Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity Network

In recent years, weakly supervised semantic segmentation (WSSS) has garnered significant attention in remote sensing image analysis due to its low annotation cost. To address the issues of inaccurate and incomplete seed areas and unreliable pseudo masks in WSSS, we propose a novel WSSS method for re...

Full description

Saved in:
Bibliographic Details
Main Authors: Zheng Chen, Yuheng Lian, Jing Bai, Jingsen Zhang, Zhu Xiao, Biao Hou
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/5/808
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850053223594852352
author Zheng Chen
Yuheng Lian
Jing Bai
Jingsen Zhang
Zhu Xiao
Biao Hou
author_facet Zheng Chen
Yuheng Lian
Jing Bai
Jingsen Zhang
Zhu Xiao
Biao Hou
author_sort Zheng Chen
collection DOAJ
description In recent years, weakly supervised semantic segmentation (WSSS) has garnered significant attention in remote sensing image analysis due to its low annotation cost. To address the issues of inaccurate and incomplete seed areas and unreliable pseudo masks in WSSS, we propose a novel WSSS method for remote sensing images based on the Siamese Affinity Network (SAN) and the Segment Anything Model (SAM). First, we design a seed enhancement module for semantic affinity, which strengthens contextual relevance in the feature map by enforcing a unified constraint principle of cross-pixel similarity, thereby capturing semantically similar regions within the image. Second, leveraging the prior notion of cross-view consistency, we employ a Siamese network to regularize the consistency of CAMs from different affine-transformed images, providing additional supervision for weakly supervised learning. Finally, we utilize the SAM segmentation model to generate semantic superpixels, expanding the original CAM seeds to more completely and accurately extract target edges, thereby improving the quality of segmentation pseudo masks. Experimental results on the large-scale remote sensing datasets DRLSD and ISPRS Vaihingen demonstrate that our method achieves segmentation performance close to that of fully supervised semantic segmentation (FSSS) methods on both datasets. Ablation studies further verify the positive optimization effect of each module on segmentation pseudo labels. Our approach exhibits superior localization accuracy and precise visualization effects across different backbone networks, achieving state-of-the-art localization performance.
format Article
id doaj-art-a8f0444b614f4bdc97e105dfe32a6abb
institution DOAJ
issn 2072-4292
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-a8f0444b614f4bdc97e105dfe32a6abb2025-08-20T02:52:35ZengMDPI AGRemote Sensing2072-42922025-02-0117580810.3390/rs17050808Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity NetworkZheng Chen0Yuheng Lian1Jing Bai2Jingsen Zhang3Zhu Xiao4Biao Hou5The Key Laboratory of Intelligent Perception and Image Understanding of the Ministry of Education, School of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaThe Key Laboratory of Intelligent Perception and Image Understanding of the Ministry of Education, School of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaThe Key Laboratory of Intelligent Perception and Image Understanding of the Ministry of Education, School of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaThe Key Laboratory of Intelligent Perception and Image Understanding of the Ministry of Education, School of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaThe College of Computer Science and Electronic Engineering, Hunan University, Changsha 410082, ChinaThe Key Laboratory of Intelligent Perception and Image Understanding of the Ministry of Education, School of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaIn recent years, weakly supervised semantic segmentation (WSSS) has garnered significant attention in remote sensing image analysis due to its low annotation cost. To address the issues of inaccurate and incomplete seed areas and unreliable pseudo masks in WSSS, we propose a novel WSSS method for remote sensing images based on the Siamese Affinity Network (SAN) and the Segment Anything Model (SAM). First, we design a seed enhancement module for semantic affinity, which strengthens contextual relevance in the feature map by enforcing a unified constraint principle of cross-pixel similarity, thereby capturing semantically similar regions within the image. Second, leveraging the prior notion of cross-view consistency, we employ a Siamese network to regularize the consistency of CAMs from different affine-transformed images, providing additional supervision for weakly supervised learning. Finally, we utilize the SAM segmentation model to generate semantic superpixels, expanding the original CAM seeds to more completely and accurately extract target edges, thereby improving the quality of segmentation pseudo masks. Experimental results on the large-scale remote sensing datasets DRLSD and ISPRS Vaihingen demonstrate that our method achieves segmentation performance close to that of fully supervised semantic segmentation (FSSS) methods on both datasets. Ablation studies further verify the positive optimization effect of each module on segmentation pseudo labels. Our approach exhibits superior localization accuracy and precise visualization effects across different backbone networks, achieving state-of-the-art localization performance.https://www.mdpi.com/2072-4292/17/5/808Siamese networkSAMremote sensing imagesweakly supervised semantic segmentation (WSSS)
spellingShingle Zheng Chen
Yuheng Lian
Jing Bai
Jingsen Zhang
Zhu Xiao
Biao Hou
Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity Network
Remote Sensing
Siamese network
SAM
remote sensing images
weakly supervised semantic segmentation (WSSS)
title Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity Network
title_full Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity Network
title_fullStr Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity Network
title_full_unstemmed Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity Network
title_short Weakly Supervised Semantic Segmentation of Remote Sensing Images Using Siamese Affinity Network
title_sort weakly supervised semantic segmentation of remote sensing images using siamese affinity network
topic Siamese network
SAM
remote sensing images
weakly supervised semantic segmentation (WSSS)
url https://www.mdpi.com/2072-4292/17/5/808
work_keys_str_mv AT zhengchen weaklysupervisedsemanticsegmentationofremotesensingimagesusingsiameseaffinitynetwork
AT yuhenglian weaklysupervisedsemanticsegmentationofremotesensingimagesusingsiameseaffinitynetwork
AT jingbai weaklysupervisedsemanticsegmentationofremotesensingimagesusingsiameseaffinitynetwork
AT jingsenzhang weaklysupervisedsemanticsegmentationofremotesensingimagesusingsiameseaffinitynetwork
AT zhuxiao weaklysupervisedsemanticsegmentationofremotesensingimagesusingsiameseaffinitynetwork
AT biaohou weaklysupervisedsemanticsegmentationofremotesensingimagesusingsiameseaffinitynetwork