Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral Regularization
Due to the instrumental and imaging optics limitations, it is difficult to acquire high spatial resolution hyperspectral imagery (HSI). Super-resolution (SR) imagery aims at inferring high quality images of a given scene from degraded versions of the same scene. This paper proposes a novel hyperspec...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2015-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/15/1/2041 |
| _version_ | 1828109705211281408 |
|---|---|
| author | Wei Huang Liang Xiao Hongyi Liu Zhihui Wei |
| author_facet | Wei Huang Liang Xiao Hongyi Liu Zhihui Wei |
| author_sort | Wei Huang |
| collection | DOAJ |
| description | Due to the instrumental and imaging optics limitations, it is difficult to acquire high spatial resolution hyperspectral imagery (HSI). Super-resolution (SR) imagery aims at inferring high quality images of a given scene from degraded versions of the same scene. This paper proposes a novel hyperspectral imagery super-resolution (HSI-SR) method via dictionary learning and spatial-spectral regularization. The main contributions of this paper are twofold. First, inspired by the compressive sensing (CS) framework, for learning the high resolution dictionary, we encourage stronger sparsity on image patches and promote smaller coherence between the learned dictionary and sensing matrix. Thus, a sparsity and incoherence restricted dictionary learning method is proposed to achieve higher efficiency sparse representation. Second, a variational regularization model combing a spatial sparsity regularization term and a new local spectral similarity preserving term is proposed to integrate the spectral and spatial-contextual information of the HSI. Experimental results show that the proposed method can effectively recover spatial information and better preserve spectral information. The high spatial resolution HSI reconstructed by the proposed method outperforms reconstructed results by other well-known methods in terms of both objective measurements and visual evaluation. |
| first_indexed | 2024-04-11T11:07:46Z |
| format | Article |
| id | doaj.art-b4527a7fa2614ce88b33ab9a57af656f |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T11:07:46Z |
| publishDate | 2015-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-b4527a7fa2614ce88b33ab9a57af656f2022-12-22T04:28:14ZengMDPI AGSensors1424-82202015-01-011512041205810.3390/s150102041s150102041Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral RegularizationWei Huang0Liang Xiao1Hongyi Liu2Zhihui Wei3School of Computer Science and Engineering, Nanjing University of Science & Technology, Nanjing 210094, ChinaSchool of Computer Science and Engineering, Nanjing University of Science & Technology, Nanjing 210094, ChinaSchool of Science, Nanjing University of Science & Technology, Nanjing 210094, ChinaSchool of Computer Science and Engineering, Nanjing University of Science & Technology, Nanjing 210094, ChinaDue to the instrumental and imaging optics limitations, it is difficult to acquire high spatial resolution hyperspectral imagery (HSI). Super-resolution (SR) imagery aims at inferring high quality images of a given scene from degraded versions of the same scene. This paper proposes a novel hyperspectral imagery super-resolution (HSI-SR) method via dictionary learning and spatial-spectral regularization. The main contributions of this paper are twofold. First, inspired by the compressive sensing (CS) framework, for learning the high resolution dictionary, we encourage stronger sparsity on image patches and promote smaller coherence between the learned dictionary and sensing matrix. Thus, a sparsity and incoherence restricted dictionary learning method is proposed to achieve higher efficiency sparse representation. Second, a variational regularization model combing a spatial sparsity regularization term and a new local spectral similarity preserving term is proposed to integrate the spectral and spatial-contextual information of the HSI. Experimental results show that the proposed method can effectively recover spatial information and better preserve spectral information. The high spatial resolution HSI reconstructed by the proposed method outperforms reconstructed results by other well-known methods in terms of both objective measurements and visual evaluation.http://www.mdpi.com/1424-8220/15/1/2041compressive sensingdictionary learningsuper-resolutionhyperspectral imagespectral similaritysparse representation |
| spellingShingle | Wei Huang Liang Xiao Hongyi Liu Zhihui Wei Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral Regularization Sensors compressive sensing dictionary learning super-resolution hyperspectral image spectral similarity sparse representation |
| title | Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral Regularization |
| title_full | Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral Regularization |
| title_fullStr | Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral Regularization |
| title_full_unstemmed | Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral Regularization |
| title_short | Hyperspectral Imagery Super-Resolution by Compressive Sensing Inspired Dictionary Learning and Spatial-Spectral Regularization |
| title_sort | hyperspectral imagery super resolution by compressive sensing inspired dictionary learning and spatial spectral regularization |
| topic | compressive sensing dictionary learning super-resolution hyperspectral image spectral similarity sparse representation |
| url | http://www.mdpi.com/1424-8220/15/1/2041 |
| work_keys_str_mv | AT weihuang hyperspectralimagerysuperresolutionbycompressivesensinginspireddictionarylearningandspatialspectralregularization AT liangxiao hyperspectralimagerysuperresolutionbycompressivesensinginspireddictionarylearningandspatialspectralregularization AT hongyiliu hyperspectralimagerysuperresolutionbycompressivesensinginspireddictionarylearningandspatialspectralregularization AT zhihuiwei hyperspectralimagerysuperresolutionbycompressivesensinginspireddictionarylearningandspatialspectralregularization |