Skip to Main content Skip to Navigation
Journal articles

Learning the Morphological Diversity

Abstract : This article proposes a new method for image separation into a linear combination of morphological components. Sparsity in global dictionaries is used to extract the cartoon and oscillating content of the image. Complicated texture patterns are extracted by learning adapted local dictionaries that sparsify patches in the image. These global and local sparsity priors together with the data fidelity define a non-convex energy and the separation is obtained as a stationary point of this energy. This variational optimization is extended to solve more general inverse problems such as inpainting. A new adaptive morphological component analysis algorithm is derived to find a stationary point of the energy. Using adapted dictionaries learned from data allows to circumvent some difficulties faced by fixed dictionaries. Numerical results demonstrate that this adaptivity is indeed crucial to capture complex texture patterns.
Complete list of metadata

Cited literature [55 references]  Display  Hide  Download
Contributor : Gabriel Peyré Connect in order to contact the contributor
Submitted on : Wednesday, July 7, 2010 - 10:35:03 PM
Last modification on : Tuesday, January 18, 2022 - 3:24:22 PM
Long-term archiving on: : Thursday, December 1, 2016 - 5:02:25 AM


Files produced by the author(s)



Gabriel Peyré, Jalal M. Fadili, Jean-Luc Starck. Learning the Morphological Diversity. SIAM Journal on Imaging Sciences, Society for Industrial and Applied Mathematics, 2010, 3 (3), pp.646-669. ⟨10.1137/090770783⟩. ⟨hal-00415782v2⟩



Record views


Files downloads