2015 | OriginalPaper | Buchkapitel
Alternating Direction Method of Multiplier for Euler’s Elastica-Based Denoising
verfasst von : Maryam Yashtini, Sung Ha Kang
Erschienen in: Scale Space and Variational Methods in Computer Vision
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Inspired by recent numerical developments, we propose a new version of alternating direction method of multiplier (ADMM) for Euler’s Elastica-based denoising model. The main contribution is to design a simple and fast method, which it is also easy to choose its parameters values. regularizer for instance the so called staircasing effect. The solution of each subproblem is given in a closed form using a discrete Fourier transform, a soft shrinkage operator, and a coupled Fourier transform. Compared to other methods, this algorithm has less parameters and we provide some insight on their values. provide some insights on how their values need to be determined. Numerical experiments on image denoising application demonstrate the effectiveness of the proposed scheme.