A novel mosaic adaptive attention network for multispectral imaging reconstruction

Due to the sparse sampling of each spectral band in MSFA, the more spectral bands, the lower the spatial resolution of each frequency band. It is still a challenge to obtain high-resolution snapshot images in both spectral and spatial domains in a short exposure. In addition, the resolution of exist...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhongqiang Zhang, Fanyang Meng, Dahua Gao, Haoyong Li
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Geocarto International
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/10106049.2025.2493233
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849323701059715072
author Zhongqiang Zhang
Fanyang Meng
Dahua Gao
Haoyong Li
author_facet Zhongqiang Zhang
Fanyang Meng
Dahua Gao
Haoyong Li
author_sort Zhongqiang Zhang
collection DOAJ
description Due to the sparse sampling of each spectral band in MSFA, the more spectral bands, the lower the spatial resolution of each frequency band. It is still a challenge to obtain high-resolution snapshot images in both spectral and spatial domains in a short exposure. In addition, the resolution of existing narrow band MSFA-based imaging systems in the spectral dimension is limited by the number of filters. The number of filters is equal to the number of bands in a multispectral image. To overcome these difficulties, we propose a novel multispectral imaging system. By sampling the spectra with random broadband filter-based MSFA, more spectral information is preserved in the generated spectral mosaic images. We further present an end-to-end mosaic adaptive attention network (MAAN) for multispectral image reconstruction. It contains a mosaic convolution block (MCB), four residual dense attention blocks (RDAB), and two spatial attention blocks (SAB). The MCB is used to extract mosaic features based on mosaic patterns. The RDAB is designed to learn deeper mosaic features, where channel attention block (CAB) can highlight the useful channel feature information. The SAB can enhance the useful spatial structure features. The experimental results show that the proposed MAAN algorithm outperforms existing CNN-based multispectral image reconstruction algorithms.
format Article
id doaj-art-fb450d48685d4aaeaac363087337606e
institution Kabale University
issn 1010-6049
1752-0762
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series Geocarto International
spelling doaj-art-fb450d48685d4aaeaac363087337606e2025-08-20T03:48:57ZengTaylor & Francis GroupGeocarto International1010-60491752-07622025-12-0140110.1080/10106049.2025.2493233A novel mosaic adaptive attention network for multispectral imaging reconstructionZhongqiang Zhang0Fanyang Meng1Dahua Gao2Haoyong Li3Department of Broadband Communication, Peng Cheng Laboratory, Shenzhen, Guangdong, PR ChinaDepartment of Broadband Communication, Peng Cheng Laboratory, Shenzhen, Guangdong, PR ChinaSchool of Artificial Intelligence, Xidian University, Xi’an, Shaanxi, PR ChinaSchool of Artificial Intelligence, Xidian University, Xi’an, Shaanxi, PR ChinaDue to the sparse sampling of each spectral band in MSFA, the more spectral bands, the lower the spatial resolution of each frequency band. It is still a challenge to obtain high-resolution snapshot images in both spectral and spatial domains in a short exposure. In addition, the resolution of existing narrow band MSFA-based imaging systems in the spectral dimension is limited by the number of filters. The number of filters is equal to the number of bands in a multispectral image. To overcome these difficulties, we propose a novel multispectral imaging system. By sampling the spectra with random broadband filter-based MSFA, more spectral information is preserved in the generated spectral mosaic images. We further present an end-to-end mosaic adaptive attention network (MAAN) for multispectral image reconstruction. It contains a mosaic convolution block (MCB), four residual dense attention blocks (RDAB), and two spatial attention blocks (SAB). The MCB is used to extract mosaic features based on mosaic patterns. The RDAB is designed to learn deeper mosaic features, where channel attention block (CAB) can highlight the useful channel feature information. The SAB can enhance the useful spatial structure features. The experimental results show that the proposed MAAN algorithm outperforms existing CNN-based multispectral image reconstruction algorithms.https://www.tandfonline.com/doi/10.1080/10106049.2025.2493233Multispectral imagemosaic adaptive attention networkmosaic convolution blockresidual dense attention block
spellingShingle Zhongqiang Zhang
Fanyang Meng
Dahua Gao
Haoyong Li
A novel mosaic adaptive attention network for multispectral imaging reconstruction
Geocarto International
Multispectral image
mosaic adaptive attention network
mosaic convolution block
residual dense attention block
title A novel mosaic adaptive attention network for multispectral imaging reconstruction
title_full A novel mosaic adaptive attention network for multispectral imaging reconstruction
title_fullStr A novel mosaic adaptive attention network for multispectral imaging reconstruction
title_full_unstemmed A novel mosaic adaptive attention network for multispectral imaging reconstruction
title_short A novel mosaic adaptive attention network for multispectral imaging reconstruction
title_sort novel mosaic adaptive attention network for multispectral imaging reconstruction
topic Multispectral image
mosaic adaptive attention network
mosaic convolution block
residual dense attention block
url https://www.tandfonline.com/doi/10.1080/10106049.2025.2493233
work_keys_str_mv AT zhongqiangzhang anovelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction
AT fanyangmeng anovelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction
AT dahuagao anovelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction
AT haoyongli anovelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction
AT zhongqiangzhang novelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction
AT fanyangmeng novelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction
AT dahuagao novelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction
AT haoyongli novelmosaicadaptiveattentionnetworkformultispectralimagingreconstruction