Open Access
Table 1
Three types of methods to unmix hyperspectral images with example papers.
| Classic method | Stationary unmixing | Non-stationary unmixing | |
|---|---|---|---|
| General approach | One-dimensional spectral fit of individual pixels. | Matrix factorization. To each component is associated a spectral vector and an amplitude map | To each component is associated a spectral matrix and an amplitude map. |
| Correlation between pixels | No. Fits pixels individually. | Yes. Treats cube as a whole. | Yes. Treats cube as a whole. |
| Spectral variability | Yes. Varying spectral shapes for each component. | No. One spectral shape per component. | Yes. Varying spectral shapes for each component. |
| Examples in astrophysics | Lovisari et al. (2024), Mayer et al. (2023), Sasaki et al. (2022) | GMCA (Picquenot et al. 2019; Bobin et al. 2015) | ROHSA (Marchal et al. 2019), SUSHI (this work) |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.