Graph regularised sparse NMF factorisation for imagery de‐noising
When utilising non‐negative matrix factorisation (NMF) to decompose a data matrix into the product of two low‐rank matrices with non‐negative entries, the noisy components of data may be introduced into the matrix. Many approaches have been proposed to address the problem. Different from them, the a...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2018-06-01
|
| Series: | IET Computer Vision |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/iet-cvi.2017.0263 |
| _version_ | 1827817580680708096 |
|---|---|
| author | Yixian Fang Huaxiang Zhang Yuwei Ren |
| author_facet | Yixian Fang Huaxiang Zhang Yuwei Ren |
| author_sort | Yixian Fang |
| collection | DOAJ |
| description | When utilising non‐negative matrix factorisation (NMF) to decompose a data matrix into the product of two low‐rank matrices with non‐negative entries, the noisy components of data may be introduced into the matrix. Many approaches have been proposed to address the problem. Different from them, the authors consider the group sparsity and the geometric structure of data by introducing ℓ2,1‐norm and local structure preserving regularisation in the formulated objective function. A graph regularised sparse NMF de‐noising approach is proposed to learn discriminative representations for the original data. Since the non‐differentiability of ℓ2,1‐norm increases the computational cost, they propose an effective iterative multiplicative update algorithm to solve the objective function by using the Frobenius‐norm of transpose coefficient matrix. Experimental results on facial image datasets demonstrate the superiority of the proposed approach over several state‐of‐the‐art approaches. |
| first_indexed | 2024-03-12T00:36:16Z |
| format | Article |
| id | doaj.art-8a80875d6fe54281b85fcbeab0765cca |
| institution | Directory Open Access Journal |
| issn | 1751-9632 1751-9640 |
| language | English |
| last_indexed | 2024-03-12T00:36:16Z |
| publishDate | 2018-06-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Computer Vision |
| spelling | doaj.art-8a80875d6fe54281b85fcbeab0765cca2023-09-15T09:36:27ZengWileyIET Computer Vision1751-96321751-96402018-06-0112446647510.1049/iet-cvi.2017.0263Graph regularised sparse NMF factorisation for imagery de‐noisingYixian Fang0Huaxiang Zhang1Yuwei Ren2School of Information Science and Engineering, Shandong Normal UniversityJinan250014Shandong ProvincePeople's Republic of ChinaSchool of Information Science and Engineering, Shandong Normal UniversityJinan250014Shandong ProvincePeople's Republic of ChinaSchool of Information Science and Engineering, Shandong Normal UniversityJinan250014Shandong ProvincePeople's Republic of ChinaWhen utilising non‐negative matrix factorisation (NMF) to decompose a data matrix into the product of two low‐rank matrices with non‐negative entries, the noisy components of data may be introduced into the matrix. Many approaches have been proposed to address the problem. Different from them, the authors consider the group sparsity and the geometric structure of data by introducing ℓ2,1‐norm and local structure preserving regularisation in the formulated objective function. A graph regularised sparse NMF de‐noising approach is proposed to learn discriminative representations for the original data. Since the non‐differentiability of ℓ2,1‐norm increases the computational cost, they propose an effective iterative multiplicative update algorithm to solve the objective function by using the Frobenius‐norm of transpose coefficient matrix. Experimental results on facial image datasets demonstrate the superiority of the proposed approach over several state‐of‐the‐art approaches.https://doi.org/10.1049/iet-cvi.2017.0263graph regularised sparse NMF factorisationimagery denoisingnonnegative matrix factorisationdata matrix decompositiontwo low-rank matrices productgroup sparsity |
| spellingShingle | Yixian Fang Huaxiang Zhang Yuwei Ren Graph regularised sparse NMF factorisation for imagery de‐noising IET Computer Vision graph regularised sparse NMF factorisation imagery denoising nonnegative matrix factorisation data matrix decomposition two low-rank matrices product group sparsity |
| title | Graph regularised sparse NMF factorisation for imagery de‐noising |
| title_full | Graph regularised sparse NMF factorisation for imagery de‐noising |
| title_fullStr | Graph regularised sparse NMF factorisation for imagery de‐noising |
| title_full_unstemmed | Graph regularised sparse NMF factorisation for imagery de‐noising |
| title_short | Graph regularised sparse NMF factorisation for imagery de‐noising |
| title_sort | graph regularised sparse nmf factorisation for imagery de noising |
| topic | graph regularised sparse NMF factorisation imagery denoising nonnegative matrix factorisation data matrix decomposition two low-rank matrices product group sparsity |
| url | https://doi.org/10.1049/iet-cvi.2017.0263 |
| work_keys_str_mv | AT yixianfang graphregularisedsparsenmffactorisationforimagerydenoising AT huaxiangzhang graphregularisedsparsenmffactorisationforimagerydenoising AT yuweiren graphregularisedsparsenmffactorisationforimagerydenoising |