RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating Mechanism

With the rapid development of remote sensing (RS) technology, it has become more and more convenient to obtain multi-temporal RS images, which provides new opportunities for the research and development of change detection (CD) technology. However, existing methods still have shortcomings in recogni...

Full description

Saved in:
Bibliographic Details
Main Authors: Chuanlu Li, Xiaorong Xue, Caijia Zeng, Yifan Xu, Xingbiao Xu, Siyue Zhao
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11014070/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849761327843639296
author Chuanlu Li
Xiaorong Xue
Caijia Zeng
Yifan Xu
Xingbiao Xu
Siyue Zhao
author_facet Chuanlu Li
Xiaorong Xue
Caijia Zeng
Yifan Xu
Xingbiao Xu
Siyue Zhao
author_sort Chuanlu Li
collection DOAJ
description With the rapid development of remote sensing (RS) technology, it has become more and more convenient to obtain multi-temporal RS images, which provides new opportunities for the research and development of change detection (CD) technology. However, existing methods still have shortcomings in recognizing complex targets and accurately distinguishing the edges of change regions when performing CD in complex backgrounds. Therefore, we propose a novel CD model called RegCDNet, which is specifically designed to address the needs of RS image CD. The model employs RegNet as the backbone network for feature extraction, using a simple and efficient strategy to fuse shallow features. We designed an ac-dramit feature enhancement module (AFEM) that combines atrous convolution and a transformer containing a dual attention mechanism, which can efficiently capture long-range dependencies between pixels while focusing on local information. A dual-gated fusion module (DGFM) is also designed, which employs a dual gating mechanism for feature fusion and can dynamically adjust the fusion weights. The experiment achieved 91.22%, 92.84% and 82.52% F1 metrics on the LEVIR-CD, WHU-CD and SYSU-CD datasets, respectively.
format Article
id doaj-art-6d4b764c24234869b2ce2644dbdce4df
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-6d4b764c24234869b2ce2644dbdce4df2025-08-20T03:06:04ZengIEEEIEEE Access2169-35362025-01-0113914669147910.1109/ACCESS.2025.357292811014070RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating MechanismChuanlu Li0https://orcid.org/0009-0001-5203-4275Xiaorong Xue1https://orcid.org/0000-0002-6443-4794Caijia Zeng2https://orcid.org/0009-0006-2435-8869Yifan Xu3https://orcid.org/0009-0002-1652-7555Xingbiao Xu4https://orcid.org/0009-0004-8408-0710Siyue Zhao5School of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou, ChinaSchool of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou, ChinaSchool of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou, ChinaSchool of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou, ChinaSchool of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou, ChinaSchool of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou, ChinaWith the rapid development of remote sensing (RS) technology, it has become more and more convenient to obtain multi-temporal RS images, which provides new opportunities for the research and development of change detection (CD) technology. However, existing methods still have shortcomings in recognizing complex targets and accurately distinguishing the edges of change regions when performing CD in complex backgrounds. Therefore, we propose a novel CD model called RegCDNet, which is specifically designed to address the needs of RS image CD. The model employs RegNet as the backbone network for feature extraction, using a simple and efficient strategy to fuse shallow features. We designed an ac-dramit feature enhancement module (AFEM) that combines atrous convolution and a transformer containing a dual attention mechanism, which can efficiently capture long-range dependencies between pixels while focusing on local information. A dual-gated fusion module (DGFM) is also designed, which employs a dual gating mechanism for feature fusion and can dynamically adjust the fusion weights. The experiment achieved 91.22%, 92.84% and 82.52% F1 metrics on the LEVIR-CD, WHU-CD and SYSU-CD datasets, respectively.https://ieeexplore.ieee.org/document/11014070/Change detectionmulti-temporal remote sensing imagesatrous convolutionattention mechanismdual gating mechanism
spellingShingle Chuanlu Li
Xiaorong Xue
Caijia Zeng
Yifan Xu
Xingbiao Xu
Siyue Zhao
RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating Mechanism
IEEE Access
Change detection
multi-temporal remote sensing images
atrous convolution
attention mechanism
dual gating mechanism
title RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating Mechanism
title_full RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating Mechanism
title_fullStr RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating Mechanism
title_full_unstemmed RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating Mechanism
title_short RegCDNet: A RegNet-Based Framework for Remote Sensing Image Change Detection Combining Feature Enhancement and Gating Mechanism
title_sort regcdnet a regnet based framework for remote sensing image change detection combining feature enhancement and gating mechanism
topic Change detection
multi-temporal remote sensing images
atrous convolution
attention mechanism
dual gating mechanism
url https://ieeexplore.ieee.org/document/11014070/
work_keys_str_mv AT chuanluli regcdnetaregnetbasedframeworkforremotesensingimagechangedetectioncombiningfeatureenhancementandgatingmechanism
AT xiaorongxue regcdnetaregnetbasedframeworkforremotesensingimagechangedetectioncombiningfeatureenhancementandgatingmechanism
AT caijiazeng regcdnetaregnetbasedframeworkforremotesensingimagechangedetectioncombiningfeatureenhancementandgatingmechanism
AT yifanxu regcdnetaregnetbasedframeworkforremotesensingimagechangedetectioncombiningfeatureenhancementandgatingmechanism
AT xingbiaoxu regcdnetaregnetbasedframeworkforremotesensingimagechangedetectioncombiningfeatureenhancementandgatingmechanism
AT siyuezhao regcdnetaregnetbasedframeworkforremotesensingimagechangedetectioncombiningfeatureenhancementandgatingmechanism