Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance Maps
Spectral unmixing (SU) aims at decomposing the mixed pixel into basic components, called endmembers with corresponding abundance fractions. Linear mixing model (LMM) and nonlinear mixing models (NLMMs) are two main classes to solve the SU. This paper proposes a new nonlinear unmixing method base on...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/24/4117 |
| _version_ | 1797544494193704960 |
|---|---|
| author | Zhicheng Wang Lina Zhuang Lianru Gao Andrea Marinoni Bing Zhang Michael K. Ng |
| author_facet | Zhicheng Wang Lina Zhuang Lianru Gao Andrea Marinoni Bing Zhang Michael K. Ng |
| author_sort | Zhicheng Wang |
| collection | DOAJ |
| description | Spectral unmixing (SU) aims at decomposing the mixed pixel into basic components, called endmembers with corresponding abundance fractions. Linear mixing model (LMM) and nonlinear mixing models (NLMMs) are two main classes to solve the SU. This paper proposes a new nonlinear unmixing method base on general bilinear model, which is one of the NLMMs. Since retrieving the endmembers’ abundances represents an ill-posed inverse problem, prior knowledge of abundances has been investigated by conceiving regularizations techniques (e.g., sparsity, total variation, group sparsity, and low rankness), so to enhance the ability to restrict the solution space and thus to achieve reliable estimates. All the regularizations mentioned above can be interpreted as denoising of abundance maps. In this paper, instead of investing effort in designing more powerful regularizations of abundances, we use plug-and-play prior technique, that is to use directly a state-of-the-art denoiser, which is conceived to exploit the spatial correlation of abundance maps and nonlinear interaction maps. The numerical results in simulated data and real hyperspectral dataset show that the proposed method can improve the estimation of abundances dramatically compared with state-of-the-art nonlinear unmixing methods. |
| first_indexed | 2024-03-10T14:01:19Z |
| format | Article |
| id | doaj.art-b8c3762dc0ec4cbdb953ff7b445af49d |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T14:01:19Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-b8c3762dc0ec4cbdb953ff7b445af49d2023-11-21T01:07:51ZengMDPI AGRemote Sensing2072-42922020-12-011224411710.3390/rs12244117Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance MapsZhicheng Wang0Lina Zhuang1Lianru Gao2Andrea Marinoni3Bing Zhang4Michael K. Ng5The Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaThe Department of Mathematics, Hong Kong Baptist University, Hong Kong, ChinaThe Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaThe Department of Physics and Technology, UiT The Arctic University of Norway, NO-9037 Tromsø, NorwayThe Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaThe Department of Mathematics, The University of Hong Kong, Hong Kong, ChinaSpectral unmixing (SU) aims at decomposing the mixed pixel into basic components, called endmembers with corresponding abundance fractions. Linear mixing model (LMM) and nonlinear mixing models (NLMMs) are two main classes to solve the SU. This paper proposes a new nonlinear unmixing method base on general bilinear model, which is one of the NLMMs. Since retrieving the endmembers’ abundances represents an ill-posed inverse problem, prior knowledge of abundances has been investigated by conceiving regularizations techniques (e.g., sparsity, total variation, group sparsity, and low rankness), so to enhance the ability to restrict the solution space and thus to achieve reliable estimates. All the regularizations mentioned above can be interpreted as denoising of abundance maps. In this paper, instead of investing effort in designing more powerful regularizations of abundances, we use plug-and-play prior technique, that is to use directly a state-of-the-art denoiser, which is conceived to exploit the spatial correlation of abundance maps and nonlinear interaction maps. The numerical results in simulated data and real hyperspectral dataset show that the proposed method can improve the estimation of abundances dramatically compared with state-of-the-art nonlinear unmixing methods.https://www.mdpi.com/2072-4292/12/24/4117hyperspectral imageryplug-and-playdenoisingnonlinear unmixing |
| spellingShingle | Zhicheng Wang Lina Zhuang Lianru Gao Andrea Marinoni Bing Zhang Michael K. Ng Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance Maps Remote Sensing hyperspectral imagery plug-and-play denoising nonlinear unmixing |
| title | Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance Maps |
| title_full | Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance Maps |
| title_fullStr | Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance Maps |
| title_full_unstemmed | Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance Maps |
| title_short | Hyperspectral Nonlinear Unmixing by Using Plug-and-Play Prior for Abundance Maps |
| title_sort | hyperspectral nonlinear unmixing by using plug and play prior for abundance maps |
| topic | hyperspectral imagery plug-and-play denoising nonlinear unmixing |
| url | https://www.mdpi.com/2072-4292/12/24/4117 |
| work_keys_str_mv | AT zhichengwang hyperspectralnonlinearunmixingbyusingplugandplaypriorforabundancemaps AT linazhuang hyperspectralnonlinearunmixingbyusingplugandplaypriorforabundancemaps AT lianrugao hyperspectralnonlinearunmixingbyusingplugandplaypriorforabundancemaps AT andreamarinoni hyperspectralnonlinearunmixingbyusingplugandplaypriorforabundancemaps AT bingzhang hyperspectralnonlinearunmixingbyusingplugandplaypriorforabundancemaps AT michaelkng hyperspectralnonlinearunmixingbyusingplugandplaypriorforabundancemaps |