

Image Modeling compiles papers presented at a workshop on image modeling in Rosemont, Illinois on August 6-7, 1979. This book discusses the mosaic models for textures, image segmentation as an estimation problem, and comparative analysis of line-drawing modeling schemes. The statistical models for the image restoration problem, use of Markov random fields as models of texture, and mathematical models of graphics are also elaborated. This text likewise covers the univariate and multivariate random field models for images, stochastic image models generated by random tessellations of the plane, and long crested wave models. Other topics include the Boolean model and random sets, structural basis for image description, and structure in co-occurrence matrices for texture analysis. This publication is useful to specialists and professionals working in the field of image processing.

Autoren/Hrsg.

Rosenfeld, Azriel

Inhaltsverzeichnis

1;Front Cover;1
2;Image Modeling;4
3;Copyright Page;5
4;Table of Contents;6
5;List of Contributors;10
6;Preface;14
7;Chapter 1.
Mosaic Models for Textures;16
7.1;1. MOSAIC MODELS;16
7.2;2. PROPERTIES OF MOSAIC MODELS;19
7.3;REFERENCES;22
8;Chapter 2.
Image Segmentation as an Estimation Problem;24
8.1;1. INTRODUCTION;24
8.2;2. OVERALL STRATEGY;25
8.3;3. REGION TESTING;26
8.4;4. EVALUATION OF INTRINSIC BOUNDARY AMBIGUITY;35
8.5;5. APPROXIMATE BOUNDARY;38
8.6;6. EXPERIMENTAL RESULTS;39
8.7;7. CONCLUSIONS;42
8.8;REFERENCES;42
9;Chapter 3. Toward a Structural Textural Analyzer Based on Statistical Methods;44
9.1;1. INTRODUCTION;44
9.2;2. BACKGROUND;46
9.3;3. A MODEL FOR TEXTURE BASED ON MATHEMATICAL TILING;50
9.4;4. FORMALIZING THE CONCEPT OF PERIOD PARALLELOGRAM
UNIT PATTERNS;57
9.5;5. FINDING THE SIZE, SHAPE, AND ORIENTATION OF A PERIOD PARALLELOGAM UNIT PATTERN OF APERIODIC TEXTURE;61
9.6;6. FURTHER PROPERTIES OF THE INERTIA MEASURE;68
9.7;7. CONCLUSIONS;74
9.8;REFERENCES;75
10;Chapter 4.
Stochastic Boundary Estimation and Object Recognition;78
10.1;1. INTRODUCTION;78
10.2;2. BOUNDARY ESTIMATION AS LIKELIHOOD MAXIMIZATION;81
10.3;3. THE RIPPLE FILTER: A REGION GROWING/SHRINKING
BOUNDARY ESTIMATOR;86
10.4;4. SEQUENTIAL BOUNDARY FINDER;94
10.5;5. MORE ON BOUNDARY MODELS;98
10.6;6. BOUNDARY ERROR ESTIMATION;100
10.7;7. OBJECT RECOGNITION;104
10.8;8. COMMENTS;106
10.9;REFERENCES;108
11;Chapter 5.
Edge Detection in Textures;110
11.1;1. INTRODUCTION;110
11.2;2. TEXTURE MODELS;111
11.3;3. A ONE-DIMENSIONAL EDGE DETECTOR;112
11.4;4. ANALYSIS OF ek;113
11.5;5. DISCUSSION;123
11.6;REFERENCES;123
12;Chapter 6.
Comparative Analysis of Line-Drawing Modeling Schemes;126
12.1;1. INTRODUCTION;126
12.2;2. LINE-DRAWING QUANTIZATION;127
12.3;3. APPROXIMANTS;128
12.4;4. QUANTIZATION;131
12.5;5. LINK PROBABILITIES;135
12.6;6. EXPERIMENTAL RESULTS;142
12.7;7. CONCLUSION;144
12.8;APPENDIX : Line Segments in an m X n Lattice Field;144
12.9;REFERENCES;145
13;Chapter 7.
Statistical Models for the Image Restoration Problem;148
13.1;1. INTRODUCTION;148
13.2;2. AN OBJECT IS POSITIVE;150
13.3;3. SOME OBJECTS ARE BOUNDED ABOVE AND BELOW;153
13.4;4. SOME OBJECTS ARE POWER SPECTRA;155
13.5;5. THE RECONCILIATION MODEL OF KIKUCHI AND SOFFER;156
13.6;6. SOME OBJECTS ARE SIMPLY CONNECTED;158
13.7;7. AN OBJECT IS A PROBABILITY LAW: MAXIMUM INFORMATION RESTORATION;162
13.8;8. SUMMARY;166
13.9;REFERENCES;167
14;Chapter 8.
Syntactic Image Modeling Using Stochastic Tree Grammars;168
14.1;1. INTRODUCTION;168
14.2;2. TREE GRAMMARS AND STOCHASTIC TREE GRAMMARS;168
14.3;3. APPLICATION OF TREE GRAMMARS TO IMAGE MODELING;177
14.4;4. CONCLUDING REMARKS;182
14.5;REFERENCES;184
15;Chapter 9.
Edge and Region Analysis for Digital Image Data;186
15.1;1. INTRODUCTION;186
15.2;2. THE SLOPED-FACET MODEL;187
15.3;3. SLOPED FACET PARAMETER ESTIMATION AND SIGNIFICANCE MEASURE;188
15.4;4. USING THE SLOPED-FACET MODEL;194
15.5;5. LITERATURE REVIEW;195
15.6;6. BAYESIAN EDGE DETECTION AND REGION ANALYSIS;197
15.7;7. CONCLUSION;199
15.8;REFERENCES;199
16;Chapter 10.
The Use of Markov Random Fields as Models of Texture;200
16.1;1. MRFS AS STATISTICAL MODELS OF TEXTURE;200
16.2;2. AN MRF SIMULATION ALGORITHM;206
16.3;3. MRF PARAMETER ESTIMATION;210
16.4;ACKNOWLEDGMENT;213
16.5;REFERENCES;213
17;Chapter 11.
On the Noise in Images Produced by Computed Tomography;214
17.1;1. IMAGE RECONSTRUCTION FROM PROJECTIONS;214
17.2;2. NOISE IN IMAGE RECONSTRUCTION;219
17.3;3. ILLUSTRATIONS;220
17.4;4. DISCUSSION;224
17.5;ACKNOWLEDGMENTS;228
17.6;REFERENCES;228
18;Chapter 12.
Mathematical Models of Graphics;230
18.1;1. INTRODUCTION;230
18.2;2. MATHEMATICAL FRAMEWORK;230
18.3;3. JOINT PROBABILITY MODEL;231
18.4;4. CONDITIONAL PROBABILITY MODEL;233
18.5;5. CONTOUR MODELS;235
18.6;6. PATTERN RECOGNITION MODELS;236
18.7;7. DEPENDENCE ON RESOLUTION;237
18.8;8. PROSPECTS;237
18.9;ACKNOWLEDGMENT;237
18.10;REFERENCES;237
19;Chapter 13.
Nonstationary Statistical Image Models (and Their Application to Image Data Compression);240
19.1;1. INTRODUCTION;240
19.2;2. CONVENTIONAL STATISTICAL IMAGE MODELS;241
19.3;3. NONSTATIONARY STATISTICAL IMAGE MODELS;243
19.4;4. TRANSFORMATION TO STATIONARY BEHAVIOR;245
19.5;5. A
SYSTEM FOR PROCESSING NONSTATIONARY IMAGES;247
19.6;6. EXAMPLES OF APPLICATIONS OF NONSTATIONARY MODELS;247
19.7;7. CLOSING REMARKS;253
19.8;REFERENCES;253
20;Chapter 14.
Markov Mesh Models;254
20.1;1. INTRODUCTION;254
20.2;2. THE MARKOV MESH;254
20.3;3. SOME COMMENTS ON RELATIONS TO OTHER CONTEMPORARY
AND LATER REFERENCES;256
20.4;ACKNOWLEDGMENT;258
20.5;REFERENCES;258
21;Chapter 15.
Univariate and Multivariate Random Field Models for Images;260
21.1;1. INTRODUCTION;260
21.2;2. PERIODIC UNIVARIATE RANDOM FIELD;261
21.3;3. PARAMETER ESTIMATION;265
21.4;4. CHOICE OF APPROPRIATE NEIGHBORS;266
21.5;5. SEGMENTATION OF AN IMAGE;267
21.6;6. IMAGE COMPRESSION AND RESTORATION;268
21.7;7. MULTIVARIATE RANDOM FIELD;269
21.8;8. CONCLUSIONS;271
21.9;APPENDIX 1;272
21.10;REFERENCES;273
22;Chapter 16.
Image Models in Pattern Theory;274
22.1;1. INTRODUCTION;274
22.2;2. ELEMENTS OF PATTERN THEORY;274
22.3;3. GENERAL IMAGE MODELS;279
22.4;4. EXAMPLES;283
22.5;REFERENCES;290
23;Chapter 17.
A Survey of Geometrical Probability in the Plane, with Emphasis on Stochastic image Modeling;292
23.1;1. INTRODUCTION;292
23.2;2. MOBILE GEOMETRICAL OBJECTS;293
23.3;3. MEASURES AND INVARIANT MEASURES;295
23.4;4. A SINGLE RANDOM MOBILE OBJECT;297
23.5;5. TWO OR MORE RANDOM MOBILE OBJECTS;300
23.6;6. TWO STANDARD TECHNIQUES;304
23.7;7. POISSON MODELS;307
23.8;8. EXPECTED NUMBERS AND ERGODIC DISTRIBUTIONS OF n-FIGURES;309
23.9;9. ASSORTED TOPICS;310
23.10;10. SQUARE LATTICE ANALOGS;313
23.11;REFERENCES;314
24;Chapter 18.
Stochastic Image Models Generated by Random Tessellations of the Plane;316
24.1;1. INTRODUCTION;316
24.2;2. PRELIMINARIES;317
24.3;3. CONSTRUCTION PROCEDURE;318
24.4;4. SECOND-ORDER PROPERTIES;327
24.5;5. APPLICATIONS;333
24.6;6. SUMMARY AND CONCLUSIONS;339
24.7;REFERENCES;339
25;Chapter 19.
Long Crested Wave Models;342
25.1;1. INTRODUCTION;342
25.2;2. THE STANDARD MODEL;342
25.3;3. NARROW-BAND NOISE MODEL;348
25.4;4. FITTING THE MODEL TO REAL DATA;351
25.5;5. GEOMETRICAL PROPERTIES;352
25.6;6. DISCUSSION;355
25.7;REFERENCES;355
26;Chapter 20.
The Boolean Model and Random Sets;358
26.1;NOTATION;358
26.2;A COUNTERPOINT;359
26.3;1. CONSTRUCTION OF THE BOOLEAN SETS;360
26.4;1.* RANDOM SETS: DEFINITION AND BASIC PROPERTIES;361
26.5;2. THE FUNCTIONAL MOMENT OF THE BOOLEAN MODEL;363
26.6;2.* INFINITE DIVISIBILITY;366
26.7;3. CONVEX PRIMARY GRAINS;366
26.8;3.* SEMI-MARKOV RACS;368
26.9;4. CONNECTIVITY NUMBER;370
26.10;4.* DIGITIZATION;372
26.11;5. SPECIFIC BOOLEAN MODELS;374
26.12;5.* ESTIMATION PROBLEMS;375
26.13;6. DERIVED MODELS;378
26.14;6.* THE ROSE OF MODELS;384
26.15;REFERENCES;384
27;Chapter 21.
Scene Modeling: A Structural Basis for Image Description;386
27.1;1. INTRODUCTION;386
27.2;2. STATISTICAL VERSUS STRUCTURAL COMPLEXITY;387
27.3;3. NECESSITY FOR STRUCTURAL MODELS;387
27.4;4. A PARADIGM FOR STRUCTURAL MODELING;388
27.5;5. DISCUSSION;403
27.6;ACKNOWLEDGMENTS;404
27.7;REFERENCES;404
28;Chapter 22.
Pictorial Feature Extraction and Recognition via Image Modeling;406
28.1;1. INTRODUCTION;406
28.2;2. TWO-DIMENSIONAL TIME SERIES MODEL;408
28.3;3. G MATRIX EIGENVALUE APPROACH;421
28.4;4. PIXEL-VECTOR CLUSTERING TECHNIQUE;432
28.5;ACKNOWLEDGMENT;435
28.6;REFERENCES;435
29;Chapter 23.
Finding Structure in Co-Occurrence Matrices for Texture Analysis;438
29.1;1. INTRODUCTION;438
29.2;2. CO-OCCURRENCE MATRICES FOR TEXTURE CLASSIFICATION;439
29.3;3. CONTINGENCY TABLES AND x2 SIGNIFICANCE TESTS;440
29.4;4. AN
ALGORITHM FOR SELECTING CO-OCCURRENCE MATRICES FOR TEXTURE CLASSIFICATION;445
29.5;5. EXPERIMENTS;449
29.6;6. DISCUSSION AND CONCLUSIONS;454
29.7;APPENDIX: ALINEAR DISCRIMINANT CLASSIFIER;456
29.8;REFERENCES;459