Olfactory simulation technology

Author: Yin Yong
Publisher:
Publish Date: 2005-04-01
Features: This book takes the two major components of olfactory simulation technology—sensor arrays and pattern recognition technology—as its main thread, providing a relatively concise, comprehensive, and systematic introduction to the relevant technical composition and basic theories of olfactory simulation. The content includes: multi-sensor array fusion technology, temperature and humidity compensation technology for sensors, feature signal extraction technology, sample screening technology, statistical pattern recognition theory, neural networks, genetic algorithms, and various theoretical and technical aspects, as well as some application fields and achievements of olfactory simulation technology. These contents reflect the author's recent scientific research activities in learning, research, and practice. This book can serve as a reference and learning material for technicians, university teachers, and graduate students in the following fields: quality analysis and control of chemical, food, agricultural, and pharmaceutical products; environmental testing and fire detection; medical diagnosis; transportation.
Introduction Olfactory simulation technology is an emerging interdisciplinary technology. It has developed rapidly over the past decade and is widely used in food analysis, quality identification of fragrances and spices, environmental testing, and medical and health fields. This book takes the two major components of olfactory simulation technology—sensor arrays and pattern recognition technology—as its main thread, providing a relatively concise, comprehensive, and systematic introduction to the relevant technical composition and basic theories of olfactory simulation. The content includes: multi-sensor array fusion technology, temperature and humidity compensation technology for sensors, feature signal extraction technology, sample screening technology, statistical pattern recognition theory, neural networks, genetic algorithms, and various theoretical and technical aspects, as well as some application fields and achievements of olfactory simulation technology. These contents reflect the author's recent scientific research activities in learning, research, and practice. This book can serve as a reference and learning material for technicians, university teachers, and graduate students in the following fields: quality analysis and control of chemical, food, agricultural, and pharmaceutical products; environmental testing and fire detection; medical diagnosis; transportation.
Table of Contents Chapter 1 Introduction 1
Section 1 Introduction to Olfactory Simulation Technology 3
1. Human Olfactory Mechanism 3
2. Principles of Olfactory Simulation Technology 5
3. Main Related Technologies in Olfactory Simulation Technology 6
4. Development History of Olfactory Simulation Technology 12
Section 2 Applications and Development Prospects 12
References 15
Chapter 2 Sensors and Arrays in Olfactory Simulation Technology 18
Section 1 Principles for Selecting Gas-Sensitive Sensors 18
Section 2 Introduction to the Detection Principles of Common Gas-Sensitive Sensors 19
1. Metal Oxide Gas-Sensitive Sensors 19
2. Mass-Type Gas-Sensitive Sensors 22
3. Electrochemical Gas-Sensitive Sensors 22
4. Conductive Polymer Gas-Sensitive Sensors 22
Section 3 Working Conditions and Characteristics of Gas-Sensitive Sensors 23
1. Working Conditions 23
2. Main Characteristic Parameters 24
3. Basic Characteristics 25
4. Influence of Materials and Sensitive Films on the Performance of Gas-Sensitive Sensors 29
Section 4 Response Models of Gas-Sensitive Sensors and Arrays 29
Section 5 Construction Methods of Gas-Sensitive Sensor Arrays 30
1. Determining the Initial Array 30
2. Determining the Final Array 31
3. Examples of Array Construction 34
References 39
Chapter 3 Sample Screening and Feature Information Extraction Technology 40
Section 1 The Necessity of Sample Screening 40
Section 2 Common Sample Screening Techniques 42
1. Sample Screening Using Pattern Classification Methods 42
2. Sample Screening Using Robust Regression Methods 43
3. Sample Screening Using Outlier Discrimination Methods 43
Section 3 Feature Information Extraction Technology 44
1. The Problem Statement 44
2. Some Basic Concepts 44
3. Measurement Information Acquisition Technology 45
4. Techniques for Eliminating Abnormal Data in Measurement Information 47
5. Feature Extraction Technology 48
Section 4 Temperature and Humidity Compensation Methods for Gas-Sensitive Sensor Measurements 56
1. Knowledge-Based Temperature and Humidity Compensation Ideas 57
2. Knowledge-Based Temperature and Humidity Compensation Methods 57
References 60
Chapter 4 Common Statistical Pattern Recognition Methods 61
Section 1 KNN and Its Improved Methods 61
1. Basic KNN Method 61
2. Improvements of KNN Method 62
Section 2 Fisher Discriminant Method 63
1. Basic Idea of Fisher Discriminant Method 63
2. Mathematical Description of Fisher Discriminant Method 63
Section 3 Introduction to Principal Component Regression 65
Section 4 Partial Least Squares Method 66
1. Basic Principles 67
2. Derivation of Computational Methods 67
Section 5 Clustering Analysis Method 69
1. Similarity and Distance 69
2. Hierarchical Clustering Method 70
3. One-Time Classification Method 71
4. Mapping Classification Method 72
References 73
Chapter 5 Artificial Neural Network Pattern Recognition Methods 74
Section 1 Overview 74
1. Brief History, Current Status, and Characteristics of Neural Network Research 74
2. Forward Neural Network Model 76
3. Nonlinear Function Approximation Theory of Forward Neural Networks 80
Section 2 BP Neural Network Learning Algorithm 83
1. BP Learning Algorithm Based on Batch Processing 83
2. BP Learning Algorithm Based on Recursive Least Squares 84
Section 3 RBF Neural Network Learning Algorithm 86
1. Overview of Common Learning Algorithms 87
2. A Gaussian Kernel-Based RBF Neural Network Learning Algorithm 88
Section 4 Self-Organizing Artificial Neural Networks 93
1. Basic Ideas and Learning Principles 93
2. Learning Algorithms 95
References 96
Chapter 6 Genetic Algorithms and Genetic Neural Networks 98
Section 1 Overview 98
1. Research History and Development Directions of Genetic Algorithms 99
2. Basic Characteristics of Genetic Algorithms 100
Section 2 Basic Theories of Genetic Algorithms 102
1. Basic Genetic Operators 102
2. Theoretical Basis of Genetic Algorithms 104
3. Convergence of Genetic Algorithms 107
Section 3 Key Parameters and Basic Steps of Genetic Algorithms 108
1. Determination of Key Parameters 108
2. Basic Steps of the Algorithm 110
Section 4 Species Formation and Niche Techniques in Genetic Algorithms 110
Section 5 Deception Problems in Genetic Algorithms 111
Section 6 Genetic Algorithm-Based LBF Neural Networks 111
1. Binary-Coded Genetic Neural Networks 112
2. Decimal-Coded Genetic Neural Networks 116
Section 7 Genetic Algorithm-Based RBF Neural Networks 119
References 120
Chapter 7 Applications of Olfactory Simulation Technology 122
Section 1 Identification of Alcoholic Beverages 122
1. Evaluation of the Quality of Alcoholic Aromas 123
2. Identification or Quality Classification of Alcoholic Beverages 123
3. Discussion on Sampling Methods 124
4. Stability Discrimination of Alcoholic Beverage Quality 125
Section 2 Quality Detection of Fruit and Vegetable Juices 128
Section 3 Freshness Discrimination of Meat Products 128
Section 4 Environmental Detection 129
Section 5 Applications in Medical Diagnosis 130
Section 6 Applications in Fire Detection 131
References 132
Appendix 1 C Program for Batch-Processing BP Neural Network 133
Appendix 2 C Program for Gaussian Kernel-Based RBF Neural Network 147
Appendix 3 C Program for Binary-Coded Genetic LBF Neural Network 162
Appendix 4 C Program for Decimal-Coded Genetic LBF Neural Network 179
Appendix 5 C Program for Decimal-Coded Genetic RBF Neural Network 194