Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization
Hyperspectral (HS) unmixing is the process of decomposing an HS image into material-specific spectra (endmembers) and their spatial distributions (abundance maps). Existing unmixing methods have two limitations with respect to noise robustness. First, if the input HS image is highly noisy, even if t...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10476645/ |
| _version_ | 1797202220299583488 |
|---|---|
| author | Kazuki Naganuma Shunsuke Ono |
| author_facet | Kazuki Naganuma Shunsuke Ono |
| author_sort | Kazuki Naganuma |
| collection | DOAJ |
| description | Hyperspectral (HS) unmixing is the process of decomposing an HS image into material-specific spectra (endmembers) and their spatial distributions (abundance maps). Existing unmixing methods have two limitations with respect to noise robustness. First, if the input HS image is highly noisy, even if the balance between sparse and piecewise-smooth regularizations for abundance maps is carefully adjusted, noise may remain in the estimated abundance maps or undesirable artifacts may appear. Second, existing methods do not explicitly account for the effects of stripe noise, which is common in HS measurements, in their formulations, resulting in significant degradation of unmixing performance when such noise is present in the input HS image. To overcome these limitations, we propose a new robust HS unmixing method based on constrained convex optimization. Our method employs, in addition to the two regularizations for the abundance maps, regularizations for the HS image reconstructed by mixing the estimated abundance maps and endmembers. This strategy makes the unmixing process much more robust in highly noisy scenarios, under the assumption that the abundance maps used to reconstruct the HS image with desirable spatio-spectral structure are also expected to have desirable properties. Furthermore, our method is designed to accommodate a wider variety of noise including stripe noise. To solve the formulated optimization problem, we develop an efficient algorithm based on a preconditioned primal-dual splitting method, which can automatically determine appropriate stepsizes based on the problem structure. Experiments on synthetic and real HS images demonstrate the advantages of our method over existing methods. |
| first_indexed | 2024-04-24T07:59:58Z |
| format | Article |
| id | doaj.art-f76faa7e81984555a6ed79c5a764f35f |
| institution | Directory Open Access Journal |
| issn | 1939-1404 2151-1535 |
| language | English |
| last_indexed | 2024-04-24T07:59:58Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-f76faa7e81984555a6ed79c5a764f35f2024-04-17T23:00:09ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352024-01-01178117813810.1109/JSTARS.2024.337955810476645Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain RegularizationKazuki Naganuma0https://orcid.org/0000-0002-7180-3017Shunsuke Ono1https://orcid.org/0000-0001-7890-5131Department of Computer Science, Tokyo Institute of Technology, Yokohama, JapanDepartment of Computer Science, Tokyo Institute of Technology, Yokohama, JapanHyperspectral (HS) unmixing is the process of decomposing an HS image into material-specific spectra (endmembers) and their spatial distributions (abundance maps). Existing unmixing methods have two limitations with respect to noise robustness. First, if the input HS image is highly noisy, even if the balance between sparse and piecewise-smooth regularizations for abundance maps is carefully adjusted, noise may remain in the estimated abundance maps or undesirable artifacts may appear. Second, existing methods do not explicitly account for the effects of stripe noise, which is common in HS measurements, in their formulations, resulting in significant degradation of unmixing performance when such noise is present in the input HS image. To overcome these limitations, we propose a new robust HS unmixing method based on constrained convex optimization. Our method employs, in addition to the two regularizations for the abundance maps, regularizations for the HS image reconstructed by mixing the estimated abundance maps and endmembers. This strategy makes the unmixing process much more robust in highly noisy scenarios, under the assumption that the abundance maps used to reconstruct the HS image with desirable spatio-spectral structure are also expected to have desirable properties. Furthermore, our method is designed to accommodate a wider variety of noise including stripe noise. To solve the formulated optimization problem, we develop an efficient algorithm based on a preconditioned primal-dual splitting method, which can automatically determine appropriate stepsizes based on the problem structure. Experiments on synthetic and real HS images demonstrate the advantages of our method over existing methods.https://ieeexplore.ieee.org/document/10476645/Constrained optimizationhyperspectral (HS) unmixingmixed noiseprimal-dual splittingstripe noise |
| spellingShingle | Kazuki Naganuma Shunsuke Ono Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Constrained optimization hyperspectral (HS) unmixing mixed noise primal-dual splitting stripe noise |
| title | Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization |
| title_full | Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization |
| title_fullStr | Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization |
| title_full_unstemmed | Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization |
| title_short | Toward Robust Hyperspectral Unmixing: Mixed Noise Modeling and Image-Domain Regularization |
| title_sort | toward robust hyperspectral unmixing mixed noise modeling and image domain regularization |
| topic | Constrained optimization hyperspectral (HS) unmixing mixed noise primal-dual splitting stripe noise |
| url | https://ieeexplore.ieee.org/document/10476645/ |
| work_keys_str_mv | AT kazukinaganuma towardrobusthyperspectralunmixingmixednoisemodelingandimagedomainregularization AT shunsukeono towardrobusthyperspectralunmixingmixednoisemodelingandimagedomainregularization |