AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images

The semantic segmentation of high-resolution remote sensing images (HRRSIs) faces persistent challenges in handling complex architectural structures and shadow occlusions, limiting the effectiveness of existing deep learning approaches. To address these limitations, we propose an attention-focused f...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiarui Li, Shuli Cheng
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/16/23/4392
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846123947156832256
author Jiarui Li
Shuli Cheng
author_facet Jiarui Li
Shuli Cheng
author_sort Jiarui Li
collection DOAJ
description The semantic segmentation of high-resolution remote sensing images (HRRSIs) faces persistent challenges in handling complex architectural structures and shadow occlusions, limiting the effectiveness of existing deep learning approaches. To address these limitations, we propose an attention-focused feature enhancement network (AFENet) with a novel encoder–decoder architecture. The encoder architecture combines ResNet50 with a parallel multistage feature enhancement group (PMFEG), enabling robust feature extraction through optimized channel reduction, scale expansion, and channel reassignment operations. Building upon this foundation, we develop a global multi-scale attention mechanism (GMAM) in the decoder that effectively synthesizes spatial information across multiple scales by learning comprehensive global–local relationships. The architecture is further enhanced by an efficient feature-weighted fusion module (FWFM) that systematically integrates remote spatial features with local semantic information to improve segmentation accuracy. Experimental results across diverse scenarios demonstrate that AFENet achieves superior performance in building structure detection, exhibiting enhanced segmentation connectivity and completeness compared to state-of-the-art methods.
format Article
id doaj-art-df3e0b92646445f096a883ea99803585
institution Kabale University
issn 2072-4292
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-df3e0b92646445f096a883ea998035852024-12-13T16:30:41ZengMDPI AGRemote Sensing2072-42922024-11-011623439210.3390/rs16234392AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing ImagesJiarui Li0Shuli Cheng1School of Computer Science and Technology, Xinjiang University, Urumq 830046, ChinaSchool of Computer Science and Technology, Xinjiang University, Urumq 830046, ChinaThe semantic segmentation of high-resolution remote sensing images (HRRSIs) faces persistent challenges in handling complex architectural structures and shadow occlusions, limiting the effectiveness of existing deep learning approaches. To address these limitations, we propose an attention-focused feature enhancement network (AFENet) with a novel encoder–decoder architecture. The encoder architecture combines ResNet50 with a parallel multistage feature enhancement group (PMFEG), enabling robust feature extraction through optimized channel reduction, scale expansion, and channel reassignment operations. Building upon this foundation, we develop a global multi-scale attention mechanism (GMAM) in the decoder that effectively synthesizes spatial information across multiple scales by learning comprehensive global–local relationships. The architecture is further enhanced by an efficient feature-weighted fusion module (FWFM) that systematically integrates remote spatial features with local semantic information to improve segmentation accuracy. Experimental results across diverse scenarios demonstrate that AFENet achieves superior performance in building structure detection, exhibiting enhanced segmentation connectivity and completeness compared to state-of-the-art methods.https://www.mdpi.com/2072-4292/16/23/4392remote sensingsemantic segmentationmulti-scale featureattention mechanism
spellingShingle Jiarui Li
Shuli Cheng
AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images
Remote Sensing
remote sensing
semantic segmentation
multi-scale feature
attention mechanism
title AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images
title_full AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images
title_fullStr AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images
title_full_unstemmed AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images
title_short AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images
title_sort afenet an attention focused feature enhancement network for the efficient semantic segmentation of remote sensing images
topic remote sensing
semantic segmentation
multi-scale feature
attention mechanism
url https://www.mdpi.com/2072-4292/16/23/4392
work_keys_str_mv AT jiaruili afenetanattentionfocusedfeatureenhancementnetworkfortheefficientsemanticsegmentationofremotesensingimages
AT shulicheng afenetanattentionfocusedfeatureenhancementnetworkfortheefficientsemanticsegmentationofremotesensingimages