Block-Gaussian-Mixture Priors for Hyperspectral Denoising and Inpainting

• Tipo: Artigo em Revista Científica Internacional
• Data: 2021

Autores

Ana Teodoro (Autor) FCUP; Bioucas Dias, JM (Autor) Outra; Figueiredo, MAT (Autor) Outra

Revista

Título: IEEE Transactions on Geoscience and Remote Sensing

Vol. 59

Páginas: 2478-2486

ISSN: 0196-2892

Editora: IEEE

Indexação

ISI Web of Knowledge - 11 Citações

Scopus - 11 Citações

Outras Informações

ID Authenticus: P-00T-M1G

DOI: 10.1109/tgrs.2020.3006757

Abstract (EN): This article proposes a denoiser for hyperspectral (HS) images that consider, not only spatial features, but also spectral features. The method starts by projecting the noisy (observed) HS data onto a lower dimensional subspace and then learns a Gaussian mixture model (GMM) from 3-D patches or blocks extracted from the projected data cube. Afterward, the minimum mean squared error (MMSE) estimates of the blocks are obtained in closed form and returned to their original positions. Experiments show that the proposed algorithm is able to outperform other state-of-the-art methods under Gaussian and Poissonian noise and to reconstruct high-quality images in the presence of stripes.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 9