Soft Computing for Recognition based on Biometrics (Studies in Computational Intelligence, 312)

We describe in this book, bio-inspired models and applications of hybrid intel- gent systems using soft computing techniques for image analysis and pattern r- ognition based on biometrics and other information sources. Soft Computing (SC) consists of several intelligent computing paradigms, including fuzzy logic, neural networks, and bio-inspired optimization algorithms, which can be used to produce powerful hybrid intelligent systems. The book is organized in five main parts, which contain a group of papers around a similar subject. The first part consists of papers with the main theme of classification methods and applications, which are basically papers that propose new models for classification to solve general pr- lems and applications. The second part contains papers with the main theme of modular neural networks in pattern recognition, which are basically papers using bio-inspired techniques, like modular neural networks, for achieving pattern r- ognition based on biometric measures. The third part contains papers with the theme of bio-inspired optimization methods and applications to diverse problems. The fourth part contains papers that deal with general theory and algorithms of bio-inspired methods, like neural networks and evolutionary algorithms. The fifth part contains papers on computer vision applications of soft computing methods. In the part of classification methods and applications there are 5 papers that - scribe different contributions on fuzzy logic and bio-inspired models with appli- tion in classification for medical images and other data. Title Preface Contents Part I Classification Algorithms and Applications Soft Computing Approaches to the Problem of Infant Cry Classification with Diagnostic Purposes Introduction The Infant Cry Automatic Recognition Process Logical-Combinatorial Approach The Connectionist Approach Genetic-Neural Approach Development of a Hybrid Classifier Classification Results Analysis Statistic Measures for Reducing Input Vectors The Fuzzy Approach Results Compressing the Cry Features Implementation and Experiments Conclusions Recommendations References Neural Networks and SVM-Based Classification of Leukocytes Using the Morphological Pattern Spectrum Introduction Image Morphology Image Morphology Definitions Morphological Pattern Spectrum Composed Feature Vector Multilayer Perceptron Support Vector Machines (SVM) Experimental Description and Results Conclusions References Hybrid System for Cardiac Arrhythmia Classification with Fuzzy K-Nearest Neighbors and Neural Networks Combined by a Fuzzy Inference System Introduction Problem Statement and Outline of our Proposal Background Theory Fuzzy K-Nearest Neighbors Algorithm Multi Layer Perceptron Backpropagation Training for a Multilayer Perceptron Gradient Descent with Momentum (GDM) Gradient Scaled Conjugate (SC) Experiments MIT-BIH Arrhythmia Database Experimental Setup Classifiers Statistical Analysis Conclusions References A Comparative Study of Blog Comments Spam Filtering with Machine Learning Techniques Introduction Background and Basic Concepts Naïve Bayes K Nearest Neighbors Neural Networks Support Vector Machines Architecture Corpus Preprocessing PP A1 Naïve Bayes A2 Support Vector Machines A3 Neural Networks A4 K-nearest Neighbors Models Results Results of A1 Naïve Results of A2 Support Vector Machines Results of A3 Neural Networks Results of A4 K-Nearest Neighbors Global Results Statistical Analysis Mean Match Pairs Conclusions References Distributed Implementation of an Intelligent Data Classifier Introduction Related Work Architecture for Building a Distributed Data Classifier Building the Global Classifier ID3 Global Data Classifier Implementation Experimental Evaluations Conclusions References Part II Pattern Recognition Modular Neural Network with Fuzzy Integration and Its Optimization Using Genetic Algorithms for Human Recognition Based on Iris, Ear and Voice Biometrics Introduction Background Iris Ear Voice Basic Concepts Modular Neural Network Fuzzy Logic Genetic Algorithms Proposed Method and Results Methodology Databases and Pre-processing Architecture and Results of the Modular Neural Network Genetic Algorithm for MNN Optimization and Results Fuzzy Integration Genetic Algorithm for Fuzzy Integrator and Results Conclusions References Comparative Study of Type-2 Fuzzy Inference System Optimization Based on the Uncertainty of Membership Functions Introduction Preliminaries Modular Neural Networks Type-2 Fuzzy Logic Genetic Algorithms Optimization Method Description Fuzzy Systems Optimization Based on the Level of Uncertainty Simulation Results Adaptive Noise Cancellation MPG Benchmark Problem Conclusions References Modular Neural Network for Human Recognition from Ear Images Using Wavelets Introduction Background Related Work Ear Ear Recognition Process Data Acquisition Image Pre-processing Neural Network Structure Neural Network Training Modular Integration Conclusions References Modular Neural Networks for Person Recognition Using the Contour Segmentation of the Human Iris Biometric Measurement Introduction Backgroun and Basic Concepts Modular Neural Network Historical Development Iris Properties Proposed Method and Problem Description Problem Description Image Pre-processing Modular Neural Network Architecture Simulation Results Results with the Initial Modular Neural Network Architecture First Modification of the Modular Neural Network Architecture Second Modification of the Modular Neural Network Architecture (Extended) Conclusions References Real Time Face Identification Using a Neural Network Approach Introduction Background and Basic Concepts Historical Development Proposed Method and Problem Description Image Acquisition (Database) Preprocessing Feature Extraction Edge Extraction Eigenface Discrete Wavelet Transform (dwt) Modular Neural Network Architecture Simulation Results Results Using Only the Pre-processed Database Results Using the Pre-processed Database with Edges Conclusions References Comparative Study of Feature Extraction Methods of Fuzzy Logic Type 1 and Type-2 for Pattern Recognition System Based on the Mean Pixels Introduction Fuzzy Image Enhancement Based Pixels Brightness Architecture Pattern Recognition with Fuzzy Extraction Features Simulation Results Type-1 and Type-2 with Blur Motion Comparison of the Results Type-1 and Type-2 Fuzzy Logic Extraction Features Conclusions References Part III Optimization Methods Application of the Bee Swarm Optimization BSO to the Knapsack Problem Introduction Knapsack Problem Types of Knapsack Problem Instances Bee Algorithm (BA) Particle Swarm Optimization (PSO) Bee Swarm Optimization (BSO) Experiments Results Uncorrelated Knapsack Problem Weakly Correlated Strongly Correlated Inverse Strongly Correlated Almost Strongly Correlated Subset-sum Uncorrelated Similar Weight Conclusions References An Approach Based on Neural Networks for Gas Lift Optimization Introduction Strategy of Optimization Based on a Neural Network Results and Discussions First Case: Produced Oil Rate by a Single Well Second Case: Produced Oil Rate by a Production System Based on Two Wells Comparative Analysis Conclusions References A New Evolutionary Method with Particle Swarm Optimization and Genetic Algorithms Using Fuzzy Systems to Dynamically Parameter Adaptation Introduction Genetic Algorithm for Optimization Particle Swarm Optimization Full Model of FPSO+FGA FPSO (Fuzzy Particle Swarm Optimization) FGA (Fuzzy Genetic Algorithm) Definition of the Fuzzy Systems Used in FPSO+FGA Benchmark Mathematical Functions Simulations Results Simulation Results with the Genetic Algorithm (GA) Simulation Results with Particle Swarm Optimization Simulation Results with FPSO+FGA Comparison Results between GA, PSO and FPSO+FGA Conclusions References Local Survival Rule for Steer an Adaptive Ant-Colony Algorithm in Complex Systems Introduction Background Graph Theory Structural Characterization SQRP Search Strategies SQRP Description SQRP Algorithms AdaNAS Model The General Model Behavior Rule AdaNAS Algorithm Experiments Generation of the Test Data Parameters Results Conclusions References Using Consecutive Swaps to Explore the Insertion Neighborhood in Tabu Search Solution of the Linear Ordering Problem Introduction Related Work Proposed Method Main Idea Improving the Neighborhood Exploration Example Experimental Results Conclusions and Future Work References A New Optimization Method Based on a Paradigm Inspired by Nature Introduction Chemical Paradigm Modeling the Chemical Paradigm Preliminary Experimental Results Conclusions References Part IV Theory and Algorithms Improvement of the Backpropagation Algorithm Using (1+1) Evolutionary Strategies Introduction Background and Basic Concepts Artificial Neural Networks The Backpropagation Algorithm (BP) Main Shortcomings of Backpropagation Algorithm Improvements to the Backpropagation Algorithm Problem Description and Proposed Method Evolutionary Computation and ANNs Evolutionary Strategies Evolutionary Strategies for Backpropagation Learning Experimental Results Summary and Conclusions References Parallel Genetic Algorithms for Architecture Optimization of Neural Networks for Pattern Recognition Introduction Theoretical Concepts Neural Networks Genetic Algorithms Parallel Genetic Algorithms Problem Statement Neural Network Structure Parallel Genetic Algorithm for Optimization Experimental Results Conclusions References Scene Recognition Based on Fusion of Color and Corner Features Introduction Corner Detection Method Edge Detection Corner Detection Windows Corner Detection Scene Segmentation RGB to HSV Color Space HSV Component Analysis Door Segmentation Recognition of Scenarios Results Conclusions References Improved Tabu Solution for the Robust Capacitated International Sourcing Problem (RoCIS) Introduction Related Work Problem Formulation Improved Tabu Solution Improving the Initial Solution Construction Improving the Neighborhood Construction Experimental Results Conclusions and Future Work References Variable Length Number Chains Generation without Repetitions Introduction Linear Congruential Method Calculation of n$_0$, a, c and m Parameters Variable Length Integer Number Chain Generation without Repetitions Variable Length Number Chains without Repetitions in the Interval (0,1), (0,1] and [0,1] References Comparative Analysis of Hybrid Techniques for an Ant Colony System Algorithm Applied to Solve a Real-World Transportation Problem Introduction Vehicle Routing Problem (VRP) Related Works VRP Variants The Ant Colony System Algorithm (ACS) State of Art RoSLoP: A Real-World Industrial Application Routing-Scheduling Constrains Loading Constrains A Reduction Technique for the Loading Elements The Exact Approach of RoSLoP The Heuristic Solution of RoSLoP Experimentation and Results The Dataset of Solomon Experimentation with RoSLoP Instances Conclusions and Future Contributions References Part V Computer Vision Applications Comparison of Fuzzy Edge Detectors Based on the Image Recognition Rate as Performance Index Calculated with Neural Networks Introduction Overview of the Tested Edge Detectors Sobel Edge Detector Improved with Fuzzy Systems Morphological Gradient Detector Improved with Fuzzy Systems Design of the Experiment General Algorithm Used for the Experiment Parameters Depend on the Database of Images The Monolithic Neural Network Results Conclusion References Intelligent Method for Contrast Enhancement in Digital Video Introduction General Concepts for an Image Quality in TFT-LCD Consumer Applications Analysis of Brightness and Contrasts in an Image Linear Point Operations in Images Transformation Function Design of the Intelligent Method for Contrast Enhancement in Digital Video Contrast Ratio and Light Leakage in TFT-LCD Devices Light Perception and Simultaneous Contrasts Brightness Perception To Preserve Level of Original Brightness Description of the ProposedMethod Proposed Neural Network and Its Training Description of the Evaluation Method and Analysis of Results Evaluation Method Using Test Pattern Image EvaluationMethod Using Real Images Conclusions References Method for Obstacle Detection and Map Reconfiguration in Wheeled Mobile Robotics Introduction General System Description Stereoscopic Vision and Obstacles Detection Surface Ground Extraction and Obstacle Detection Using Luminance and Hue Stereoscopic Vision System Module and FPGA Implementation Design of the Stereoscopic Vision Module Depth Measure from Stereo Image Map Building and Map Reconfiguration Conclusion References Automatic Dust Storm Detection Based on Supervised Classification of Multispectral Data Introduction An Overview of MODIS Data Selection and Analysis of Spectral Bands for Feature Extraction Dust Storm Detection Using the Maximum Likelihood Classifier Neuro-Probabilistic Modeling: The Probabilistic Neural Network The PNN Large Sample Size Problem Results and Discussion Conclusion References Author Index We describe in this book, bio-inspired models and applications of hybrid intelligent systems using soft computing techniques for image analysis and pattern recognition based on biometrics and other information sources. Soft Computing (SC) consists of several intelligent computing paradigms, including fuzzy logic, neural networks, and bio-inspired optimization algo-rithms, which can be used to produce powerful hybrid intelligent systems. The book is organized in five main parts, which contain a group of papers around a similar subject.