This book shows how wavelets can be used to enhance computational intelligence for chaotic and complex pattern recognition problems.  By integrating wavelets with other soft computing techniques such as neurocomputing and fuzzy logic, complicated and noisy pattern recognition problems can be solved effectively. The book focuses on applications in intelligent transportation systems (ITS) where a number of very complicated pattern recognition problems have eluded researchers over the past few decades. Advancing the frontiers of computational intelligence, this book: Describes ingenious computational models based on novel problem solving and computing techniques such as Case-Based Reasoning, Neurocomputing, and Wavelets, and presents examples to illustrate their importance and use. Presents a multi-paradigm intelligent systems approach to the freeway traffic incident detection and construction work zone management problems. Advocates application and integration of wavelets, neural networks and fuzzy logic for modeling the complex traffic flow behaviors leading to effective and efficient control and management solutions. Presents efficient, reliable, and robust algorithms for automatic detection of incidents on freeways. Wavelets in Intelligent Transportation Systems is an invaluable resource for computational intelligence researchers and transportation engineers involved in the application of advanced computational techniques for ITS.

Table of Contents:
Preface xiii Acknowledgements xv About the Authors xvii 1 Introduction 1 2 Introduction to Wavelet Analysis 3 2.1 Introduction 3 2.2 Basic Concept of Wavelets and Wavelet Analysis 5 2.3 Mathematical Foundations 9 2.4 The Discrete Wavelet Transform (DWT) 14 2.5 Multi-resolution Analysis 16 2.6 Wavelet Bases 18 2.7 Computing the DWT 23 3 Feature Extraction for Traffic Incident Detection Using the Wavelet Transform and Linear Discriminant Analysis 27 3.1 Introduction 27 3.2 Incident Detection Algorithms 29 3.3 Discrete Wavelet Transform (DWT) of Traffic Signals 31 3.4 Linear Discriminant Analysis (LDA) 34 3.5 Data Acquisition 37 3.6 Results 38 4 Adaptive Conjugate Gradient Neural Network-Wavelet Model for Traffic Incident Detection 45 4.1 Introduction 45 4.2 Improving Traffic Incident Detection 46 4.3 Adaptive Conjugate Gradient Neural Network Model 48 4.4 Incident Detection Results Using Various Approaches 52 4.5 Effect of Data Filtering Using the DWT 55 4.6 Relative Contribution of DWT and LDA for Feature Extraction 58 4.7 Effects of Freeway Geometry on Incident Detection 58 4.8 Conclusion 65 5 Enhancing Fuzzy Neural Network Algorithms Using Wavelets 67 5.1 Introduction 67 5.2 Discrete Wavelet Transform 69 5.3 Architecture 69 5.4 Training of the Network 72 5.5 Filtering of the Traffic Data Using the DWT 73 5.6 Incident Detection Results 73 6 Fuzzy-Wavelet Radial Basis Function Neural Network Model for Freeway Incident Detection 79 6.1 Introduction 79 6.2 A New Traffic Incident Detection Methodology 80 6.3 Selection of Type and Number of Traffic Data 83 6.4 Wavelet-Based De-noising 85 6.5 Fuzzy Data Clustering 86 6.6 Radial Basis Function Neural Network Classifier 89 6.7 Fuzzy-Wavelet RBFNN Model for Incident Detection 91 6.8 Example 93 6.9 Conclusion 97 7 Comparison of the Fuzzy-Wavelet RBFNN Freeway Incident Detection Model with the California Algorithm 99 7.1 Introduction 99 7.2 California Algorithm #8 100 7.3 Evaluation of the Model 101 7.4 Concluding Remarks 116 8 Incident Detection Algorithm Using a Wavelet Energy Representation of Traffic Patterns 119 8.1 Introduction 119 8.2 Freeway Incident Detection and Patterns in Traffic Flow 120 8.3 Discrete Wavelet Transform and Signal Energy 129 8.4 Traffic Pattern Feature Enhancement and De-noising 131 8.5 Pattern Classification Using Radial Basis Function Neural Network 136 8.6 Wavelet Energy Freeway Incident Detection Algorithm 137 8.7 Model Testing 140 8.8 Conclusion 145 9 Parametric Evaluation of the Wavelet Energy Freeway Incident Detection Algorithm 147 9.1 Introduction 147 9.2 Factors to Consider in Rural Freeway Incident Detection 148 9.3 Evaluation and Parametric Investigation 150 9.4 Parametric Evaluation Using Simulated Data on Typical Urban Freeways 152 9.5 False Alarm Performance in the Vicinity of On-and Off Ramps 156 9.6 Evaluation on Rural Freeways 157 9.7 Evaluation Using Real Data 163 9.8 Performance Summary and Conclusion 164 10 Case-Based Reasoning Model for Work Zone Traffic Management 167 10.1 Introduction 167 10.2 Work Zones and Traffic Management 169 10.3 Case-Based Reasoning 170 10.4 Objectives 173 10.5 Scope and Categorization of Parameters 174 10.6 A Four-Set Case Model for the Work Zone Traffic Management Domain 176 10.7 Hierarchical Object-Oriented Case Model 179 10.8 Case Representation 181 10.9 Similarity Measures 185 10.10 Case Retrieval 185 10.11 Creation of the Case Base 186 10.12 Creation of Work Zone Traffic Control Plans Using the CBR System 187 10.13 Illustrative Examples 189 10.14 Conclusion 195 11 Mesoscopic-Wavelet Freeway Work Zone Flow and Congestion Model 197 11.1 Introduction 197 11.2 Macroscopic Models 200 11.3 Microscopic Models 200 11.4 A Mesoscopic Flow Model for a Freeway Work Zone 203 11.5 Traffic Feature Enhancement Using Discrete Wavelet Transform 211 11.6 Concluding Remarks 213 References 215 Index 223