Mehnen / Köppen / Koeppen Applications of Soft Computing

From Theory to Praxis
1. Auflage 2009
ISBN: 978-3-540-89619-7
Verlag: Springer Berlin Heidelberg
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 58, 536 Seiten

Reihe: Advances in Intelligent and Soft Computing

WSC2008Chair's Welcome Message Dear Colleague, The World Soft Computing (WSC) conference is an annual international online conference on applied and theoretical soft computing technology. This WSC 2008 is the thirteenth conference in this series and it has been a great success. We received a lot of excellent paper submissions which were peer-reviewed by an international team of experts. Only60 papers out of111 submissions were selected for online publication. This assured a high quality standard for this online conference. The corresponding online statistics are a proof of the great world-wide interest in the WSC 2008 conference. The conference website had a total of33,367di?erent human user accessesfrom43 countries with around100 visitors every day,151 people signed up to WSC to discuss their scienti?c disciplines in our chat rooms and the forum. Also audio and slide presentations allowed a detailed discussion of the papers. The submissions and discussions showed that there is a wide range of soft computing applications to date. The topics covered by the conference range from applied to theoretical aspects of fuzzy, neuro-fuzzy and rough sets over to neural networks to single and multi-objective optimisation. Contributions aboutparticleswarmoptimisation,geneexpressionprogramming,clustering, classi?cation,supportvectormachines,quantumevolutionandagentsystems have also been received. One whole session was devoted to soft computing techniques in computer graphics, imaging, vision and signal processing.

Autoren/Hrsg.

Mehnen, Jörn

Köppen, Mario

Koeppen, Mario

Saad, Ashraf

Tiwari, Ashutosh

Inhaltsverzeichnis

1;Preface;6
2;Organization;8
3;Contents;13
4;List of Contributors;19
5;Part I Fuzzy, Neuro-Fuzzy and Rough Sets Applications;32
5.1;Fuzzy Group Decision Making for Management of Oil Spill Responses;33
5.1.1;Introduction;33
5.1.2;Fuzzy Decision Making Methods;35
5.1.3;Case Study;35
5.1.4;Fuzzy Evaluation Method;37
5.1.4.1;First Order Fuzzy Evaluation Method;38
5.1.4.2;Second Order Fuzzy Evaluation Method;38
5.1.5;Results;40
5.1.6;Conclusions;41
5.1.7;References;41
5.2;Sensor Fusion Map Building-Based on Fuzzy Logic Using Sonar and SIFT Measurements;43
5.2.1;Introduction;43
5.2.2;Sensor Models;44
5.2.2.1;Sonar Model;44
5.2.2.2;Vision-SIFT-Descriptor Model;45
5.2.2.3;SIFT-Descriptor Model;45
5.2.3;Sensor Fusion Based on Fuzzy Set Theory;47
5.2.3.1;Dombi Operator;47
5.2.3.2;Fuzzy Maps from Sensor Measurements;47
5.2.4;Map Building Experimental Results;48
5.2.5;Conclusion;51
5.2.6;References;51
5.3;Rough Sets in Medical Informatics Applications;53
5.3.1;Introduction;53
5.3.2;Rough Set Theory;54
5.3.3;Rough Sets in Medical Image Segmentation;55
5.3.4;Rough Sets in Medical Classification;56
5.3.5;Rough Sets in Medical Data Mining;57
5.3.6;Rough Sets in Medical Decision Support Systems;58
5.3.7;Conclusions;59
5.3.8;References;59
5.4;A Real Estate Management System Based on Soft Computing;61
5.4.1;Introduction;61
5.4.2;Soft Data Server (SDS);62
5.4.2.1;Fuzzy Data Comparison;65
5.4.3;ImmoSoftDataServerWeb (ISDSW);65
5.4.3.1;Features;66
5.4.3.2;Fuzzy Attributes in the Application;66
5.4.3.3;Geographic Search;67
5.4.4;System Architecture;68
5.4.5;Concluding Remarks and Future Work;69
5.4.6;References;70
5.5;Proportional Load Balancing Using Scalable Object Grouping Based on Fuzzy Clustering;71
5.5.1;Introduction;71
5.5.2;Background and Related Works;72
5.5.3;Scalable Object Grouping Based on Fuzzy Clustering;74
5.5.4;Load Distribution Based on Equal Proportions;75
5.5.5;Experimental Evaluation;76
5.5.5.1;Simulation Result;77
5.5.6;Conclusions and Future Work;78
5.5.7;References;79
6;Part II Neural Network Applications;81
6.1;Multilevel Image Segmentation UsingOptiMUSIG Activation Function with Fixed andVariable Thresholding: A Comparative Study;82
6.1.1;Introduction;82
6.1.2;Multilayer Self-Organizing Neural Network (MLSONN);84
6.1.3;Optimized Multilevel Sigmoidal (OptiMUSIG) Activation Function;84
6.1.4;Principle of MLSONN Based Optimized Multilevel Image Segmentation;86
6.1.5;Results;87
6.1.5.1;Performance Evaluation of OptiMUSIG Activation Function;90
6.1.6;Discussions and Conclusion;90
6.1.7;References;90
6.2;Artificial Neural Networks Modeling to Reduce Industrial Air Pollution;92
6.2.1;Introduction;92
6.2.2;A Brief Description of the Nitric Acid Plant;93
6.2.3;A Short Overview of Artificial Neural Networks Modeling;94
6.2.4;ANN Model Training and Analyzing;95
6.2.5;Discussion;98
6.2.6;Conclusions;99
6.2.7;References;99
6.3;Wavelet Neural Network as a Multivariate Processing Tool in Electronic Tongues;101
6.3.1;Introduction;101
6.3.2;Theory;102
6.3.2.1;WNN Algorithm;103
6.3.3;Application in Chemical Sensing;106
6.3.4;Results and Discussion;106
6.3.5;Conclusions;107
6.3.6;References;108
6.4;Design of ANFIS Networks Using Hybrid Genetic and SVD Method for the Prediction of Coastal Wave Impacts;110
6.4.1;Introduction;110
6.4.2;Study Area;111
6.4.3;Basis of Design and Measuring Manner of Sea Wave Impact Tester;112
6.4.4;Modelling Using ANFIS;112
6.4.5;Application of Genetic Algorithm to the Design of ANFIS;114
6.4.6;Application of Singular Value Decomposition to the Design of ANFIS Networks;115
6.4.7;Genetic/SVD Based ANFIS Prediction of Coastal Wave Impacts;115
6.4.8;Conclusion;118
6.4.9;References;118
6.5;A Neuro-Fuzzy Control for TCP Network Congestion;120
6.5.1;Introduction;120
6.5.2;Improved Neural Network AQM;121
6.5.3;ANFIS AQM;122
6.5.4;Simulations and Results;124
6.5.4.1;Simulation 1;124
6.5.4.2;Simulation 2;125
6.5.4.3;Simulation 3;125
6.5.5;Conclusions;127
6.5.6;References;128
6.6;Use of Remote Sensing Technology for GIS Based Landslide Hazard Mapping;129
6.6.1;Introduction;130
6.6.1.1;Digital Techniques for Landslide Change Detection;131
6.6.1.2;GIS Data Analysis and Modeling for Landslide Risk Assessment;131
6.6.2;Study Area;132
6.6.3;Data Requirements;132
6.6.4;Construction of Spatial Database Using GIS;132
6.6.5;The Artificial Neural Network;133
6.6.6;Landslide Susceptibility Forecast Mapping and Verification;135
6.6.7;The Analytical Hierarchic Process;136
6.6.8;Conclusion;138
6.6.9;References;138
6.7;An Analysis of the Disturbance on TCP Network Congestion;140
6.7.1;Introduction;140
6.7.2;Dynamics of TCP/AQM Networks;141
6.7.3;Improved Neural Network AQM;142
6.7.4;Simulations and Results;143
6.7.4.1;Simulation 1;143
6.7.4.2;Simulation 2;144
6.7.4.3;Simulation 3;146
6.7.4.4;Simulation 4;146
6.7.5;Conclusions;147
6.7.6;References;148
7;Part III Applications of Evolutionary Computations;149
7.1;RAM Analysis of the Press Unit in a Paper Plant Using Genetic Algorithm and Lambda-Tau Methodology;150
7.1.1;Introduction;150
7.1.2;GABLT Technique;152
7.1.3;RAM Index;154
7.1.4;An Illustration with Application;154
7.1.5;Discussion and Conclusion;157
7.1.6;Managerial Implications Drawn;159
7.1.7;References;159
7.2;A Novel Approach to Reduce High-Dimensional Search Spaces for the Molecular Docking Problem;161
7.2.1;Introduction;161
7.2.2;Motivation;162
7.2.3;Search Space Reduction;164
7.2.3.1;Computing the Protein’s Surface;164
7.2.3.2;Optimising along the Surface;166
7.2.4;Results;167
7.2.5;Conclusion and Outlook;169
7.2.6;References;169
7.3;GA Inspired Heuristic for Uncapacitated Single Allocation Hub Location Problem;171
7.3.1;Introduction;171
7.3.2;Mathematical Formulation;172
7.3.3;Previous Work;173
7.3.4;Proposed GA Heuristic Method;174
7.3.4.1;Representation;174
7.3.4.2;Objective Function;174
7.3.4.3;Genetic Operators;175
7.3.4.4;Other GA Aspects;176
7.3.5;Computational Results;176
7.3.6;Conclusions;179
7.3.7;References;180
7.4;Evolutionary Constrained Design of Seismically Excited Buildings: Sensor Placement;181
7.4.1;Introduction;181
7.4.2;Optimal Placement of Sensors;183
7.4.2.1;Defining the Optimization Problem;183
7.4.2.2;Constrained GA;184
7.4.2.3;Proposed Constrained GA;184
7.4.3;Benchmark Building;185
7.4.3.1;Sample LQG Control System;185
7.4.4;Simulation Results;186
7.4.5;Conclusion;189
7.4.6;References;189
7.5;Applying Evolution Computation Model to the Development and Transition of Virtual Community under Web2.0;192
7.5.1;Introduction;192
7.5.2;Related Work;193
7.5.2.1;Social Network and Virtual Community;193
7.5.2.2;Particle Swarm Intelligence;194
7.5.3;Proposed Model;195
7.5.3.1;Initia1 Individual;195
7.5.3.2;Link Construction;196
7.5.3.3;Evaluation;197
7.5.3.4;Self-adaption;197
7.5.4;Moving Path Simulation;198
7.5.4.1;Evolution Process;198
7.5.4.2;Overview Result;198
7.5.5;Conclusion;200
7.5.6;References;200
7.6;Genetic Algorithms in Chemistry: Success or Failure Is in the Genes;202
7.6.1;General Introduction to Genetic Algorithms in Chemistry;202
7.6.2;Types of Problems in Chemistry Where Genetic Algorithms Are Used;203
7.6.2.1;Genetic Regression;203
7.6.2.2;Structure Elucidation;205
7.6.2.3;Other Uses;208
7.6.3;Conclusions;208
7.6.4;References;209
8;Part IV Other Soft Computing Applications;211
8.1;Multi-objective Expansion Planning of Electrical Distribution Networks Using Comprehensive Learning Particle Swarm Optimization;212
8.1.1;Introduction;212
8.1.2;Problem Formulation;213
8.1.3;PSO, CLPSO and Particle Encoding/Decoding;214
8.1.3.1;CLPSO [12];215
8.1.3.2;Proposed Encoding Scheme;215
8.1.3.3;Proposed Decoding Scheme;216
8.1.4;Multi-objective Optimization;216
8.1.4.1;Pareto-Optimality Principle;217
8.1.4.2;Fitness Assignment and Elite Preserving Scheme;217
8.1.5;Complete Distribution System Expansion Planning Algorithm;217
8.1.6;Simulation Results and Performance Assessment;218
8.1.7;Conclusions;220
8.1.8;References;221
8.2;Prediction of Compressive Strength of Cement Using Gene Expression Programming;222
8.2.1;Introduction;222
8.2.2;Data Modeling Tools;223
8.2.2.1;Artificial Neural Networks;223
8.2.2.2;Fuzzy Logic Model;224
8.2.3;Background of Gene Expression Programming;224
8.2.3.1;Initialization of Population;224
8.2.3.2;Replication and Selection;225
8.2.3.3;Genetic Operators;225
8.2.3.4;GEP Algorithm;226
8.2.4;Data Collection;226
8.2.5;Model Construction and Results;228
8.2.6;Summary and Future Work;230
8.2.7;References;230
8.3;Fault-Tolerant Nearest Neighbor Classifier Based on Reconfiguration of Analog Hardware in Low Power Intelligent Sensor Systems;232
8.3.1;Introduction;232
8.3.2;Reconfigurable Hardware Implementation of Nearest Classifier;234
8.3.3;Optimization of Reconfigurable Prototypes;235
8.3.3.1;Particle Swarm Optimization;235
8.3.3.2;Objective Function and Gaussian Model;235
8.3.3.3;Optimization Procedure;236
8.3.4;Experiments and Results;237
8.3.5;Conclusion;240
8.3.6;References;241
8.4;Text Documents Classification by Associating Terms with Text Categories;242
8.4.1;Introduction;242
8.4.2;Related Work;243
8.4.2.1;Text Categorization;243
8.4.2.2;Association Rule Mining;243
8.4.3;Building an Associative Text Classifier;244
8.4.3.1;Association Rule Generation;244
8.4.3.2;Prediction of Classes Associated with New Documents;246
8.4.4;Experimental Result;247
8.4.4.1;Experiment Data;247
8.4.4.2;Experimental Results and Analysis;247
8.4.5;Conclusion and Future Work;249
8.4.6;References;249
8.5;Applying Methods of Soft Computing to Space Link Quality Prediction;251
8.5.1;Background;251
8.5.2;Specific Optimization Aims;252
8.5.2.1;Space Up- and Downlink Quality Identification and Determination;253
8.5.2.2;Identification of Correlations between Environmental Conditions and Space Link Quality;253
8.5.2.3;The Long-Term Impact of Climate Changes on Short-Range Satellite Communication;254
8.5.2.4;The Rapid Determination of Spacecraft Orbital Elements;254
8.5.2.5;The Short-Term Prediction of Space Communication Link Quality;255
8.5.2.6;Automated Identification of Imprecise Ground Stations;255
8.5.2.7;Automated Determination of Spacecraft Health and Orbit Changes;255
8.5.3;Data Mining Architecture;255
8.5.4;Prediction Model Development;258
8.5.5;Conclusion;259
8.5.6;References;259
8.6;A Novel Multicriteria Model Applied to Cashew Chestnut Industrialization Process;261
8.6.1;Introduction;261
8.6.2;The Industrial Process of Breaking Cashew Chestnuts;263
8.6.3;Approaching the Critical Processes;264
8.6.4;ZAPROS Method;265
8.6.5;Modeling the Industrialization Process of the Cashew Chestnut with ZAPROS;267
8.6.6;Conclusions;269
8.6.7;References;270
9;Part V Design of Fuzzy, Neuro-Fuzzy and Rough Sets Techniques;271
9.1;Selection of Aggregation Operators with Decision Attitudes;272
9.1.1;Introduction;272
9.1.2;Aggregation Operators;272
9.1.3;Categories of Aggregation Operators;273
9.1.3.1;Quasi-Linear Means;273
9.1.3.2;Ordered Weighted Averaging;274
9.1.3.3;Weighted Median;274
9.1.3.4;T-Conorm and T-Norm;274
9.1.3.5;Weighted Gamma Operator;275
9.1.3.6;OWMAX and OWMIN;275
9.1.3.7;Leximin Ordering;276
9.1.4;AODA Model;276
9.1.5;A Numerical Example;279
9.1.6;Conclusions;281
9.1.7;References;281
9.2;A New Approach Based on Artificial Neural Networks for High Order Bivariate Fuzzy Time Series;282
9.2.1;Introduction;282
9.2.2;Fuzzy Time Series;283
9.2.3;The Proposed Method;284
9.2.4;Application;286
9.2.5;Conclusion;289
9.2.6;References;289
9.3;A Genetic Fuzzy System with Inconsistent Rule Removal and Decision Tree Initialization;291
9.3.1;Introduction;291
9.3.2;Mamdani Fuzzy Models;292
9.3.3;Proposed Identification Method;293
9.3.3.1;Mamdani FM Initialization Using C4.5 Algorithm;293
9.3.3.2;Merging of Fuzzy Sets;294
9.3.3.3;Coding of the Mamdani FM;294
9.3.3.4;MOEA Optimization of the Initial Population;295
9.3.3.5;Heuristic Rule and Rule Condition Removal;295
9.3.4;Experiments;296
9.3.4.1;Datasets;296
9.3.4.2;Experimental Setup;297
9.3.4.3;MG, Lorenz and Gas Datasets: Results Comparison;297
9.3.4.4;Electric Data: Results Comparison;298
9.3.5;Conclusions;299
9.3.6;References;299
9.4;Robust Expectation Optimization Model Using the Possibility Measure for the Fuzzy Random Programming Problem;301
9.4.1;Introduction;301
9.4.2;Formulation of Fuzzy Random Programming Problem;303
9.4.2.1;Fuzzy Random Variables;303
9.4.2.2;Fuzzy Random Programming Problem;303
9.4.2.3;Expectation OptimizationModel Using a Possibility Measure;305
9.4.3;Robust Expectation Optimization Model;305
9.4.4;Numerical Example;308
9.4.5;Conclusion;309
9.4.6;References;309
9.5;Improving Mining Fuzzy Rules with Artificial Immune Systems by Uniform Population;311
9.5.1;Introduction;311
9.5.2;Buffered IFRAIS;312
9.5.3;Buffered IFRAIS with Uniform Population;315
9.5.4;Experimental Results;316
9.5.5;Conclusion;318
9.5.6;References;318
9.6;Incremental Locally Linear Fuzzy Classifier;320
9.6.1;Introduction;320
9.6.2;Local Linear Neuro-Fuzzy Classifier;321
9.6.2.1;Rule Consequent Parameters;322
9.6.2.2;Rule Antecedent Structure;323
9.6.3;Proposed Algorithm for Structure Optimization;324
9.6.4;Experiments;326
9.6.4.1;Comparison with Conventional Classifiers;326
9.6.4.2;Comparison with Piecewise Linear Classifiers;327
9.6.4.3;Comparison with Decision Tree Classifiers;328
9.6.5;Conclusions;328
9.6.6;References;328
9.7;On Criticality of Paths in Networks with Imprecise Durations and Generalized Precedence Relations;330
9.7.1;Introduction;330
9.7.2;Terminology and Representation;331
9.7.3;Necessary Criticality in Interval Networks with GPRs;333
9.7.4;Possible Criticality in Interval Networks with GPRs;333
9.7.4.1;Necessarily Non-critical Path;334
9.7.4.2;Possibly Critical Path;334
9.7.5;Criticality in Networks with Fuzzy Durations and GPRs;335
9.7.6;Conclusions;338
9.7.7;References;338
10;Part VI Design of Evolutionary Computation Techniques;340
10.1;Parallel Genetic Algorithm Approach to Automated Discovery of Hierarchical Production Rules;341
10.1.1;Introduction;341
10.1.2;Hierarchical Production Rules;342
10.1.2.1;Subsumption Relation;343
10.1.2.2;Coefficient of Similarity;343
10.1.3;Design of Genetic Algorithm;344
10.1.3.1;Encoding;344
10.1.3.2;Fitness Function;344
10.1.3.3;GA Operators;345
10.1.3.4;Parallel Genetic Algorithm Scheme;345
10.1.4;Experimental Setup and Results;346
10.1.5;Conclusions and Future Direction;348
10.1.6;References;349
10.2;Two Hybrid Genetic Algorithms for Solving the Super-Peer Selection Problem;351
10.2.1;Introduction;351
10.2.2;Mathematical Representation;352
10.2.3;Proposed Hybrid GA Methods;353
10.2.3.1;Description of HGA1 Representation;353
10.2.3.2;Description of HGA2 Representation;353
10.2.3.3;Objective Function;354
10.2.3.4;Selection;354
10.2.3.5;Crossover and Mutation;355
10.2.3.6;Caching GA;356
10.2.3.7;Other GA Aspects;356
10.2.4;Computational Results;357
10.2.5;Conclusions;359
10.2.6;References;359
10.3;A Genetic Algorithm for the Constrained Coverage Problem;361
10.3.1;Introduction;361
10.3.2;Voronoi Diagram;362
10.3.3;Constrained Space p-Center and Coverage Problem;363
10.3.4;Genetic Algorithm for Solving the Coverage Problem;364
10.3.4.1;Parameters and Operators in Genetic Algorithm;364
10.3.4.2;Genetic Algorithm;367
10.3.5;Simulation Results;368
10.3.6;Conclusion;369
10.3.7;References;369
10.4;Using Multi-objective Evolutionary Algorithms in the Optimization of Polymer Injection Molding;371
10.4.1;Introduction;371
10.4.2;Multi-objective Evolutionary Algorithms;372
10.4.3;Injection Molding Optimization;373
10.4.4;Results and Discussion;376
10.4.5;Conclusions;378
10.4.6;References;379
10.5;A Multiobjective Extremal Optimization Algorithm for Efficient Mapping in Grids;380
10.5.1;Introduction;380
10.5.2;Extremal Optimization;381
10.5.2.1;Multiobjective Extremal Optimization;382
10.5.3;Mapping in Grids;383
10.5.4;Experimental Results;384
10.5.5;Conclusions;388
10.5.6;References;389
10.6;Interactive Incorporation of User Preferences in Multiobjective Evolutionary Algorithms;390
10.6.1;Introduction;390
10.6.2;Interactive Preference Articulation;391
10.6.2.1;Pairwise Solution Comparison Scheme;392
10.6.2.2;Preference Estimation Based on Pairwise Similarity;392
10.6.2.3;Preference Controlled Selection Mechanism;393
10.6.2.4;Artificial Benchmark Preference Model;394
10.6.3;Results;396
10.6.3.1;Interactive Optimization with Multiple Targets;398
10.6.4;Conclusion;399
10.6.5;References;399
10.7;Improvement of Quantum Evolutionary Algorithm with a Functional Sized Population;400
10.7.1;Introduction;400
10.7.2;Best Structure for QEA;401
10.7.3;Functional Sized Population QEA (FSQEA);402
10.7.4;Finding the Best Parameters;405
10.7.5;Experimental Results;406
10.7.6;Conclusion;408
10.7.7;References;408
10.8;Optimal Path Planning for Controllability of Switched Linear Systems Using Multi-level Constrained GA;410
10.8.1;Introduction;410
10.8.2;Path Planning for Controllability of Switched Linear Systems;412
10.8.2.1;General Switched Linear System;412
10.8.2.2;Path Planning for Controllability;412
10.8.3;Proposed Approach;413
10.8.3.1;Constrained Genetic Algorithms;413
10.8.3.2;Problem Formulation;414
10.8.3.3;Path Planning for Controllability Using MLCGA;415
10.8.4;Simulation Example;416
10.8.5;Conclusion;418
10.8.6;References;418
11;Part VII Design for Other Soft Computing Techniques;420
11.1;Particle Swarm Optimization for Inference Procedures in the Generalized Gamma Family Based on Censored Data;421
11.1.1;Introduction;421
11.1.2;Problem Formulation;422
11.1.3;Generalized Gamma Family;424
11.1.4;Inference Procedures;425
11.1.5;Particle Swarm Optimization (PSO);426
11.1.6;Applications;427
11.1.7;Conclusion;431
11.1.8;References;431
11.2;SUPER-SAPSO: A New SA-Based PSO Algorithm;433
11.2.1;Introduction;433
11.2.2;Particle Swarm Optimisation;434
11.2.3;SAPSO Hybrid Algorithm;434
11.2.4;Proposed Algorithm (SUPER-SAPSO);435
11.2.5;Experimental Results;436
11.2.6;Conclusions and Future Works;439
11.2.7;References;440
11.3;Testing of Diversity Strategy and Ensemble Strategy in SVM-Based Multiagent Ensemble Learning;441
11.3.1;Introduction;441
11.3.2;Methodology Formulation;442
11.3.3;Experiment Study;446
11.3.3.1;Data Description and Experiment Design;446
11.3.3.2;Experiment Results;447
11.3.4;Conclusions;450
11.3.5;References;450
11.4;Probability Collectives: A Decentralized, Distributed Optimization for Multi-Agent Systems;451
11.4.1;Introduction;451
11.4.2;Distributed, Decentralized, Cooperative MAS;452
11.4.3;Collective Intelligence (COIN) Using Probability Collectives (PC) Theory;453
11.4.3.1;Probability Collectives (PC);453
11.4.3.2;Advantages of PC;453
11.4.4;COIN Formulation;454
11.4.5;Results for Rosenbrock Function Using COIN;457
11.4.6;Conclusions and Future Work;459
11.4.7;References;460
12;Part VIII Computer Graphics, Imaging, Vision and Signal Processing;461
12.1;Shape from Focus Based on Bilateral Filtering and Principal Component Analysis;462
12.1.1;Introduction;462
12.1.2;Related Work;463
12.1.3;Proposed Algorithm;464
12.1.3.1;Motivation;464
12.1.3.2;Bilateral Filtering;464
12.1.3.3;Transformation into Eigenspace;465
12.1.3.4;Depth Map Generation;466
12.1.4;Results and Discussion;466
12.1.4.1;Performance Measures;466
12.1.4.2;Analysis in Eigenspace;467
12.1.4.3;Accuracy Comparison;467
12.1.5;Conclusion;470
12.1.6;References;470
12.2;Detecting Hidden Information from Watermarked Signal Using Granulation Based Fitness Approximation;472
12.2.1;Introduction;473
12.2.2;AFFG Framework - The Main Idea;474
12.2.3;Spread Spectrum Watermarking;475
12.2.4;Recovering PN Sequence;476
12.2.5;Empirical Results;477
12.2.6;Concluding Remarks;479
12.2.7;References;480
12.3;Fuzzy Approaches for Colour Image Palette Selection;482
12.3.1;Introduction;482
12.3.2;Fuzzy c-Means;483
12.3.3;Fast FCM with Random Sampling (RSFCM);484
12.3.4;Fast Generalized FCM Scheme (EnFCM);485
12.3.5;Anisotropic Mean Shift Based FCM (AMSFCM);486
12.3.6;Experimental Results;488
12.3.7;Conclusions;490
12.3.8;References;491
12.4;Novel Face Recognition Approach Using Bit-Level Information and Dummy Blank Images in Feedforward Neural Network;492
12.4.1;Introduction;492
12.4.2;Bit Level Information;493
12.4.3;Feedforward Neural Network;494
12.4.4;Face Expression Database;495
12.4.5;Dummy Blank Images;495
12.4.6;Data and Analysis;496
12.4.7;Conclusion;498
12.4.8;References;498
12.5;ICA for Face Recognition Using Different Source Distribution Models;500
12.5.1;Introduction;501
12.5.2;Algorithm;502
12.5.3;Probability Density Functions;502
12.5.4;Experimental Results;504
12.5.5;Conclusions;506
12.5.6;References;506
12.6;Object Recognition Using Particle Swarm Optimization on Moment Descriptors;508
12.6.1;Introduction;508
12.6.2;Getting Bitmap Image;509
12.6.3;Finding Boundary;509
12.6.4;Moments;510
12.6.5;Similarity Measure;510
12.6.6;Optimization Using Particle Swarm;511
12.6.6.1;Particle Swarm Optimization;511
12.6.6.2;Objective Function;513
12.6.6.3;Steps of the PSO Algorithm;513
12.6.6.4;Test Results Using PSO;514
12.6.7;Conclusion and Future Work;515
12.6.8;References;516
12.7;Perceptual Shaping in Digital Image Watermarking Using LDPC Codes and Genetic Programming;518
12.7.1;Introduction;518
12.7.2;Proposed Scheme;520
12.7.3;Experimental Results;524
12.7.4;Conclusion;526
12.7.5;References;526
12.8;Voice Conversion by Mapping the Spectral and Prosodic Features Using Support Vector Machine;528
12.8.1;Introduction;529
12.8.2;Proposed Voice Conversion System;530
12.8.3;Mapping the Spectral Characteristics;531
12.8.3.1;Features Extraction and Time Alignment;531
12.8.3.2;Training and Testing the Proposed System;532
12.8.4;Mapping the Prosodic Features;532
12.8.5;Synthesis of Speech Signal Using Modified Features;533
12.8.6;Performance Evaluation of the Proposed System;534
12.8.7;Summary and Conclusion;535
12.8.8;Future Work;536
12.8.9;References;536
13;Author Index;538
14;Subject Index;540