Biometric Authentication: A Machine Learning Approach

Preface. 1. Overview. Introduction. Biometric Authentication Methods. Face Recognition: Reality and Challenge. Speaker Recognition: Reality and Challenge. Road Map of the Book. 2. Biometric Authentication Systems. Introduction. Design Tradeoffs. Feature Extraction. Adaptive Classifiers. Visual-Based Feature Extraction and Pattern Classification. Audio-Based Feature Extraction and Pattern Classification. Concluding Remarks. 3. Expectation-Maximization Theory. Introduction. Traditional Derivation of EM. An Entropy Interpretation. Doubly-Stochastic EM. Concluding Remarks. 4. Support Vector Machines. Introduction. Fisher's Linear Discriminant Analysis. Linear SVMs: Separable Case. Linear SVMs: Fuzzy Separation. Nonlinear SVMs. Biometric Authentication Application Examples. 5. Multi-Layer Neural Networks. Introduction. Neuron Models. Multi-Layer Neural Networks. The Back-Propagation Algorithms. Two-Stage Training Algorithms. Genetic Algorithm for Multi-Layer Networks. Biometric Authentication Application Examples. 6. Modular and Hierarchical Networks. Introduction. Class-Based Modular Networks. Mixture-of-Experts Modular Networks. Hierarchical Machine Learning Models. Biometric Authentication Application Examples. 7. Decision-Based Neural Networks. Introduction. Basic Decision-Based Neural Networks. Hierarchical Design of Decision-Based Learning Models. Two-Class Probabilistic DBNNs. Multiclass Probabilistic DBNNs. Biometric Authentication Application Examples. 8. Biometric Authentication by Face Recognition. Introduction. Facial Feature Extraction Techniques. Facial Pattern Classification Techniques. Face Detection and Eye Localization. PDBNN Face Recognition System Case Study. Application Examples for Face Recognition Systems. Concluding Remarks. 9. Biometric Authentication by Voice Recognition. Introduction. Speaker Recognition. Kernel-Based Probabilistic Speaker Models. Handset and Channel Distortion. Blind Handset-Distortion Compensation. Speaker Verification Based

Sun-Yuan Kung is a professor of electrical engineering at Princeton University. His research and teaching interests include VLSI signal processing; neural networks; digital signal, image, and video processing; and multimedia information systems. His books include VLSI Array Processors and Digital Neural Networks (Prentice Hall PTR). Man-Wai Mak is an assistant professor at The Hong Kong Polytechnic University and chairman of the IEEE Hong Kong Section Computer Chapter. His research interests include speaker recognition, machine learning, and neural networks. Shang-Hung Lin is a senior architect at Nvidia, a leader in video and imaging products.