A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image Compression
Deep learning-based remote sensing image compression methods show great potential, but traditional convolutional networks mainly focus on local feature extraction and show obvious limitations in dynamic feature learning and global context modeling. Remote sensing images contain multiscale local feat...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Spolecnost pro radioelektronicke inzenyrstvi
2025-06-01
|
| Series: | Radioengineering |
| Subjects: | |
| Online Access: | https://www.radioeng.cz/fulltexts/2025/25_02_0324_0341.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849421506644279296 |
|---|---|
| author | Q. Z. Fang S. B. Gu J. G. Wang L. L. Zhang |
| author_facet | Q. Z. Fang S. B. Gu J. G. Wang L. L. Zhang |
| author_sort | Q. Z. Fang |
| collection | DOAJ |
| description | Deep learning-based remote sensing image compression methods show great potential, but traditional convolutional networks mainly focus on local feature extraction and show obvious limitations in dynamic feature learning and global context modeling. Remote sensing images contain multiscale local features and global low-frequency information, which are challenging to extract and fuse efficiently. To address this, we propose a Feature Dynamic Enhancement and Global Collaboration Guidance Network (FDEGCNet). First, we propose an Omni-Dimensional Attention Model (ODAM), which dynamically captures the key salient features in the image content by adaptively adjusting the feature extraction strategy to enhance the model’s sensitivity to key information. Second, a Hyperprior Efficient Attention Model (HEAM) is designed to combine multi-directional convolution and pooling operations to efficiently capture cross-dimensional contextual information and facilitate the interaction and fusion of multi-scale features. Finally, the Multi-Kernel Convolutional Attention Model (MCAM) integrates global branching to extract frequency domain context and enhance local feature representation through multi-scale convolutions. The experimental results show that FDEGCNet achieves significant improvement and maintains low computational complexity regarding image quality evaluation metrics (PSNR, MSSSIM, LPIPS, and VIFp) compared to the advanced compression models. Code is available at https://github.com/shiboGu12/FDEGCNet |
| format | Article |
| id | doaj-art-0b209f4cc50e457e8b3eb6bb4bbaf322 |
| institution | Kabale University |
| issn | 1210-2512 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Spolecnost pro radioelektronicke inzenyrstvi |
| record_format | Article |
| series | Radioengineering |
| spelling | doaj-art-0b209f4cc50e457e8b3eb6bb4bbaf3222025-08-20T03:31:26ZengSpolecnost pro radioelektronicke inzenyrstviRadioengineering1210-25122025-06-01342324341A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image CompressionQ. Z. FangS. B. GuJ. G. WangL. L. ZhangDeep learning-based remote sensing image compression methods show great potential, but traditional convolutional networks mainly focus on local feature extraction and show obvious limitations in dynamic feature learning and global context modeling. Remote sensing images contain multiscale local features and global low-frequency information, which are challenging to extract and fuse efficiently. To address this, we propose a Feature Dynamic Enhancement and Global Collaboration Guidance Network (FDEGCNet). First, we propose an Omni-Dimensional Attention Model (ODAM), which dynamically captures the key salient features in the image content by adaptively adjusting the feature extraction strategy to enhance the model’s sensitivity to key information. Second, a Hyperprior Efficient Attention Model (HEAM) is designed to combine multi-directional convolution and pooling operations to efficiently capture cross-dimensional contextual information and facilitate the interaction and fusion of multi-scale features. Finally, the Multi-Kernel Convolutional Attention Model (MCAM) integrates global branching to extract frequency domain context and enhance local feature representation through multi-scale convolutions. The experimental results show that FDEGCNet achieves significant improvement and maintains low computational complexity regarding image quality evaluation metrics (PSNR, MSSSIM, LPIPS, and VIFp) compared to the advanced compression models. Code is available at https://github.com/shiboGu12/FDEGCNethttps://www.radioeng.cz/fulltexts/2025/25_02_0324_0341.pdfremote sensing image compressionconvolutional networksmultiscale convolutionattention modelmultiscale local featuresglobal low-frequency information |
| spellingShingle | Q. Z. Fang S. B. Gu J. G. Wang L. L. Zhang A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image Compression Radioengineering remote sensing image compression convolutional networks multiscale convolution attention model multiscale local features global low-frequency information |
| title | A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image Compression |
| title_full | A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image Compression |
| title_fullStr | A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image Compression |
| title_full_unstemmed | A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image Compression |
| title_short | A Feature Dynamic Enhancement and Global Collaboration Guidance Network for Remote Sensing Image Compression |
| title_sort | feature dynamic enhancement and global collaboration guidance network for remote sensing image compression |
| topic | remote sensing image compression convolutional networks multiscale convolution attention model multiscale local features global low-frequency information |
| url | https://www.radioeng.cz/fulltexts/2025/25_02_0324_0341.pdf |
| work_keys_str_mv | AT qzfang afeaturedynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression AT sbgu afeaturedynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression AT jgwang afeaturedynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression AT llzhang afeaturedynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression AT qzfang featuredynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression AT sbgu featuredynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression AT jgwang featuredynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression AT llzhang featuredynamicenhancementandglobalcollaborationguidancenetworkforremotesensingimagecompression |