Open AccessWeighted total generalised variation scheme for image restoration
Author(s) -
Liu Xinwu
Publication year - 2016
Publication title -
iet image processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.401
H-Index - 45
eISSN - 1751-9667
pISSN - 1751-9659
DOI - 10.1049/iet-ipr.2015.0013
Subject(s) - smoothing , convergence (economics) , scheme (mathematics) , variation (astronomy) , image (mathematics) , feature (linguistics) , image restoration , mathematical optimization , mathematics , algorithm , computer science , image processing , artificial intelligence , computer vision , mathematical analysis , linguistics , philosophy , physics , astrophysics , economics , economic growth
Total generalised variation (TGV) methods are highly efficient for eliminating the staircase artefacts. However, with the aim of further avoiding over‐smoothing edges, this study investigates a new weighted second‐order TGV scheme for image restoration. Computationally, an alternating split Bregman algorithm is employed to obtain the optimal solution recursively. Moreover, the rigorous convergence analysis of the resulting algorithm is also described in brief. In comparison with the results of current state‐of‐the‐art regulariser techniques, numerical simulations distinctly demonstrate the competitive performance of the proposed strategy in feature preservation and staircasing effect suppression.