Mobile Robot Navigation with Intelligent Infrared Image Interpretation (eBook)

William L. Fehlman II is a lieutenant colonel in the United States Army. He received a BS in Mathematics from SUNY Fredonia in 1990 and MS in Applied Mathematics from Rensselaer Polytechnic Institute in 2000. He earned a PhD in Applied Science from The College of William and Mary in 2008, and is currently assigned as an Assistant Professor of Mathematics at the United States Military Academy, West Point, New York. His research interests include pattern classification, multi-sensor data fusion, and autonomous robotic systems. Mark K. Hinders holds a BS, MS and PhD in Aerospace and Mechanical Engineering from Boston University, and is currently Professor of Applied Science at the College of William and Mary in Virginia. Before coming to Williamsburg in 1993, Professor Hinders was Senior Scientist at Massachusetts Technological Laboratory, Inc., and also Research Assistant Professor at Boston University. Before that Dr Hinders was an Electromagnetics Research Engineer at the USAF Rome Laboratory located at Hanscom AFB, MA. Professor Hinders conducts research in wave propagation and scattering phenomena, applied to medical imaging, intelligent robotics, security screening, remote sensing and nondestructive evaluation. He and his students study the interaction of acoustic, ultrasonic, elastic, thermal, electromagnetic and optical waves with materials, tissues and structures.

Preface 5
List of Symbols 7
List of Figures 11
List of Tables 20
Contents 25
1 Introduction and Overview 28
1.1 Purpose of Book 28
1.2 Non-Heat Generating Objects 33
1.3 Autonomous Robotic Systems 35
1.4 Infrared Thermography 42
1.5 Overview of the Book 47
References 49
2 Data Acquisition 51
2.1 Introduction 51
2.2 Robotic Thermal Imaging System 51
2.3 Data Collection 67
2.4 Summary 71
References 71
3 Thermal Feature Generation 73
3.1 Introduction 73
3.2 “Ugly Duckling” Features 74
3.3 Thermal Image Representation 77
3.4 Meteorological Features 80
3.5 Micro Features 81
3.6 Macro Features 95
3.7 Thermal Feature Application 110
3.8 Curvature Algorithm 114
3.9 Summary 117
References 117
4 Thermal Feature Selection 120
4.1 Introduction 120
4.2 “No Free Lunch” Classifiers 121
4.3 Preliminary Feature Analysis 123
4.4 Classifiers 130
4.5 Model Performance and Feature Selection 140
4.6 Sensitivity Analysis 168
References 183
5 Adaptive Bayesian Classification Model 185
5.1 Introduction 185
5.2 Distance Metrics for Hyperconoidal Clusters 186
5.3 Adaptive Bayesian Classifier Design 217
5.4 Adaptive Bayesian Classifier Appraisal 221
5.5 Adaptive Bayesian Classification Model Design 247
5.6 Adaptive Bayesian Classification Model Application 251
5.7 Summary 275
References 278
6 Conclusions and Future Research Directions 279
6.1 Introduction 279
6.2 Contributions 280
6.3 Limitation of a Thermal Infrared Imaging System 281
6.4 Future Research 283
6.5 Concluding Remarks 292
References 293
Index 294