Total Variation Regularization for Super-Resolution Reconstruction Using Two Norm Formulations
- Login to Download
- 1 Credits
Resource Overview
Detailed Documentation
Total variation regularization super-resolution reconstruction can be implemented using two different norm-based regularization methods as mentioned in the text. These approaches utilize the L1-norm and L2-norm formulations respectively. The primary objective of total variation regularization in super-resolution reconstruction is to enhance image resolution by regularizing the image variation, which effectively preserves edges while reducing noise amplification. This methodology finds broad applications in image processing domains and has been demonstrated to significantly improve image quality through constrained optimization techniques. Implementation typically involves solving a convex optimization problem where the data fidelity term combines with the regularization term, often using gradient descent algorithms or specialized solvers like the Split Bregman method for efficient computation.
- Login to Download
- 1 Credits