Improved U-Net++ Semantic Segmentation Method for Remote Sensing Images

Remote sensing image semantic segmentation has extensive applications in land resource planning and smart cities. Due to the problems of unclear boundary segmentation and insufficient Semantic information of small targets in high-resolution remote sensing images, an improved network TU net based on...

Full description

Saved in:
Bibliographic Details
Main Authors: Yang Xu, Bin Cao, Hui Lu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10930761/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850273145721716736
author Yang Xu
Bin Cao
Hui Lu
author_facet Yang Xu
Bin Cao
Hui Lu
author_sort Yang Xu
collection DOAJ
description Remote sensing image semantic segmentation has extensive applications in land resource planning and smart cities. Due to the problems of unclear boundary segmentation and insufficient Semantic information of small targets in high-resolution remote sensing images, an improved network TU net based on U-net++ is proposed. Secondly, the attention aggregation module of the base Transformer is introduced to capture global contextual information, replacing the original multi-level skip connections of U-net++. A cross-window interaction module is designed, which significantly reduces computational complexity and achieves a lightweight model. Finally, a dynamic feature fusion block is designed at the end of the decoder to obtain multi-class and multi-scale Semantic information and enhance the final segmentation effect. TU-net conducted experiments on two datasets, where OA, mIoU, and mF1 scores were higher than mainstream models. The IoU and F1 scores of small-sized target cars in the Vaihingen dataset were 0.896 and 0.962, respectively, which were 5% and 15.8% higher than the suboptimal model; The IoU and F1 scores of the trees in the Potsdam dataset are 0.913 and 0.936, respectively, which are 6.3% and 4.3% higher than the suboptimal model. The experimental results show that the model can more accurately segment small-sized targets and target boundaries.
format Article
id doaj-art-d75addf8eddc472fb453c953c76e8b4b
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-d75addf8eddc472fb453c953c76e8b4b2025-08-20T01:51:35ZengIEEEIEEE Access2169-35362025-01-0113558775588610.1109/ACCESS.2025.355258110930761Improved U-Net++ Semantic Segmentation Method for Remote Sensing ImagesYang Xu0https://orcid.org/0000-0003-2758-1880Bin Cao1Hui Lu2Guiyang Aluminum Magnesium Design and Research Institute Company Ltd., Guiyang, ChinaSchool of Big Data and Information Engineering, Guizhou University, Guiyang, ChinaGuiyang Aluminum Magnesium Design and Research Institute Company Ltd., Guiyang, ChinaRemote sensing image semantic segmentation has extensive applications in land resource planning and smart cities. Due to the problems of unclear boundary segmentation and insufficient Semantic information of small targets in high-resolution remote sensing images, an improved network TU net based on U-net++ is proposed. Secondly, the attention aggregation module of the base Transformer is introduced to capture global contextual information, replacing the original multi-level skip connections of U-net++. A cross-window interaction module is designed, which significantly reduces computational complexity and achieves a lightweight model. Finally, a dynamic feature fusion block is designed at the end of the decoder to obtain multi-class and multi-scale Semantic information and enhance the final segmentation effect. TU-net conducted experiments on two datasets, where OA, mIoU, and mF1 scores were higher than mainstream models. The IoU and F1 scores of small-sized target cars in the Vaihingen dataset were 0.896 and 0.962, respectively, which were 5% and 15.8% higher than the suboptimal model; The IoU and F1 scores of the trees in the Potsdam dataset are 0.913 and 0.936, respectively, which are 6.3% and 4.3% higher than the suboptimal model. The experimental results show that the model can more accurately segment small-sized targets and target boundaries.https://ieeexplore.ieee.org/document/10930761/High resolution remote sensing imagesemantic segmentationfeature head refinement modulelightweightingdynamic feature fusion block
spellingShingle Yang Xu
Bin Cao
Hui Lu
Improved U-Net++ Semantic Segmentation Method for Remote Sensing Images
IEEE Access
High resolution remote sensing image
semantic segmentation
feature head refinement module
lightweighting
dynamic feature fusion block
title Improved U-Net++ Semantic Segmentation Method for Remote Sensing Images
title_full Improved U-Net++ Semantic Segmentation Method for Remote Sensing Images
title_fullStr Improved U-Net++ Semantic Segmentation Method for Remote Sensing Images
title_full_unstemmed Improved U-Net++ Semantic Segmentation Method for Remote Sensing Images
title_short Improved U-Net++ Semantic Segmentation Method for Remote Sensing Images
title_sort improved u net semantic segmentation method for remote sensing images
topic High resolution remote sensing image
semantic segmentation
feature head refinement module
lightweighting
dynamic feature fusion block
url https://ieeexplore.ieee.org/document/10930761/
work_keys_str_mv AT yangxu improvedunetsemanticsegmentationmethodforremotesensingimages
AT bincao improvedunetsemanticsegmentationmethodforremotesensingimages
AT huilu improvedunetsemanticsegmentationmethodforremotesensingimages