1 Introduction
1.1 Pattern Recognition
1.2 Uncertainties
1.3 Book Overview
References
2 Probabilistic Graphical Models
2.1 The Labeling Problem
2.2 Markov Properties
2.3 The Bayesian Decision Theory
2.3.1 Descriptive and Generative Models
2.3.2 Statistical-Structural Pattern Recognition
2.4 Summary
References
3 Type-2 Fuzzy Sets for Pattern Recognition
3.1 Type-2 Fuzzy Sets
3.2 Operations on Type-2 Fuzzy Sets
3.3 Type-2 Fuzzy Logic Systems
3.3.1 Fuzzifier
3.3.2 Rule Base and Inference
3.3.3 Type Reducer and Defuzzifier
3.4 Pattern Recognition Using Type-2 Fuzzy Sets
3.5 The Type-2 Fuzzy Bayesian Decision Theory
3.6 Summary
References
4 Type-2 Fuzzy Gaussian Mixture Models
4.1 Gaussian Mixture Models
4.2 Type-2 Fuzzy Gaussian Mixture Models
4.3 Multi-category Pattern Classification
References
5 Type-2 Fuzzy Hidden Moarkov Models
5.1 Hidden Markov Models
5.1.1 The Forward-Backward Algorithm
5.1.2 The Viterbi Algorithm
5.1.3 The Baum-Welch Algorithm
5.2 Type-2 Fuzzy Hidden Markov Models
5.2.1 Elements of a Type-2 FHMM
5.2.2 The Type-2 Fuzzy Forward-Backward Algorithm
5.2.3 The Type-2 Fuzzy Viterbi Algorithm
5.2.4 The Learning Algorithm
5.2.5 Type-Reduction and Defuzzification
5.2.6 Computational Complexity
5.3 Speech Recognition
5.3.1 Automatic Speech Recognition System
5.3.2 Phoneme Classification
5.3.3 Phoneme Recognition
5.4 Summary
References
6 Type-2 Fuzzy Markov Random Fields
6.1 Markov Random Fields
6.1.1 The Neighborhood System
6.1.2 Clique Potentials
6.1.3 Relaxation Labeling
6.2 Type-2 Fuzzy Markov Random Fields
6.2.1 The Type-2 Fuzzy Relaxation Labeling
6.2.2 Computational Complexity
6.3 Stroke Segmentation of Chinese Character
6.3.1 Gabor Filters-Based Cyclic Observations
6.3.2 Stroke Segmentation Using MRFs
6.3.3 Stroke Extraction of Handprinted Chinese Characters.
6.3.4 Stroke Extraction of Cursive Chinese Characters
6.4 Handwritten Chinese Character Recognition
6.4.1 MRFs for Character Structure Modeling
6.4.2 Handwritten Chinese Character Recognition (HCCR).
6.4.3 Experimental Results
6.5 Summary
References
7 Type-2 Fuzzy Topic Models
7.1 Latent Dirichlet Allocation
7.1.1 Factor Graph for the Collapsed LDA
7.1.2 Loopy Belief Propagation (BP)
7.1.3 An Alternative View of BP
7.1.4 Simplified BP (siBP)
7.1.5 Relationship to Previous Algorithms
7.1.6 Belief Propagation for ATM
7.1.7 Belief Propagation for RTM
7.2 Speedup Topic Modeling
7.2.1 Fast Topic Modeling Techniques
7.2.2 Residual Belief Propagation
7.2.3 Active Belief Propagation
7.3 Type-2 Fuzzy Latent Dirichlet Allocation
7.3.1 Topic Models
7.3.2 Type-2 Fuzzy Topic Models (T2 FTMs)
7.4 Topic Modeling Performance
7.4.1 Belief Propagation
7.4.2 Residual Belief Propagation
7.4.3 Active Belief Propagation
7.5 Human Action Recognition
7.5.1 Feature Extraction and Vocabulary Formation
7.5.2 Results on KTH Data Set
References
8 Conclusions and Future Work
8.1 Conclusions
8.2 Future Works
Errata to: Type-2 Fuzzy Graphical Models for Pattern Recognition
鄂公网安备 42010302001612号 