Open AccessDeblurring Algorithm Using Alternating Low Rank Augmented Lagrangian with Iterative Priors
Author(s) -
Laya Tojo,
K Gurushankar,
Vivek Maik,
M. Nirmala Devi
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1964/6/062041
Subject(s) - deblurring , augmented lagrangian method , regularization (linguistics) , prior probability , rank (graph theory) , algorithm , mathematics , convergence (economics) , mathematical optimization , iterative method , computer science , image (mathematics) , artificial intelligence , image restoration , image processing , combinatorics , bayesian probability , economic growth , economics
The paper focuses on the Enhanced Augmented Lagrangian method with sparse regularization for image deblurring. The method suggested by ALTERNATING LOW RANK AUGMENTED LAGRANGIAN WITH ITERATIVE A PRIOR is novel in the following ways. (i) Faster convergence leading to speeder execution through rank regulations (ii) using derivatives and low rank together as regularization priors (iii) penalty and regularization weights ensure that each iteration hits a global minimum with a steep descent. The proposed method begins with the lowest rank matrix, which is the sparsest matrix available. The final deblurred result is very successful in achieving good dB improvements through rank regulation.