Supervised and Unsupervised Pattern Recognition: Feature Extraction and Computational Intelligence

by Evangelia Miche Tzanakou (Editor)

Overview

There are many books on neural networks, some of which cover computational intelligence, but none that incorporate both feature extraction and computational intelligence, as Supervised and Unsupervised Pattern Recognition does. This volume describes the application of a novel, unsupervised pattern recognition scheme to the classification of various types of waveforms and images. This substantial collection of recent research begins with an introduction to Neural Networks, classifiers, and feature extraction methods. It then addresses unsupervised and fuzzy neural networks and their applications to handwritten character recognition and recognition of normal and abnormal visual evoked potentials. The third section deals with advanced neural network architectures-including modular design-and their applications to medicine and three-dimensional NN architecture simulating brain functions. The final section discusses general applications and simulations, such as the establishment of a brain-computer link, speaker identification, and face recognition. In the quickly changing field of computational intelligence, every discovery is significant. Supervised and Unsupervised Pattern Recognition gives you access to many notable findings in one convenient volume.

Product Details

ISBN-13:	9781351835558
Publisher:	CRC Press
Publication date:	12/19/2017
Series:	Industrial Electronics
Sold by:	Barnes & Noble
Format:	eBook
Pages:	392
File size:	14 MB
Note:	This product may take a few minutes to download.

About the Author

Evangelia Miche Tzanakou

Table of Contents

classifiers-an overview Criteria for optimal classifier design Categorizing the Classifiers Classifiers Neural Networks Comparison of Experimental Results System Performance Assessment Analysis of Prediction Rates from Bootstrapping Assessment ARTIFICIAL NEURAL NETWORKS: DEFINITIONS, METHODS, APPLICATIONS Definitions Training Algorithm Some Applications A SYSTEM FOR HANDWRITTEN DIGIT RECOGNITION Preprocessing of Handwritten Digit Images Zernike Moments (ZM) for Characterization of Image Patterns Dimensionality Reduction Analysis of Prediction Error Rates from Bootstrapping Assessment Summary OTHER TYPES OF FEATURE EXTRACTION METHODS Introduction Wavelets Invariant Moments Entropy Cepstrum Analysis Fractal Dimension Entropy SGLD Texture Features FUZZY NEURAL NETWORKS Pattern Recognition Optimization System Design Clustering APPLICATION TO HANDWRITTEN DIGITS Introduction to Character Recognition Data Collection Results Discussion Summary A UNSUPERVISED NEURAL NETWORK SYSTEM FOR VISUAL EVOKED POTENTIALS Data Collection and Preprocessing System Design Results Discussion CLASSIFICATION OF MAMMOGRAMS USING A MODULAR NEURAL NETWORK Methods and System Overview Modular Neural Networks Neural Network Training Classification Results The Process of Obtaining Results ALOPEX Parameters Generalization Conclusions VISUAL OPHTHALMOLOGIST: AN AUTOMATED SYSTEM FOR CLASSIFICATION OF RETINAL DAMAGE System Overview Modular Neural Networks Applications to Ophthalmology Results Discussion A THREE-DIMENSIONAL NEURAL NETWORK ARCHITECTURE The Neural Network Architecture Simulations Discussion A FEATURE EXTRACTION ALGORITHM USING CONNECTIVITY STRENGTHS AND MOMENT INVARIANTS ALOPEX Algorithms Moment Invariants and ALOPEX Results and Discussion MULTILAYER PERCEPTRONS WITH ALOPEX: 2D-TEMPLATE MAT

Editorial Reviews

Describes the application of supervised and unsupervised pattern recognition schemes to the classification of various types of waveforms and images. Micheli-Tzanakou (biomedical engineering, (Rutgers U.) treats experimental and theoretical contributions equally and examines interchanges between the two. The chapters span a variety of problems in signal and image processing, using mainly neural networks for classification and template matching. Much of the work was co-authored with students of Micheli-Tzanakou. Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews

"This book is an excellent source of knowledge of state-of-the-art feature extraction…Supervised and unsupervised learning and training schemes are notable finds…Exciting applications of signal and image analysis and recognition…This book provides in-depth guidance and inspiring ideas to new applications of signal and image analysis and recognition."
—Tonglei Li, Ph.D., Purdue University, School of Pharmacy
"…great efforts have been made in a number of communities to explore solutions to pattern recognition problems…this book describes their efforts made over ten researchers in the Neuroelectric and Neurocomputing Laboratories at Rutgers University. Along with concise introductory materials in pattern recognition, this volume presents several applications of supervised and unsupervised schemes to the classification of various types of signals and images…Unlike other books in neural networks, this book gives an emphasis on feature extraction as well, which provides a systematic way to deal with pattern recognition problems in terms of neural networks and computational intelligence…it is worth noting that each chapter contains an extensive bibliography that provides a reliable list of good references. We believe that readers will find this list very useful to understand the materials in the book and cautious beginners in the related fields might benefit from this list as well…helpful to a broad audience of graduate students, researchers, practicing engineers and professionals in computer and information science, electrical engineering, and biomedical informatics…this book reflects the long-term continuous endeavors of a research group for conducting innovatory researches, which could provide some useful hints to those novices in related fields…pioneering volume…welcomed by all interested in the fields of pattern recognition and computational intelligence…the editor's serious attempt to address the aforementioned issue must be welcomed by all interested in the fields of pattern recognition and computational intelligence and, therefore, this book deserves all credit."
—Ke Chen, National Laboratory of Machine Perception and The Center for Information Science, Peking University, Beijing, China
Promo Copy

From the Publisher