Semi-Supervised Remote Sensing Building Change Detection with Joint Perturbation and Feature Complementation

IF 4.2 2区 地球科学 Q2 ENVIRONMENTAL SCIENCES
Remote Sensing Pub Date : 2024-09-14 DOI:10.3390/rs16183424
Zhanlong Chen, Rui Wang, Yongyang Xu
{"title":"Semi-Supervised Remote Sensing Building Change Detection with Joint Perturbation and Feature Complementation","authors":"Zhanlong Chen, Rui Wang, Yongyang Xu","doi":"10.3390/rs16183424","DOIUrl":null,"url":null,"abstract":"The timely updating of the spatial distribution of buildings is essential to understanding a city’s development. Deep learning methods have remarkable benefits in quickly and accurately recognizing these changes. Current semi-supervised change detection (SSCD) methods have effectively reduced the reliance on labeled data. However, these methods primarily focus on utilizing unlabeled data through various training strategies, neglecting the impact of pseudo-changes and learning bias in models. When dealing with limited labeled data, abundant low-quality pseudo-labels generated by poorly performing models can hinder effective performance improvement, leading to the incomplete recognition results of changes to buildings. To address this issue, we propose a feature multi-scale information interaction and complementation semi-supervised method based on consistency regularization (MSFG-SemiCD), which includes a multi-scale feature fusion-guided change detection network (MSFGNet) and a semi-supervised update method. Among them, the network facilitates the generation of multi-scale change features, integrates features, and captures multi-scale change targets through the temporal difference guidance module, the full-scale feature fusion module, and the depth feature guidance fusion module. Moreover, this enables the fusion and complementation of information between features, resulting in more complete change features. The semi-supervised update method employs a weak-to-strong consistency framework to achieve model parameter updates while maintaining perturbation invariance of unlabeled data at both input and encoder output features. Experimental results on the WHU-CD and LEVIR-CD datasets confirm the efficacy of the proposed method. There is a notable improvement in performance at both the 1% and 5% levels. The IOU in the WHU-CD dataset increased by 5.72% and 6.84%, respectively, while in the LEVIR-CD dataset, it improved by 18.44% and 5.52%, respectively.","PeriodicalId":48993,"journal":{"name":"Remote Sensing","volume":"65 1","pages":""},"PeriodicalIF":4.2000,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Remote Sensing","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.3390/rs16183424","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0

Abstract

The timely updating of the spatial distribution of buildings is essential to understanding a city’s development. Deep learning methods have remarkable benefits in quickly and accurately recognizing these changes. Current semi-supervised change detection (SSCD) methods have effectively reduced the reliance on labeled data. However, these methods primarily focus on utilizing unlabeled data through various training strategies, neglecting the impact of pseudo-changes and learning bias in models. When dealing with limited labeled data, abundant low-quality pseudo-labels generated by poorly performing models can hinder effective performance improvement, leading to the incomplete recognition results of changes to buildings. To address this issue, we propose a feature multi-scale information interaction and complementation semi-supervised method based on consistency regularization (MSFG-SemiCD), which includes a multi-scale feature fusion-guided change detection network (MSFGNet) and a semi-supervised update method. Among them, the network facilitates the generation of multi-scale change features, integrates features, and captures multi-scale change targets through the temporal difference guidance module, the full-scale feature fusion module, and the depth feature guidance fusion module. Moreover, this enables the fusion and complementation of information between features, resulting in more complete change features. The semi-supervised update method employs a weak-to-strong consistency framework to achieve model parameter updates while maintaining perturbation invariance of unlabeled data at both input and encoder output features. Experimental results on the WHU-CD and LEVIR-CD datasets confirm the efficacy of the proposed method. There is a notable improvement in performance at both the 1% and 5% levels. The IOU in the WHU-CD dataset increased by 5.72% and 6.84%, respectively, while in the LEVIR-CD dataset, it improved by 18.44% and 5.52%, respectively.
利用联合扰动和特征补充进行半监督式遥感建筑物变化检测
及时更新建筑物的空间分布对于了解城市的发展至关重要。深度学习方法在快速准确地识别这些变化方面有着显著的优势。目前的半监督变化检测(SSCD)方法有效地减少了对标记数据的依赖。然而,这些方法主要侧重于通过各种训练策略利用非标记数据,而忽略了模型中伪变化和学习偏差的影响。在处理有限的标注数据时,性能不佳的模型所产生的大量低质量伪标签会阻碍性能的有效提高,导致对建筑物变化的识别结果不完整。针对这一问题,我们提出了一种基于一致性正则化的特征多尺度信息交互与互补半监督方法(MSFG-SemiCD),它包括一个多尺度特征融合引导的变化检测网络(MSFGNet)和一种半监督更新方法。其中,该网络通过时差引导模块、全尺度特征融合模块和深度特征引导融合模块,促进多尺度变化特征的生成、特征的整合和多尺度变化目标的捕捉。此外,这还能实现特征之间的信息融合与互补,从而获得更完整的变化特征。半监督更新方法采用弱到强一致性框架实现模型参数的更新,同时保持输入和编码器输出特性的未标记数据的扰动不变性。在 WHU-CD 和 LEVIR-CD 数据集上的实验结果证实了所提方法的有效性。在 1% 和 5% 的水平上,性能都有显著提高。WHU-CD 数据集的 IOU 分别提高了 5.72% 和 6.84%,而 LEVIR-CD 数据集的 IOU 则分别提高了 18.44% 和 5.52%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Remote Sensing
Remote Sensing REMOTE SENSING-
CiteScore
8.30
自引率
24.00%
发文量
5435
审稿时长
20.66 days
期刊介绍: Remote Sensing (ISSN 2072-4292) publishes regular research papers, reviews, letters and communications covering all aspects of the remote sensing process, from instrument design and signal processing to the retrieval of geophysical parameters and their application in geosciences. Our aim is to encourage scientists to publish experimental, theoretical and computational results in as much detail as possible so that results can be easily reproduced. There is no restriction on the length of the papers. The full experimental details must be provided so that the results can be reproduced.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信