Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixing
Abstract Advances have been achieved in hyperspectral unmixing using the existing manifold Non‐negative Matrix Factorisation methods, although most of these methods only exploit the preliminary structural information, that is, the nearest neighbour graph. Consequently, the performance of these metho...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-08-01
|
| Series: | IET Computer Vision |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/cvi2.12168 |
| _version_ | 1827863135078318080 |
|---|---|
| author | Xue Li Sifan Cao Dan Huang Ming Zhang Yiwei Li |
| author_facet | Xue Li Sifan Cao Dan Huang Ming Zhang Yiwei Li |
| author_sort | Xue Li |
| collection | DOAJ |
| description | Abstract Advances have been achieved in hyperspectral unmixing using the existing manifold Non‐negative Matrix Factorisation methods, although most of these methods only exploit the preliminary structural information, that is, the nearest neighbour graph. Consequently, the performance of these methods would be degraded when considering only the geometrical structure due to the diverse distribution of the hyperspectral data, that is, the close pixels could belong to different categories or the distant points could be sampled from the same classes. In this context, the present study worked from the perspective of both global and local data relationships to develop and propose a novel approach—the Global centralised and Structured discriminative Non‐negative Matrix Factorisation (GSNMF)—to achieve a further effective representation of hyperspectral unmixing. GSNMF involved maintaining the global centralised clustering and the local structured discriminative regularisation, based on which it could perfectly mine the structure information and drive a discriminative representation of the data. Experiments comparing the application of GSNMF and the state‐of‐the‐art methods to synthetic data demonstrated the superiority of GSNMF. In addition, the consistency of the fractional abundances obtained using GSNMF with the real distributions of spectral data was evaluated on two real‐world datasets. |
| first_indexed | 2024-03-12T14:04:35Z |
| format | Article |
| id | doaj.art-3d3b0027dc5e4109a9c131f31d755ef6 |
| institution | Directory Open Access Journal |
| issn | 1751-9632 1751-9640 |
| language | English |
| last_indexed | 2024-03-12T14:04:35Z |
| publishDate | 2023-08-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Computer Vision |
| spelling | doaj.art-3d3b0027dc5e4109a9c131f31d755ef62023-08-21T14:42:49ZengWileyIET Computer Vision1751-96321751-96402023-08-0117554956410.1049/cvi2.12168Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixingXue Li0Sifan Cao1Dan Huang2Ming Zhang3Yiwei Li4School of Computer Science and Technology Anhui University of Technology Ma'anshan ChinaDepartment of Electronic Engineering Chinese University of HongKong HongKong ChinaChina Research and Development Academy of Machinery Equipment Beijing ChinaJiangsu Expressway Company Limited Nanjing ChinaCollege of Science and Technology Wenzhou‐Kean University Wenzhou ChinaAbstract Advances have been achieved in hyperspectral unmixing using the existing manifold Non‐negative Matrix Factorisation methods, although most of these methods only exploit the preliminary structural information, that is, the nearest neighbour graph. Consequently, the performance of these methods would be degraded when considering only the geometrical structure due to the diverse distribution of the hyperspectral data, that is, the close pixels could belong to different categories or the distant points could be sampled from the same classes. In this context, the present study worked from the perspective of both global and local data relationships to develop and propose a novel approach—the Global centralised and Structured discriminative Non‐negative Matrix Factorisation (GSNMF)—to achieve a further effective representation of hyperspectral unmixing. GSNMF involved maintaining the global centralised clustering and the local structured discriminative regularisation, based on which it could perfectly mine the structure information and drive a discriminative representation of the data. Experiments comparing the application of GSNMF and the state‐of‐the‐art methods to synthetic data demonstrated the superiority of GSNMF. In addition, the consistency of the fractional abundances obtained using GSNMF with the real distributions of spectral data was evaluated on two real‐world datasets.https://doi.org/10.1049/cvi2.12168clusteringfeature extractionhyperspectral unmixinglocal affinitynon‐negative matrix factorisation |
| spellingShingle | Xue Li Sifan Cao Dan Huang Ming Zhang Yiwei Li Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixing IET Computer Vision clustering feature extraction hyperspectral unmixing local affinity non‐negative matrix factorisation |
| title | Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixing |
| title_full | Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixing |
| title_fullStr | Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixing |
| title_full_unstemmed | Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixing |
| title_short | Global centralised and structured discriminative non‐negative matrix factorisation for hyperspectral unmixing |
| title_sort | global centralised and structured discriminative non negative matrix factorisation for hyperspectral unmixing |
| topic | clustering feature extraction hyperspectral unmixing local affinity non‐negative matrix factorisation |
| url | https://doi.org/10.1049/cvi2.12168 |
| work_keys_str_mv | AT xueli globalcentralisedandstructureddiscriminativenonnegativematrixfactorisationforhyperspectralunmixing AT sifancao globalcentralisedandstructureddiscriminativenonnegativematrixfactorisationforhyperspectralunmixing AT danhuang globalcentralisedandstructureddiscriminativenonnegativematrixfactorisationforhyperspectralunmixing AT mingzhang globalcentralisedandstructureddiscriminativenonnegativematrixfactorisationforhyperspectralunmixing AT yiweili globalcentralisedandstructureddiscriminativenonnegativematrixfactorisationforhyperspectralunmixing |