DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite images
Abstract Semantic segmentation of high-resolution images from remote sensing is crucial across various sectors. However, due to limitations in computational resources and the complexity of network architectures, many sophisticated semantic segmentation models struggle with efficiency in real-world a...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-84134-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559664294100992 |
|---|---|
| author | Naga Surekha Jonnala Renuka Chowdary Bheemana Krishna Prakash Shonak Bansal Arpit Jain Vaibhav Pandey Mohammad Rashed Iqbal Faruque K. S. Al-mugren |
| author_facet | Naga Surekha Jonnala Renuka Chowdary Bheemana Krishna Prakash Shonak Bansal Arpit Jain Vaibhav Pandey Mohammad Rashed Iqbal Faruque K. S. Al-mugren |
| author_sort | Naga Surekha Jonnala |
| collection | DOAJ |
| description | Abstract Semantic segmentation of high-resolution images from remote sensing is crucial across various sectors. However, due to limitations in computational resources and the complexity of network architectures, many sophisticated semantic segmentation models struggle with efficiency in real-world applications, leading to an interest in developing lightweight model like borders. These models often employ a dual-branch structure, which balances processing speed and performance effectively. Yet, this design typically falls short in leveraging shallow structural information to enrich the dual branches with comprehensive multiscale data. Additionally, the lightweight components struggle to capture the global contextual details of feature sets efficiently. When compared to state-of-the-art models, lightweight semantic segmentation models usually exhibit performance gaps. To address these issues, we introduce a novel approach that incorporates a deep-shallow interaction mechanism with an attention module to improve water body segmentation efficiency. This method spatially adjusts feature representations to better identify water-related data, utilizing a U-Net frame work to enhance the accuracy of edge detection in water zones by providing more precise local positioning information. The attention mechanism processes and merges low and high-level data separately in different dimensions, allowing for the effective distinction of water areas from their surroundings by blending spatial attributes with in-depth context insights. Experimental outcomes demonstrate a remarkable 95% accuracy, showcasing the proposed method’s superiority over existing models. |
| format | Article |
| id | doaj-art-6822e613b29b410da56d2e64bdcc0caa |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-6822e613b29b410da56d2e64bdcc0caa2025-01-05T12:16:52ZengNature PortfolioScientific Reports2045-23222025-01-0115111710.1038/s41598-024-84134-4DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite imagesNaga Surekha Jonnala0Renuka Chowdary Bheemana1Krishna Prakash2Shonak Bansal3Arpit Jain4Vaibhav Pandey5Mohammad Rashed Iqbal Faruque6K. S. Al-mugren7Department of ECE, NRI Institute of TechnologyDepartment of ECE, NRI Institute of TechnologyDepartment of ECE, NRI Institute of TechnologyDepartment of Electronics and Communication Engineering, Chandigarh UniversityDepartment of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation VadeshawaramDepartment of Computer Science and Engineering, Thapar Institute of Engineering and TechnologySpace Science Centre (ANGKASA), Institute of Climate Change (IPI), Universiti Kebangsaan MalaysiaPhysics Department, Science College, Princess Nourah Bint Abdulrahman UniversityAbstract Semantic segmentation of high-resolution images from remote sensing is crucial across various sectors. However, due to limitations in computational resources and the complexity of network architectures, many sophisticated semantic segmentation models struggle with efficiency in real-world applications, leading to an interest in developing lightweight model like borders. These models often employ a dual-branch structure, which balances processing speed and performance effectively. Yet, this design typically falls short in leveraging shallow structural information to enrich the dual branches with comprehensive multiscale data. Additionally, the lightweight components struggle to capture the global contextual details of feature sets efficiently. When compared to state-of-the-art models, lightweight semantic segmentation models usually exhibit performance gaps. To address these issues, we introduce a novel approach that incorporates a deep-shallow interaction mechanism with an attention module to improve water body segmentation efficiency. This method spatially adjusts feature representations to better identify water-related data, utilizing a U-Net frame work to enhance the accuracy of edge detection in water zones by providing more precise local positioning information. The attention mechanism processes and merges low and high-level data separately in different dimensions, allowing for the effective distinction of water areas from their surroundings by blending spatial attributes with in-depth context insights. Experimental outcomes demonstrate a remarkable 95% accuracy, showcasing the proposed method’s superiority over existing models.https://doi.org/10.1038/s41598-024-84134-4Semantic segmentationU-NetAttention moduleSatellite imagesDeep learningNeural network |
| spellingShingle | Naga Surekha Jonnala Renuka Chowdary Bheemana Krishna Prakash Shonak Bansal Arpit Jain Vaibhav Pandey Mohammad Rashed Iqbal Faruque K. S. Al-mugren DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite images Scientific Reports Semantic segmentation U-Net Attention module Satellite images Deep learning Neural network |
| title | DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite images |
| title_full | DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite images |
| title_fullStr | DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite images |
| title_full_unstemmed | DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite images |
| title_short | DSIA U-Net: deep shallow interaction with attention mechanism UNet for remote sensing satellite images |
| title_sort | dsia u net deep shallow interaction with attention mechanism unet for remote sensing satellite images |
| topic | Semantic segmentation U-Net Attention module Satellite images Deep learning Neural network |
| url | https://doi.org/10.1038/s41598-024-84134-4 |
| work_keys_str_mv | AT nagasurekhajonnala dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages AT renukachowdarybheemana dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages AT krishnaprakash dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages AT shonakbansal dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages AT arpitjain dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages AT vaibhavpandey dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages AT mohammadrashediqbalfaruque dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages AT ksalmugren dsiaunetdeepshallowinteractionwithattentionmechanismunetforremotesensingsatelliteimages |