Pattern Recognition and Artificial Intelligence, Towards an Integration (eBook)

Proceedings of an International Workshop held in Amsterdam, May 18-20, 1988

E.S. Gelsema, L.N. Kanal (Herausgeber)

eBook Download: PDF

2014 | 1. Auflage
510 Seiten
Elsevier Science (Verlag)
978-1-4832-9945-7 (ISBN)

Infos

This volume brings together the results of research into the methodology and applications of pattern recognition, with particular emphasis given to the incorporation of artificial intelligence methodologies into pattern recognition systems.The first part of this volume covers image analysis and processing software, systems and algorithms. Pattern analysis and classifier design are dealt with in part two, while the last part deals with model based and expert systems, including uncertainty calculus methods in pattern analysis and object recognition. A number of specific application areas are considered, including such diverse topics as fingerprinting, astronomy, molecular biology and pathology.

Inhalt

Front Cover 1
Pattern Recognition and Artificial Intelligence: Towards an Integration 4
Copyright Page 5
Table of Contents 14
PREFACE 6
ACKNOWLEDGEMENTS 12
PART I: IMAGE PROCESSING 18
SECTION I: SYSTEMS 20
CHAPTER 1. ACUITY: IMAGE ANALYSIS FOR THE PERSONAL COMPUTER 22
1. Introduction 22
2. Using Acuity 24
3. Preparing an Experiment 24
4. Examining the Resulting Data 30
5. Exporting Data from an Experiment 31
6. Summary 32
References 32
CHAPTER 2. LILY: A SOFTWARE PACKAGE FOR IMAGE PROCESSING 34
1. INTRODUCTION 34
2. PROGRAMMING LANGUAGE CHOICE AND IT'S CONSEQUENCES 35
3. STRUCTURE OF THE PACKAGE 35
4. DATA REPRESENTATION, STRUCTURES AND SYNTAX 38
5. DOCUMENTATION 42
6. MAINTENANCE AND DEVELOPMENT TOOLS 44
7. IMAGE PROCESSING APPLICATIONS 44
8. CONCLUSION 48
REFERENCES 50
CHAPTER 3. REAL TIME PROCESSING OF IMAGES 52
1. INTRODUCTION 52
2. DIFFERENT REAL-TIME REQUIREMENTS 52
3. SYSTEMS WITH OPTIMAL RESPONSE-TIME 53
4. SYSTEMS WITH OPTIMAL THROUGHPUT 57
5. CONCLUSION 58
REFERENCES 58
DISCUSSIONS PART I , SECTION I 60
SECTION II: ALGORITHMS 64
CHAPTER 4. A NEW PROCEDURE FOR LINE ENHANCEMENT APPLIED TO FINGERPRINTS 66
1. ROTATION–INVARIANT OPERATORS 66
2. LINE DETECTION 69
3. THE FINGERPRINT APPLICATION 75
CONCLUSIONS 77
REFERENCES 77
CHAPTER 5. AN EDGE DETECTION MODEL BASED ON NON-LINEAR LAPLACEFILTERING 80
1. INTRODUCTION 80
2. EDGE DETECTION SCHEME 81
3. EVALUATION PROCEDURE 85
4. EXPERIMENTAL RESULTS 86
5. COMPARISONS 88
6. CONCLUSIONS 88
ACKNOWLEDGEMENT 89
REFERENCES 90
CHAPTER 6. PATTERN RECOGNITION BY DETECTION OF LOCAL SYMMETRIES 92
1 INTRODUCTION 92
2 MODELING THE LOCAL NEIGHBOURHOODS BY HARMONIC FUNCTIONS 93
3 DETECTION OF LOCAL SYMMETRIES 98
4 APPLICATIONS AND EXPERIMENTS 99
5 CONCLUSION 106
ACKNOWLEDGEMENTS 106
REFERENCES 107
CHAPTER 7. ACCURATE MEASUREMENT OF SHAPE AT LOW RESOLUTION 108
1. INTRODUCTION 108
2. THE METHOD FOR HIGH ACCURACY 111
3. CONCLUSION 118
ACKNOWLEDGEMENT 118
FOOTNOTES AND REFERENCES 118
CHAPTER 8. COMPUTING VISIBILITY PROPERTIES OF POLYGONS 120
Abstract 120
1. INTRODUCTION 120
2. VISIBILITY FROM A POINT 120
3. VISIBILITY FROM AN EDGE 121
4. DETERMINING THE VISIBILITY REGION FROM AN EDGE 123
5. DETECTING THE VISIBILITY OF A POLYGON FROM AN EDGE 126
6. DETECTING VISIBILITY BETWEEN TWO EDGES OF A POLYGON 129
REFERENCES 136
CHAPTER 9. USING VANISHING POINTS TO LOCATE OBJECTS WITH SIX DEGREES OF FREEDOM 140
1. INTRODUCTION 140
2. THE PROBLEM STATEMENT 143
3. SOLUTION USING VANISHING POINTS 145
4. ERROR ANALYSIS 148
5. PRELIMINARY EXPERIMENTAL RESULTS 152
6. CONCLUSION 156
ACKNOWLEDGMENTS 156
REFERENCES 156
CHAPTER 10. GRAPH CONSTRUCTION AND MATCHING FOR 3D OBJECT RECOGNITION 158
1. INTRODUCTION 158
2. PASSIVE APPROACH 159
3. OVERVIEW OF IMPLEMENTATION 159
4. 2D GRAPH BUILDING 160
5. PRIMITIVE MATCHING AND 3D TRANSFORMS 162
6. CONCLUSIONS 170
REFERENCES 171
CHAPTER 11. THREE–DIMENSIONAL RECONSTRUCTION OF MYOCARDIAL CONTRAST PERFUSION FROM BIPLANE CINEANGIOGRAMS. 172
1. INTRODUCTION 172
2. THREE-DIMENSIONAL RECONSTRUCTION BY MEANS OF LINEAR PROGRAMMING TECHNIQUES 173
3. DETERMINATION OF THE OPTIMAL BIPLANE ANGIOGRAPHIC VIEWS 174
4. MYOCARDIAL PERFUSION IMAGE ACQUISITION 175
5. DIGITIZATION AND PREPROCESSING OF THE SELECTED IMAGES 175
6. RECONSTRUCTION OF GEOMETRY 176
7. RECONSTRUCTION OF REGIONAL MYOCARDIAL PERFUSION 179
8. FIRST EXPERIMENTAL RESULTS 181
9. CONCLUSIONS 183
ACKNOWLEDGEMENTS 184
REFERENCES 184
CHAPTER 12. AUTOMATED CENTERLINE TRACING IN CORONARY ANGIOGRAMS 186
1. INTRODUCTION 186
2. METHODOLOGY 188
3. POSTPROCESSING 191
4. VALIDATION PROCEDURE 192
5. RESULTS 194
6. REPRODUCIBILITY 196
7. DISCUSSION 198
ACKNOWLEDGEMENTS 199
REFERENCES 199
CHAPTER 13. SHAPE ESTIMATION IN COMPUTER TOMOGRAPHY FROM MINIMALDATA 202
1. INTRODUCTION 202
2. THEORY 203
3. ERROR ANALYSIS FOR AREA ESTIMATIONS 209
4. EXPERIMENTAL RESULTS 210
5. CONCLUSION 214
ACKNOWLEDGEMENT 217
REFERENCES AND NOTES 217
DISCUSSIONS PART I, SECTION II 218
PART II: PATTERN RECOGNITION 226
Chapter 14. Classifier Design with Parzen Windows 228
1. INTRODUCTION 228
2. CLASSIFIER DESIGN IN PRACTICE 229
3. PARZEN WINDOW DENSITY ESTIMATES 232
4. PERFORMANCE OF CLASSIFIERS ON REAL DATA SETS 240
5. CONCLUSIONS 241
6. REFERENCES 244
CHAPTER 15. DISCRIMINANT ANALYSIS IN A NON-PROBABILISTIC CONTEXTBASED ON FUZZY LABELS 246
1. Problem definition 246
2. The choice of the error criterion 247
3. Classification and error estimation 248
4. Class separation and feature selection 250
5. Discussion 250
6. Conclusion and summary 251
7. References 251
CHAPTER 16. INCOMPLETE DATA SETS 254
1. Introduction 254
2. Classification of an object with missing values 255
3. Design of a classifier using an incomplete data set 256
4. Estimation of missing values 258
5. Experiments 261
6. Final remarks 270
REFERENCES 271
CHAPTER 17. A Structural Look at Pattern Recognition From the Point of View of Rate-Distortion Theory 274
ABSTRACT 274
1.0 Introduction 274
2.0 Rate - Distortion Theory 275
3.0 Modeling of the Pattern Recognition Process 279
4.0 Two-Class Pattern Recognition 281
5.0 Correlated Patterns 288
6.0 Conclusions 291
References: 291
CHAPTER 18. MAPPING TECHNIQUES FOR EXPLORATORY PATTERN ANALYSIS 294
1. INTRODUCTION 294
2. WHAT ARE MAPPING METHODS AND HOW ARE THEY USED? 295
3. OVERVIEW 297
4. EXPERIMENTAL COMPARISON 308
5. CONCLUSIONS 315
BIBLIOGRAPHY 315
CHAPTER 19. A MODEL FOR THE CLASSIFICATION OF CYTOLOGICAL SPECIMENS 318
1. Introduction 318
2. Cell-class and cell-feature approaches 319
3. The inclusion of between-specimen variability: a hierarchicalmodel 320
4. A two-level compound model for specimen classification 322
5. Cell-class approach: a model allowing a random proportion ofabnormal cells 324
6. Cell-feature approach: specimen discriminant analysis 326
7. Conclusions 330
8. References 331
CHAPTER 20. Astronomical Object Classification 334
1. INTRODUCTION 334
2. CURRENT PROBLEMS IN AUTOMATED ASTRONOMICAL CLASSIFICATION 335
3. CONCLUSIONS 341
References 343
DISCUSSIONS PART II 346
PART III: ARTIFICIAL INTELIGENCEAND PATTERN RECOGNITION 350
CHAPTER 21. OF BRITTLENESS AND BOTTLENECKS:CHALLENGES IN THE CREATION OF PATTERN-RECOGNITIONAND EXPERT-SYSTEM MODELS 352
1. INTRODUCTION 353
2. THE KNOWLEDGE-ACQUISITION BOTTLENECK 354
3. BRITTLENESS 357
4. DOMAIN MODELING IN PATTERN RECOGNITION 359
5. THE ANALOG OF BRITTLENESS 361
6. DISCUSSION 362
ACKNOWLEDGMENTS 366
REFERENCES 367
CHAPTER 22. Constructing Alternate Preferred Lines of Reasoning inInconsistent Knowledge Environments 370
Abstract 370
1. Introduction 370
2. Representation 371
3. Knowledge of Uncertainty 372
4. Lines of Reasoning 373
5. Knowledge of Causality 374
6. The Methodology 375
7. Search for the Model-trees 378
8. Extensions of the Method 380
9. Conclusion 382
Bibliography 383
CHAPTER 23. AN ANALYSIS OF FIVE STRATEGIES FOR REASONING IN UNCERTAINTIES AND THEIR SUITABILITY FOR PATHOLOGY 384
1. INTRODUCTION 385
2. DEFINITIONS AND COMBINATORICS 386
3. ANALYSIS 388
4. CONCLUSION 393
ACKNOWLEDGEMENTS 395
REFERENCES 396
CHAPTER 24. COMBINING THE CLASSIFICATION RESULTS OF INDEPENDENT CLASSIFIERS BASED ON THE DEMPSTER/SHAFER THEORY OF EVIDENCE 398
1. INTRODUCTION 398
2. COMBINATION PRINCIPLE 399
3. IMPLEMENTATION 400
4. CLASSIFICATION RESULTS 408
5. CONCLUSION 410
REFERENCES 410
CHAPTER 25. CLUSAN: A KNOWLEDGE BASE FOR APPROXIMATE REASONING IN EXPLORATORY DATA ANALYSIS 412
1. INTRODUCTION 412
2. THE CLUSAN/DELFI EXPERT SYSTEM SET-UP 414
3. LOW-LEVEL CLUSTERING TENDENCY 417
4. PROBLEM EXAMPLES 420
5. CONCLUSIONS 427
REFERENCES 427
CHAPTER 26. A TWO STEPS DECISION METHOD 430
ABSTRACT. 430
1. DECISION SURFACES TO DECIDE FOR CLASSES WITHIN A CATEGORY 430
2. DECISION BETWEEN CATEGORIES 437
3. CONCLUSION 439
REFERENCES. 440
CHAPTER 27. A KNOWLEDGE-BASED SYSTEM FOR THETHREEDIMENSIONAL RECONSTRUCTION OF THE CEREBRAL BLOOD VESSELS FROM A PAIR OF STEREOSCOPIC ANGIOGRAMS 442
1. INTRODUCTION 442
2. THE FIRST DELINEATE, THEN MATCH PARADIGM 443
3. BLOOD VESSEL SEGMENT DELINEATION 444
4. HIGH LEVEL FEATURE MATCHING 445
5. CONCLUSION 449
6. ACKNOWLEDGMENTS 452
REFERENCES 452
CHAPTER 28. PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE IN MOLECULAR BIOLOGY 454
1. INTRODUCTION 454
2. THREE DIMENSIONAL STRUCTURE DETERMINATION 455
3. PRIMARY STRUCTURE (SEQUENCE) ANALYSES 458
4. PREDICTION OF PROTEIN STRUCTURE FROM SEQUENCE 459
5. PLANNING MOLECULAR GENETICS EXPERIMENTS 460
6. CONCLUDING REMARKS 461
ACKNOWLEDGEMENTS 461
REFERENCES 461
CHAPTER 29. HYPOTHESIS COMBINATION AND CONTEXT SENSITIVE CLASSIFICATION FOR CHROMOSOME ABERRATION SCORING 466
1. INTRODUCTION 466
2. CENTROMERE CANDIDATE HYPOTHESIS GENERATION 470
3. CENTROMERE CANDIDATE FEATURE MEASUREMENT 471
4. CENTROMERE CANDIDATE CLASSIFICATION 472
5. DATA SET COLLECTION AND TRAINING 474
6. RESULTS 474
7. DISCUSSION 475
ACKNOWLEDGEMENTS 476
REFERENCES 476
CHAPTER 30. AN EXPERT SYSTEM APPROACH TO THE IDENTIFICATION AND CATEGORISATION OF FEATURES OF BIOLOGICAL IMAGES 478
INTRODUCTION 478
POTENTIAL VALUE OF EXPERT SYSTEMS 479
EXAMPLES OF PROJECTS REQUIRING AN EXPERT SYSTEMS APPROACH 479
PROCEDURE FOR FEATURE RECOGNITION OF NEMATODE SECTIONS 483
DESIGN OF THE EXPERT SYSTEM 486
ACKNOWLEDGEMENTS 486
REFERENCES 486
CHAPTER 31. A Coupled Expert System for Automated Signal Interpretation 488
1. INTRODUCTION 488
2. SYSTEM DESCRIPTION 489
3. SYSTEM'S OPERATION 492
4. APPLICATIONS TO EEG DATA 495
5. DISCUSSION AND CONCLUSIONS 497
ACKNOWLEDGEMENTS 498
REFERENCES 498
DISCUSSIONS PART III 500
AUTHOR INDEX 510
SUBJECT INDEX 512

Produkt-Details

• Erscheint lt. Verlag: 28.6.2014
• Sprache: englisch
• Themenwelt: Informatik ► Grafik / Design ► Digitale Bildverarbeitung
• Themenwelt: Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik
• ISBN-10: 1-4832-9945-7 / 1483299457
• ISBN-13: 978-1-4832-9945-7 / 9781483299457