Contents -- Foreword -- Preface -- 1. Combining Information with a Bayesian Multi-class Multi-kernel Pattern Recognition Machine T. Damoulas and M. A. Girolami -- Contents -- 1.1. Introduction -- Problem area -- Past methods -- Proposed method -- 1.2. Intuition and Motivation -- 1.3. Multinomial Probit Kernel Combination -- 1.3.1. Markov chain Monte Carlo solution -- 1.3.2. Variational Bayes approximation -- Inference and intuition -- 1.4. Experiments and Results -- 1.4.1. Proof of concept on an artificial data-set -- 1.4.2. UCI and handwritten numerals data-sets -- 1.4.3. Protein fold recognition and remote homology detection -- Fold recognition -- Remote homology detection (RHD) -- 1.5. Discussion and Future Directions -- Funding -- Acknowledgment -- References -- 2. Image Quality Assessment Based on Weighted Perceptual Features D. V. Rao and L. P. Reddy -- Contents -- 2.1. Introduction -- 2.1.1. Statistical metrics -- 2.1.2. Error sensitivity metrics -- 2.1.3. Coder specific metrics -- 2.1.4. Transform based metrics -- 2.1.5. Structural approaches -- 2.2. A Model for Weighted Perceptual Features on SSIM -- 2.2.1. Image distortion model -- 2.3. Perceptual Structural Similarity Model -- 2.3.1. Weighing function based on edge strength -- 2.3.2. Perceptual structural similarity based on edge strength -- 2.3.3. Weighing function based on texture -- 2.3.4. Perceptual structural similarity based on texture -- 2.3.5. Weighing function based on contrast -- 2.3.6. Perceptual structural similarity based on local contrast -- 2.3.7. Weighing function of perceptual features -- 2.3.8. Perceptual Structural Similarity index -- 2.4. Performance Evaluation of PSSIM -- 2.5. Conclusions and Future Scope -- References -- 3. Quasi-reversible Two-dimension Fractional Differentiation for Image Entropy Reduction A. Nakib, A. Nait-Ali, H. Oulhadj and P. Siarry.

Contents -- 3.1. Introduction -- 3.2. Fractional Differentiation -- 3.3. Extension of the Fractional Differentiation to the 2D Case and its Application in Entropy Reduction -- 3.4. Statistical Properties of the Differentiated Image -- 3.4.1. Average value of the differentiated image -- 3.4.2. Entropy -- 3.5. Choice of the Optimal Order -- 3.6. Conclusion -- References -- 4. Parallel Genetic Algorithm Based Clustering for Object and Background Classification P. Kanungo, P. K. Nanda and A. Ghosh -- Contents -- 4.1. Introduction -- 4.2. Determination of Niches by Clustering -- 4.3. Proposed Methods -- 4.3.1. Featureless (FL) approach -- 4.3.2. Featured Based (FB) approach -- 4.4. GA Class Model -- 4.4.1. Crowding method -- 4.4.2. Generalized crossover (GC) -- 4.4.3. GA based clustering -- 4.4.4. Parallel Genetic Algorithm (PGA) -- 4.4.5. PGA based clustering -- 4.4.6. Algorithm -- 4.5. Results and Discussions -- 4.6. Conclusion -- Acknowledgments -- References -- 5. Bipolar Fuzzy Spatial Information: First Operations in the Mathematical Morphology Setting I. Bloch -- Contents -- 5.1. Introduction -- 5.2. Preliminaries -- 5.3. Algebraic Dilation and Erosion of Bipolar Fuzzy Sets -- 5.4. Morphological Erosion of Bipolar Fuzzy Sets -- 5.5. Morphological Dilation of Bipolar Fuzzy Sets -- 5.5.1. Dilation by duality -- 5.5.2. Dilation by adjunction -- 5.5.3. Links between both approaches -- 5.6. Properties and Interpretation -- 5.7. Illustrative Example -- 5.8. Derived Operators -- 5.8.1. Morphological gradient -- 5.8.2. Conditional dilation -- 5.8.3. Opening, closing and derived operators -- 5.9. Conclusion -- References -- 6. Approaches to Intelligent Information Retrieval G. Pasi -- Contents -- 6.1. Introduction -- 6.2. Information Retrieval -- 6.3. Personalization and Context Modeling -- 6.3.1. Content-based information filltering.

6.4. Recent Approaches to Textual Content Representation -- 6.5. An Approach to Personalized Document Indexing -- 6.6. Conclusions -- References -- 7. Retrieval of On-line Signatures H. N. Prakash and D. S. Guru -- Contents -- 7.1. Introduction -- 7.1.1. Signature analysis -- 7.1.2. Related work -- 7.1.2.1. Signature verification -- 7.1.2.2. Signature recognition -- 7.1.3. Signature retrieval -- 7.2. Retrival Model -- 7.2.1. Signature representation -- 7.2.2. Retrieval -- 7.2.2.1. Edge orientation based model -- 7.2.2.2. TSR based model -- 7.3. Experimentation -- 7.4. Conclusions -- Acknowledgment -- Appendix -- References -- 8. A Two Stage Recognition Scheme for Offline Handwritten Devanagari Words B. Shaw and S. K. Parui -- Keywords: -- Contents -- 8.1. Introduction -- 8.2. Devanagari Script -- 8.3. Handwritten Devanagari Word Database -- 8.4. Recognition Scheme Stage I -- 8.4.1. Pre-processing and feature extraction -- 8.4.1.1. Preprocessing -- 8.4.1.2. Extraction of strokes -- 8.4.1.3. Extraction of features -- 8.4.2. HMM classifier for handwritten word recognition -- 8.4.2.1. Hidden Markov models -- 8.4.2.2. Literature review of HMM classifiers for word recognition -- 8.4.3. Proposed HMM classifier -- 8.4.3.1. HMM parameters -- 8.4.3.2. Estimation of HMM parameters -- 8.4.4. Experimental results for stage I -- 8.5. Recognition Scheme Stage II -- 8.5.1. Grouping scheme -- 8.5.2. Wavelet features -- 8.5.3. Modified Bayes classifier -- 8.6. Conclusion -- References -- 9. Fall Detection from a Video in the Presence of Multiple Persons V. Vishwakarma, S. Sural and C. Mandal -- Contents -- 9.1. Introduction -- 9.2. Related Work -- 9.3. Detecting Fall in a Video Containing a Single Person -- 9.3.1. Human detection -- 9.3.1.1. Feature extraction -- 9.3.2. Fall model -- 9.3.2.1. Fall detection step -- 9.3.2.2. Fall confirmation step.

9.3.2.3. State transition -- 9.3.3. Experimental results -- 9.4. Detecting Fall in a Video Containing Multiple Humans -- 9.4.1. Foreground detection -- 9.4.2. Modeling of human shapes using ellipsoids -- 9.4.2.1. Feature extraction -- 9.4.3. Enhanced fall model -- 9.4.3.1. Fall detection step -- 9.4.3.2. Fall confirmation step -- 9.4.3.3. State transition -- 9.4.4. Experimental results -- 9.5. Conclusion and Future Work -- References -- 10. Fusion of GIS and SAR Statistical Features for Earthquake Damage Mapping at the Block Scale G. Trianni, G. Lisini, P. Gamba and F. Dell'Acqua -- Contents -- 10.1. Introduction -- 10.2. Damage Mapping Processing Chain -- 10.3. Damage Mapping Results -- 10.4. Conclusions -- Acknowledgments -- References -- 11. Intelligent Surveillance and Pose-invariant 2D Face Classification B. C. Lovell, C. Sanderson and T. Shan -- Contents -- 11.1. Introduction -- 11.2. Methods Based on ASMs and AAMs -- 11.2.1. Face modeling -- 11.2.2. Pose estimation -- 11.2.3. Frontal view synthesis -- 11.2.4. Direct pose-robust features -- 11.2.5. Remove pose effect using correlation model -- 11.2.5.1. Correlation model and pose estimation -- 11.2.5.2. Removing pose effect in appearance -- 11.2.6. Face recognition using pose-independent features -- 11.3. Methods Based on Bag-of-Features Approach -- 11.3.1. Feature extraction and illumination normalization -- 11.3.2. Bag-of-features with direct likelihood evaluation -- 11.3.3. Bag-of-features with histogram matching -- 11.4. Face Recognition Robust to Pose -- 11.5. NICTA Smart Camera -- 11.5.1. Proposed smart camera architecture -- 11.5.2. System design constraints -- 11.5.3. Hardware specification of NICTA smart camera prototype -- 11.6. Conclusions -- 11.7. Future Work -- Acknowledgments -- References.

12. Simple Machine Learning Approaches to Safety-related Systems C. Moewes, C. Otte and R. Kruse -- Contents -- 12.1. Introduction -- 12.1.1. Multiple-instance problem -- 12.1.2. Safety-related applications -- 12.1.3. Outline of this chapter -- 12.2. Separating Hyperplanes -- 12.2.1. Linear hyperplane classifiers -- 12.2.2. Soft margin hyperplanes -- 12.2.3. Support vector machines -- 12.3. Creating Prerequisites for Simple Models -- 12.3.1. Simplicity of support vector machines -- 12.3.2. Support vector pruning -- 12.3.3. Bag weighting -- 12.3.4. Combination of both methods -- 12.4. Conclusions -- Acknowledgment -- References -- 13. Nonuniform Multi Level Crossings for Signal Reconstruction N. Poojary, H. Kumar and A. Rao -- Contents -- 13.1. Introduction -- 13.2. Level Crossing Based Irregular Sampling Model -- 13.3. Weight Functions for Irregular Sampling -- 13.3.1. Linear function -- 13.3.2. Logarithmic function -- 13.3.3. Incomplete beta function -- 13.3.4. Level estimation -- 13.4. Experimental Evaluation -- 13.5. Conclusion -- References -- 14. Adaptive Web Services Brokering K. M. Gupta and D. W. Aha -- Contents -- 14.1. Introduction -- 14.2. Web Services Brokering Overview -- 14.3. A Meteorological and Oceanographic Application -- 14.4. Integrated Web Services Broker -- 14.4.1. User interface -- 14.4.2. Service discovery and mediation engine -- 14.4.3. Service discovery engine -- 14.4.3.1. Mediator -- 14.4.4. Knowledge base -- 14.5. Evaluations -- 14.5.1. Web service classi.cation evaluation -- 14.5.2. Web service method classification evaluation -- 14.5.3. Web service data category classification evaluation -- 14.5.4. Discussion -- 14.6. Conclusion -- Acknowledgment -- References.

15. Granular Support Vector Machine Based Method for Prediction of Solubility of Proteins on Over Expression in Escherichia Coli and Breast Cancer Classification P. Kumar, B. D. Kulkarni and V. K. Jayaraman.