(Downloads - 0)
For more info about our services contact : help@bestpfe.com
Table of contents
1 Introduction
1.1 Overview of Document Analysis
1.1.1 Shape Recognition
1.1.2 Classifier Fusion
1.2 Objectives and contribution of the present dissertation
1.2.1 Goals of the work
1.2.2 Framework of the thesis
1.3 Organization
2 Shape Descriptors
2.1 Introduction
2.2 Definitions
2.3 Taxonomy of Descriptors
2.3.1 Primitives of Descriptors
2.3.2 Multiresolution methods
2.3.3 Structural descriptors
2.4 Discussion
3 Ridgelets descriptors
3.1 Introduction
3.2 Theoretical Framework
3.2.1 Radon Transform
3.2.2 Multiresolution Analysis
3.3 Ridgelets Transform
3.3.1 Computation of Ridgelets coefficients
3.4 Image representation
3.4.1 Definition of a shape model
3.4.2 Definition of a similarity measure
3.4.3 Definition of a combination rule
3.5 Local Norm descriptors based on ridgelets transform
3.6 Discussion
4 Classifier Fusion
4.1 Introduction
4.2 Classifier Fusion approaches
4.2.1 Bayesian approach
4.2.2 Additive models: logistic regression approach
4.3 The problem of classifier fusion: definitions
4.4 Optimal Linear Combination Rules: IN and DN
4.4.1 IN method
4.4.2 DN method
4.5 Discussion
5 Experimental Evaluation
5.1 Introduction
5.2 Descriptors
5.3 Ridgelets descriptors
5.3.1 Robustness to resolution
5.3.2 Invariance to similarity transforms
5.3.3 Robustness to degradation and vectorial distortion
5.4 Combining Classifiers
5.4.1 Using synthetic data
5.4.2 Using Shape Databases
5.5 Combining ridgelets descriptors
5.5.1 Invariance to similarity transforms
5.5.2 Robustness to degradation and vectorial distortion
5.5.3 Discussion
6 Conclusions
6.1 Summary of the Contributions
6.2 Discussion and Conclusions
6.2.1 Ridgelets descriptors
6.2.2 Classifier Fusion
6.3 Open issues
6.4 Final Conclusion
