Spatial–Spectral Cross Fusion Attention based Hyperspectral Image Super-Resolution for Land Resource Auditing
Abstract
Hyperspectral imaging, celebrated for its detailed spectral information, finds broad application in various fields. However, the limitations inherent to optical systems often impede the direct acquisition of highresolution hyperspectral images. Hence, achieving these images has become a key focus in the research community. The process of single hyperspectral image super-resolution (HSI-SR) aims to upscale lowresolution images to a higher resolution. With the evolution of deep learning, the incorporation of Convolutional Neural Networks (CNNs) into super-resolution methods has shown considerable promise. Yet, the challenge lies in the thorough extraction of both spatial and spectral data, especially in the context of remote sensing, which can limit the model's ability to learn effectively. Additionally, Transformer-based techniques often struggle to capture the intricate relationships between spatial and spectral features, which can hinder the effectiveness of image reconstruction. To overcome these challenges, this paper presents a novel HSI-SR approach: Spatial–Spectral Cross Fusion Attention based Hyperspectral Image Super-Resolution for Land Resource Auditing, which synergizes the strengths of CNNs and the Transformer architecture. During the learning of spatial features, the method alternates between window self-attention and zero-padding window self-attention, allowing for a more comprehensive focus on feature information and the integration of different windows to achieve long-range insights. Furthermore, the cross-attention feature fusion module designed for this approach is adept at merging spatial and spectral features, thus enhancing the model's ability to learn from both types of information. The approach effectively enhances spatial-spectral integration, improving reconstruction quality. Extensive experimental assessments have demonstrated the proposed method's superiority over current industry benchmarks. PSNR improvements 0.08 over baseline in Cave.
DOI:
https://doi.org/10.31449/inf.v49i22.7634Downloads
Additional Files
Published
How to Cite
Issue
Section
License
I assign to Informatica, An International Journal of Computing and Informatics ("Journal") the copyright in the manuscript identified above and any additional material (figures, tables, illustrations, software or other information intended for publication) submitted as part of or as a supplement to the manuscript ("Paper") in all forms and media throughout the world, in all languages, for the full term of copyright, effective when and if the article is accepted for publication. This transfer includes the right to reproduce and/or to distribute the Paper to other journals or digital libraries in electronic and online forms and systems.
I understand that I retain the rights to use the pre-prints, off-prints, accepted manuscript and published journal Paper for personal use, scholarly purposes and internal institutional use.
In certain cases, I can ask for retaining the publishing rights of the Paper. The Journal can permit or deny the request for publishing rights, to which I fully agree.
I declare that the submitted Paper is original, has been written by the stated authors and has not been published elsewhere nor is currently being considered for publication by any other journal and will not be submitted for such review while under review by this Journal. The Paper contains no material that violates proprietary rights of any other person or entity. I have obtained written permission from copyright owners for any excerpts from copyrighted works that are included and have credited the sources in my article. I have informed the co-author(s) of the terms of this publishing agreement.
Copyright © Slovenian Society Informatika
