Spatial–Spectral Multiscale Sparse Unmixing for Hyperspectral Images - Signaux et Images Access content directly
Journal Articles IEEE Geoscience and Remote Sensing Letters Year : 2023

Spatial–Spectral Multiscale Sparse Unmixing for Hyperspectral Images

Abstract

We propose a simple yet efficient sparse unmixing method for hyperspectral images. It exploits the spatial and spectral properties of hyperspectral images by designing a new regularization informed by multiscale analysis. The proposed approach consists of two steps. First, a sparse unmixing is conducted on a coarse hyperspectral image resulting from a spatial smoothing of the original data. The estimated coarse abundance map is subsequently used to design two weighting terms summarizing the spatial and spectral properties of the image. They are combined to define a sparse regularization embedded into a unmixing problem associated with the original hyperspectral image at full resolution. The performance of the proposed method is assessed with numerous experiments conducted on synthetic and real data sets. It is shown to compete favorably with state-of-the-art methods from the literature with lower computational complexity.
Fichier principal
Vignette du fichier
Ince_IEEE_GRS_Letters_2023b.pdf (4.2 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04280429 , version 1 (11-11-2023)

Identifiers

Cite

Taner Ince, Nicolas Dobigeon. Spatial–Spectral Multiscale Sparse Unmixing for Hyperspectral Images. IEEE Geoscience and Remote Sensing Letters, 2023, 20, pp.1-5. ⟨10.1109/LGRS.2023.3328370⟩. ⟨hal-04280429⟩
63 View
22 Download

Altmetric

Share

Gmail Facebook X LinkedIn More