Contents -- Foreword -- Preface -- 1. Non-parametric Mixture Model Based Evolution of Level Sets N. Joshi and M. Brady -- 1.1 Introduction -- 1.2 Need for Modelling Class Distributions Non-parametrically -- 1.3 NP-windows Method for Non-parametric Estimation of PDFs -- 1.4 NPMM-ICLS Framework -- 1.5 Level Sets Method -- 1.6 NPMM-ICLS Level Sets Method -- 1.7 Results and Discussion -- 1.8 Conclusions -- Bibliography -- 2. Pattern Generation Using Level Set Based Curve Evolution A. Chattopadhyay and D. P. Mukherjee -- 2.1 Introduction -- 2.2 Background -- 2.2.1 Level set model of curve evolution -- 2.2.2 Reaction-diffusion model -- 2.2.3 Shape optimization -- 2.3 Proposed Methodology -- 2.3.1 Reaction-diffusion influenced curve evolution -- 2.3.2 Shape optimization based curve evolution -- 2.4 Results -- 2.4.1 Pattern disocclusion -- 2.5 Conclusions -- Bibliography -- 3. Stable Contour Tracking Through Tangential Evolution V. Srikrishnan and S. Chaudhuri -- 3.1 Active Contours: Introduction -- 3.2 Curve Evolution -- 3.3 Difficulties with Parametric Curves -- 3.4 Existing Solutions -- 3.5 Proposed Method -- 3.5.1 Comparison with other works -- 3.5.2 Choice of the ideal constant K -- 3.5.3 Proof of conditional boundedness -- 3.6 Applications in Image Segmentation and Tracking -- 3.7 Implementation Details -- 3.8 Results -- 3.9 Conclusions and FutureWork -- Bibliography -- 4. Information Theoretic Approaches for Next Best View Planning in Active Computer Vision C. Derichs, B. Deutsch, S. Wenhardt, H. Niemann and J. Denzler -- 4.1 Introduction -- 4.2 Information Theoretical Approaches for Next Best View Planning -- 4.2.1 General state modeling and estimation -- 4.2.2 Optimality criteria for active view planning -- 4.3 Planning Tasks -- 4.3.1 Active object recognition -- 4.3.1.1 State representation and information fusion.

4.3.1.2 Optimal action selection -- 4.3.2 Active object tracking -- 4.3.2.1 State and observation representation -- 4.3.2.2 Optimal action selection -- 4.3.2.3 Visibility -- 4.3.2.4 Multi-step action selection -- 4.3.3 Active object reconstruction -- 4.3.3.1 State and observation representation -- 4.3.3.2 Optimal action selection -- 4.4 Experiments -- 4.4.1 Evaluation for active object recognition -- 4.4.2 Evaluation for active object tracking -- 4.4.3 Evaluation for active object reconstruction -- 4.4.3.1 Reconstructing a calibration pattern -- 4.4.3.2 Reconstructing a mouse pad -- 4.5 Summary -- Bibliography -- 5. Evaluation of Linear Combination of Views for Object Recognition V. Zografos and B. F. Buxton -- 5.1 Introduction -- 5.2 Linear Combination of Views -- 5.2.1 Image synthesis -- 5.3 The Recognition System -- 5.3.1 Template matching -- 5.3.2 Optimisation -- 5.4 Experimental Results -- 5.4.1 Experiments on the CMU PIE database -- 5.4.2 Experiments on the COIL-20 database -- 5.5 Conclusion -- Bibliography -- 6. Using Object Models as Domain Knowledge in Perceptual Organization G. Harit, R. Bharatia and S. Chaudhury -- 6.1 Introduction -- 6.2 Perceptual Grouping in Video -- 6.2.1 Video data clustering -- 6.2.2 The perceptual grouping model -- 6.3 Object Model as a Pictorial Structure -- 6.3.1 Formulation of the potential between object parts -- 6.3.2 Formulation of appearance parameters for object parts -- 6.4 Spatio-Temporal Grouping -- 6.4.1 Formulation of object model evidence -- 6.4.2 The grouping algorithm -- 6.5 Results -- 6.6 Conclusions -- Bibliography -- 7. Image Representations Based on Discriminant Non-negative Matrix Factorization I. Buciu and I. Pitas -- 7.1 Introduction -- 7.2 Bregman Distance, Kullback-Leibler Divergence and Non-negative Matrix Factorization -- 7.3 Local Non-negative Matrix Factorization.

7.4 Discriminant Non-negative Matrix Factorization -- 7.5 Facial Expression Recognition Experiment -- 7.5.1 Data description -- 7.5.2 Training procedure -- 7.5.3 Feature extraction and image representation -- 7.5.4 Test procedure -- 7.5.5 Classification procedure -- 7.6 Performance Evaluation and Discussions -- 7.7 Conclusion -- Bibliography -- 8. Duplicate Image Detection in Large Scale Databases P. Ghosh, E. D. Gelasca, K. R. Ramakrishnan and B. S. Manjunath -- 8.1 Introduction -- 8.2 Related Work -- 8.3 System Overview -- 8.3.1 CFMT descriptor for images -- 8.3.2 CFMT extraction for arbitrarily shaped regions -- 8.4 Experimental Results -- 8.4.1 Performance evaluation -- 8.4.2 Results on web image database -- 8.4.3 Time performance -- 8.4.4 Results on MM270K image database -- 8.5 Conclusion and FutureWork -- Bibliography -- 9. Unsupervised Change Detection Techniques Based on Self-Organizing Feature Map Neural Network S. Patra, S. Ghosh and A. Ghosh -- 9.1 Introduction -- 9.2 Kohonen's Model of Self-Organizing Feature Map -- 9.3 Proposed Change Detection Techniques -- 9.3.1 Change detection based on 1D-SOFM -- 9.3.2 Change detection based on 2D-SOFM -- 9.3.2.1 Learning the weights -- 9.3.2.2 Proposed threshold selection techniques -- 9.4 Description of the Data Sets -- 9.4.1 Data set related to Mexico area -- 9.4.2 Data set related to Sardinia Island, Italy -- 9.5 Experimental Results -- 9.5.1 Description of experiments -- 9.5.2 Results on Mexico data -- 9.5.3 Results on Sardinia Island data -- 9.6 Discussion and Conclusion -- Bibliography -- 10. Recent Advances in Video Compression L. Liu, F. Zhu, M. Bosch, and E. J. Delp -- 10.1 Introduction and Overview of Video Coding Standards -- 10.2 Distributed Video Coding -- 10.2.1 Wyner-Ziv video coding -- 10.2.2 Rate distortion analysis -- 10.2.3 Backward channel aware Wyner-Ziv video coding.

10.3 Texture-Based Video Coding -- 10.3.1 Spatial texture models -- 10.3.2 Temporal qnalysis -- 10.3.3 A new perspective on texture-based video coding -- 10.4 Scalable Video Coding -- 10.4.1 Temporal scalability -- 10.4.2 Spatial scalability -- 10.4.3 SNR and rate scalability -- 10.5 Multi-View Coding -- 10.6 Conclusion -- Bibliography -- 11. Hardware Architecture for Ridge Extraction in Fingerprints A. Bishnu, P. Bhowmick, J. Dey, B. B. Bhattacharya, M. K. Kundu, C. A. Murthy, and T. Acharya -- 11.1 Introduction -- 11.2 Proposed Method -- 11.2.1 Theme -- 11.2.2 Combinatorial possibilities -- 11.2.3 Implementation details -- 11.2.4 Classification of a pixel -- 11.2.4.1 Preliminary classification -- 11.2.4.2 Final classification -- 11.2.5 Thinning -- 11.2.6 Algorithm -- 11.3 Evaluation and Results -- 11.3.1 Evaluation criteria for ridge/valley finding algorithms -- 11.3.2 Results -- 11.4 Parallel Algorithm and Hardware Architecture -- 11.4.1 Parallel algorithm -- 11.4.2 Hardware architecture -- 11.4.3 Time complexity -- 11.4.4 Circuit cost -- 11.5 Discussions and Conclusions -- Bibliography -- 12. Rough-Fuzzy Hybridization for Protein Sequence Analysis P. Maji and S. K. Pal -- 12.1 Introduction -- 12.2 Bio-Basis Function, Rough Set, and Fuzzy Set -- 12.2.1 Bio-basis function -- 12.2.2 Rough sets -- 12.2.3 Fuzzy set -- 12.3 Rough-Fuzzy C-Medoids Algorithm -- 12.3.1 Hard C-medoids -- 12.3.2 Fuzzy C-medoids -- 12.3.3 Rough C-medoids -- 12.3.4 Rough-fuzzy C-medoids -- 12.3.5 Selection of initial bio-basis -- 12.4 Quantitative Measure -- 12.4.1 Using homology alignment score -- 12.4.2 Using mutual information -- 12.5 Experimental Results -- 12.5.1 Description of data set -- 12.5.1.1 Five whole HIV protein sequences -- 12.5.1.2 Cai-Chou HIV data set -- 12.5.1.3 Caspase cleavage data set -- 12.5.2 Example -- 12.5.3 Performance analysis.

12.5.3.1 Optimum value of parameter 3 -- 12.5.3.2 Random versus DOR based initialization -- 12.5.3.3 Optimum values of parameters m, w, and -- 12.5.3.4 Comparative performance of different algorithms -- 12.6 Conclusion -- 13. Knowledge Reuse in the Design of Models of Computational Intelligence W. Pedrycz -- 13.1 Introduction -- 13.2 Problem Formulation -- 13.2.1 Performance index -- 13.2.2 Optimization of the level of knowledge reuse -- 13.2.3 Refinement of knowledge reuse -- 13.3 Fuzzy Rule-based Models and Their Experience-consistent Design -- 13.4 The Alignment of Information Granules -- 13.5 Granular Characterization of Experience-consistent Rule-based Models -- 13.6 Conclusions -- Bibliography -- Index.