Pattern Recognition with Neural Networks in C++

by Pandya; Abhijit S.

• ISBN13: 9780849394621
• ISBN10: 0849394627
• Format: Hardcover
• Copyright: 1995-10-17
• Publisher: CRC Press

Summary

The addition of artificial neural network computing to traditional pattern recognition has given rise to a new, different, and more powerful methodology that is presented in this interesting book. This is a practical guide to the application of artificial neural networks.Geared toward the practitioner, Pattern Recognition with Neural Networks in C++ covers pattern classification and neural network approaches within the same framework. Through the book's presentation of underlying theory and numerous practical examples, readers gain an understanding that will allow them to make judicious design choices rendering neural application predictable and effective. The book provides an intuitive explanation of each method for each network paradigm. This discussion is supported by a rigorous mathematical approach where necessary.C++ has emerged as a rich and descriptive means by which concepts, models, or algorithms can be precisely described. For many of the neural network models discussed, C++ programs are presented for the actual implementation. Pictorial diagrams and in-depth discussions explain each topic. Necessary derivative steps for the mathematical models are included so that readers can incorporate new ideas into their programs as the field advances with new developments. For each approach, the authors clearly state the known theoretical results, the known tendencies of the approach, and their recommendations for getting the best results from the method.The material covered in the book is accessible to working engineers with little or no explicit background in neural networks. However, the material is presented in sufficient depth so that those with prior knowledge will find this book beneficial. Pattern Recognition with Neural Networks in C++ is also suitable for courses in neural networks at an advanced undergraduate or graduate level. This book is valuable for academic as well as practical research.

Table of Contents

Introduction

Pattern Recognition Systems

Motivation for Artificial Neural Network Approach

A Prelude to Pattern Recognition

Statistical Pattern Recognition

Syntactic Pattern Recognition

The Character Recognition Problem

Organization of Topics

Neural Networks: An Overview

Motivation for Overviewing Biological Neural Networks

Background

Biological Neural Networks

Hierarchical Organization of the Brain

Historical Background

Artificial Neural Networks

Preprocessing

General

Dealing with Input from a Scanned Image

Image Compression

Edge Detection

Skeletonizing

Dealing with Input from a Tablet

Segmentation

Feed Forward Networks with Supervised Learning

Feed-Forward Multilayer Perceptron (FFMLP) Architecture

FFMLP in C++

Training with Back Propagation

A Primitive Example

Training Strategies and Avoiding Local Minima

Variations on Gradient Descent

Topology

ACON vs. OCON

Overtraining and Generalization

Training Set Size and Network Size

Conjugate Gradient Method

ALOPEX

Some Other Types of Neural Networks

General

Radial Basis Function Networks

Higher Order Neural Networks

Feature Extraction I: Geometric Features and Transformations

General

Geometric Features (Loops, Intersections and Endpoints)

Feature Maps

A Network Example Using Geometric Features

Feature Extraction Using Transformations

Fourier Descriptors

Gabor Transformations and Wavelets

Feature Extraction II: Principle Component Analysis

Dimensionality Reduction

Principal Components

Karhunen-Loeve (K-L) Transformation

Principal Component Neural Networks

Applications

Kohonen Networks and Learning Vector Quantization

General

K-Means Algorithm

An Introduction to the Kohonen Model

The Role of Lateral Feedback

Kohonen Self-Organizing Feature Map

Learning Vector Quantization

Variations on LVQ

Neural Associative Memories and Hopfield Networks

General

Linear Associative Memory (LAM)

Hopfield Networks

A Hopfield Example

Discussion

Bit Map Example

BAM Networks

A BAM Example

Adaptive Resonance Theory (ART)

General

Discovering the Cluster Structure

Vector Quantization

ART Philosophy

The Stability-Plasticity Dilemma

Art1: Basic Operation

Art1: Algorithm

The Gain Control Mechanism

ART2 Model

Discussion

Applications

Neocognition

Introduction

Architecture

Example of a System with Sample Training Patterns

Systems with Multiple Classifiers

General

A Framework for Combining Multiple Recognizers

Voting Schemes

The Confusion Matrix

Reliability

Some Empirical Approaches