Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral Data
In our previous work, a two-phase multiobjective sparse unmixing (Tp-MoSU) approach has been proposed, which settled the regularization parameter issues of the regularization unmixing methods. However, Tp-MoSU has limited performance in identifying the real end-members from the highly noisy data in...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9337884/ |
| _version_ | 1818893140971487232 |
|---|---|
| author | Xiangming Jiang Maoguo Gong Tao Zhan Kai Sheng Mingliang Xu |
| author_facet | Xiangming Jiang Maoguo Gong Tao Zhan Kai Sheng Mingliang Xu |
| author_sort | Xiangming Jiang |
| collection | DOAJ |
| description | In our previous work, a two-phase multiobjective sparse unmixing (Tp-MoSU) approach has been proposed, which settled the regularization parameter issues of the regularization unmixing methods. However, Tp-MoSU has limited performance in identifying the real end-members from the highly noisy data in the first phase and cannot effectively exploit the spatial-contextual information in the second phase because of the similarity measure it used. To settle these two problems, a composite spectral similarity measure is first constructed by fusing the spectral correlation angle and the Euclidean distance. It is used instead of the Frobenius norm to measure the unmixing residuals in the first phase because it considers both the shape and amplitude discrepancy between two spectra simultaneously. Then, the L<sub>2,∞</sub> norm is used instead of the l<sub>2</sub> norm to measure the unmixing residuals in the second phase, and the initialization, recombination, mutation, and local search strategies are also elaborately redesigned to help reduce this new objective, based on which the unmixing tasks of all pixels in a hyperspectral image can be completed at once. Therefore, this new measure facilitates the estimation of the abundances as a whole, and thus, the spatial-contextual information can be better exploited to improve the estimated abundances. Besides, the time efficiency for abundance estimation is also greatly improved. Experimental results demonstrate that the proposed method (termed as TpMoSU+) outperforms Tp-MoSU in both of the two phases under heavy noise and outperforms the tested regularization algorithms in estimating the abundances. |
| first_indexed | 2024-12-19T18:07:52Z |
| format | Article |
| id | doaj.art-ced9ba6e5cdd4e9f850a151cb940acf9 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-19T18:07:52Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-ced9ba6e5cdd4e9f850a151cb940acf92022-12-21T20:11:24ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01142418243110.1109/JSTARS.2021.30549269337884Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral DataXiangming Jiang0https://orcid.org/0000-0002-4650-1308Maoguo Gong1https://orcid.org/0000-0002-0415-8556Tao Zhan2https://orcid.org/0000-0002-9283-4488Kai Sheng3Mingliang Xu4https://orcid.org/0000-0002-6885-3451School of Electronic Engineering, Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education, Xidian University, Xi’an, ChinaSchool of Electronic Engineering, Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education, Xidian University, Xi’an, ChinaSchool of Computer Science and Technology, Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education, Xidian University, Xi’an, ChinaAcademy of Advanced Interdisciplinary Research, Xidian University, Xi’an, ChinaSchool of Information Engineering, Zhengzhou University, Zhengzhou, ChinaIn our previous work, a two-phase multiobjective sparse unmixing (Tp-MoSU) approach has been proposed, which settled the regularization parameter issues of the regularization unmixing methods. However, Tp-MoSU has limited performance in identifying the real end-members from the highly noisy data in the first phase and cannot effectively exploit the spatial-contextual information in the second phase because of the similarity measure it used. To settle these two problems, a composite spectral similarity measure is first constructed by fusing the spectral correlation angle and the Euclidean distance. It is used instead of the Frobenius norm to measure the unmixing residuals in the first phase because it considers both the shape and amplitude discrepancy between two spectra simultaneously. Then, the L<sub>2,∞</sub> norm is used instead of the l<sub>2</sub> norm to measure the unmixing residuals in the second phase, and the initialization, recombination, mutation, and local search strategies are also elaborately redesigned to help reduce this new objective, based on which the unmixing tasks of all pixels in a hyperspectral image can be completed at once. Therefore, this new measure facilitates the estimation of the abundances as a whole, and thus, the spatial-contextual information can be better exploited to improve the estimated abundances. Besides, the time efficiency for abundance estimation is also greatly improved. Experimental results demonstrate that the proposed method (termed as TpMoSU+) outperforms Tp-MoSU in both of the two phases under heavy noise and outperforms the tested regularization algorithms in estimating the abundances.https://ieeexplore.ieee.org/document/9337884/Composite spectral similarity measurehighly noisy data<inline-formula xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <tex-math notation="LaTeX">$ {L_{2,\infty }}$</tex-math> </inline-formula> normmultiobjective sparse unmixingspatial-contextual informationtime efficiency |
| spellingShingle | Xiangming Jiang Maoguo Gong Tao Zhan Kai Sheng Mingliang Xu Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral Data IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Composite spectral similarity measure highly noisy data <inline-formula xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <tex-math notation="LaTeX">$ {L_{2,\infty }}$</tex-math> </inline-formula> norm multiobjective sparse unmixing spatial-contextual information time efficiency |
| title | Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral Data |
| title_full | Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral Data |
| title_fullStr | Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral Data |
| title_full_unstemmed | Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral Data |
| title_short | Efficient Two-Phase Multiobjective Sparse Unmixing Approach for Hyperspectral Data |
| title_sort | efficient two phase multiobjective sparse unmixing approach for hyperspectral data |
| topic | Composite spectral similarity measure highly noisy data <inline-formula xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <tex-math notation="LaTeX">$ {L_{2,\infty }}$</tex-math> </inline-formula> norm multiobjective sparse unmixing spatial-contextual information time efficiency |
| url | https://ieeexplore.ieee.org/document/9337884/ |
| work_keys_str_mv | AT xiangmingjiang efficienttwophasemultiobjectivesparseunmixingapproachforhyperspectraldata AT maoguogong efficienttwophasemultiobjectivesparseunmixingapproachforhyperspectraldata AT taozhan efficienttwophasemultiobjectivesparseunmixingapproachforhyperspectraldata AT kaisheng efficienttwophasemultiobjectivesparseunmixingapproachforhyperspectraldata AT mingliangxu efficienttwophasemultiobjectivesparseunmixingapproachforhyperspectraldata |