Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention Network
Hyperspectral imaging (HSI) captures detailed spectral data across a wide range of wavelengths, making HSI invaluable for many computer vision applications. However, limitations of imaging systems result in low spatial resolution in HSI images. State-of-the-art research encounters challenges due to...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11098921/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849340565789868032 |
|---|---|
| author | Allen Patnaik M. K. Bhuyan Sultan Alfarhood Mejdl Safran |
| author_facet | Allen Patnaik M. K. Bhuyan Sultan Alfarhood Mejdl Safran |
| author_sort | Allen Patnaik |
| collection | DOAJ |
| description | Hyperspectral imaging (HSI) captures detailed spectral data across a wide range of wavelengths, making HSI invaluable for many computer vision applications. However, limitations of imaging systems result in low spatial resolution in HSI images. State-of-the-art research encounters challenges due to complex textural patterns and inadequate low-frequency details in captured images. Moreover, high-fidelity image reconstruction is computationally intensive. To address these issues, we developed a grouped second-order spatial features and spectral attention network (GSSAN). GSSAN incorporates a grouped second-order spatial attention (GSOSA) module to extract relevant first- and second-order spatial features. We employed a grouped spectral attention (GSA) module to extract spectral details across different bands of HSI images. Rotary positional embedding (RoPE) is incorporated into our model to encode relative distances and angular rotations between features while preserving equivariance. These modules help reconstruct clear edges and contours in HSI. Our experimental results demonstrate that GSSAN significantly outperforms state-of-the-art HSISR methods, achieving improvements in mean peak signal-to-noise ratio (MPSNR) by up to 0.183 dB and mean structural similarity index by 0.051 on the Chikusei test dataset with a scaling factor of <inline-formula><tex-math notation="LaTeX">$\times 4$</tex-math></inline-formula>. The average inference time of our model is 16.09 s. These findings indicate an improved tradeoff between reconstruction quality and computational efficiency. |
| format | Article |
| id | doaj-art-dbab430a8cc840f0819c42e8ccf2b373 |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-dbab430a8cc840f0819c42e8ccf2b3732025-08-20T03:43:52ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118199741998710.1109/JSTARS.2025.359359411098921Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention NetworkAllen Patnaik0https://orcid.org/0009-0009-1418-9805M. K. Bhuyan1https://orcid.org/0000-0003-2152-5466Sultan Alfarhood2https://orcid.org/0009-0001-1268-9613Mejdl Safran3https://orcid.org/0000-0002-7445-7121Department of Electronics and Electrical Engineering, IIT Guwahati, Guwahati, IndiaDepartment of Electronics and Electrical Engineering, IIT Guwahati, Guwahati, IndiaDepartment of Computer Science, College of Computer and Information Sciences, King Saud University, Riyadh, Saudi ArabiaDepartment of Computer Science, College of Computer and Information Sciences, King Saud University, Riyadh, Saudi ArabiaHyperspectral imaging (HSI) captures detailed spectral data across a wide range of wavelengths, making HSI invaluable for many computer vision applications. However, limitations of imaging systems result in low spatial resolution in HSI images. State-of-the-art research encounters challenges due to complex textural patterns and inadequate low-frequency details in captured images. Moreover, high-fidelity image reconstruction is computationally intensive. To address these issues, we developed a grouped second-order spatial features and spectral attention network (GSSAN). GSSAN incorporates a grouped second-order spatial attention (GSOSA) module to extract relevant first- and second-order spatial features. We employed a grouped spectral attention (GSA) module to extract spectral details across different bands of HSI images. Rotary positional embedding (RoPE) is incorporated into our model to encode relative distances and angular rotations between features while preserving equivariance. These modules help reconstruct clear edges and contours in HSI. Our experimental results demonstrate that GSSAN significantly outperforms state-of-the-art HSISR methods, achieving improvements in mean peak signal-to-noise ratio (MPSNR) by up to 0.183 dB and mean structural similarity index by 0.051 on the Chikusei test dataset with a scaling factor of <inline-formula><tex-math notation="LaTeX">$\times 4$</tex-math></inline-formula>. The average inference time of our model is 16.09 s. These findings indicate an improved tradeoff between reconstruction quality and computational efficiency.https://ieeexplore.ieee.org/document/11098921/Hyperspectral image (HSI)positional embeddingsecond-order featurestotal variation (TV) loss |
| spellingShingle | Allen Patnaik M. K. Bhuyan Sultan Alfarhood Mejdl Safran Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention Network IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Hyperspectral image (HSI) positional embedding second-order features total variation (TV) loss |
| title | Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention Network |
| title_full | Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention Network |
| title_fullStr | Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention Network |
| title_full_unstemmed | Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention Network |
| title_short | Hyperspectral Image Super-Resolution via Grouped Second-Order Spatial Features and Spectral Attention Network |
| title_sort | hyperspectral image super resolution via grouped second order spatial features and spectral attention network |
| topic | Hyperspectral image (HSI) positional embedding second-order features total variation (TV) loss |
| url | https://ieeexplore.ieee.org/document/11098921/ |
| work_keys_str_mv | AT allenpatnaik hyperspectralimagesuperresolutionviagroupedsecondorderspatialfeaturesandspectralattentionnetwork AT mkbhuyan hyperspectralimagesuperresolutionviagroupedsecondorderspatialfeaturesandspectralattentionnetwork AT sultanalfarhood hyperspectralimagesuperresolutionviagroupedsecondorderspatialfeaturesandspectralattentionnetwork AT mejdlsafran hyperspectralimagesuperresolutionviagroupedsecondorderspatialfeaturesandspectralattentionnetwork |