声明
摘要
ABSTRACT
CONTENT
CHAPTER 1 INTRODUCTION
1.1 Background and Significance
1.2 Research Scope
1.3 Main Contributions
1.4 Thesis Organization
CHAPTER 2 BASIC THEORY
2.1 Image Acquisition Model
2.2 Mathematical Modeling
2.3 Inverse Problem in Super-resolution Reconstruction
2.4 Super-resolution Reconstruction Procedure
2.5 Related Work
2.5.1 Frequency domain based image super resolution
2.5.2 Spatial domain based image super resolution
2.6 Evaluation Metrics
2.6.1 Subjective Estimation
2.6.2 Objective Estimation
CHAPTER 3 IMAGE INTERPOLATION USING NONLOCAL RESIDUAL ENHANCEMENT
3.1 Introduction
3.2 Methods
3.2.1 Curvature based image interpolation
3.2.2 Nonlocal Information Compensation
3.2.3 Residual Enhancement
3.3 Results
3.4 Conclusions
CHAPTER 4 ENHANCING THE DICTIONARY BASED SUPER RESOLUTION USING NONOLCAL TOTAL VARIATION
4.1 Introduction
4.2 Related Work
4.3 Methods
4.3.1 KSVD dictionary learning
4.3.2 Nonlocal total variation regularization
4.3.3 Iterative back projection
4.4 Results
4.5 Conclusions
CHAPTER 5 COMPRESSION ARTIFACT REMOVAL OF HIGHLY COMPRESSED IMAGES WITH DE-BLOCKING DICTIONARY AND NONLOCAL TOTAL VARIATION
5.1 Introduction
5.2 Methods
5.2.1 Dictionary Learning
5.2.2 Error Threshold Estimation
5.2.3 Non-local Total Variation Regularization
5.3 Results
5.4 Conclusions
CHAPTER 6 DISCUSSION
6.1 Summary
6.2 Future Work
APPENDIX A IMAGE SUPER RESOLUTION SYSTEM
REFERENCES
ACKNOWLEDGEMENTS
RESEARCH ACHIEVEMENTS