Awesome-Optical-Remote-Sensing-Datasets-and-Methods

This repository is for summarizing the latest optical remote sensing datasets and methods, which are listed in our review article Deep learning change detection techniques for optical remote sensing imagery: Status, Perspectives and Challenges published in International Journal of Applied Earth Observation and Geoinformation.

Introduction

Change detection (CD) aims to compare and analyze images of identical geographic areas but different dates, whereby revealing spatio-temporal change patterns of Earth’s surface. With the implementation of the High-Resolution Earth Observation Project, an integrated sky-to-ground observation system has been continuously developed and improved. The accumulation of massive multi-modal, multi-angle, and mul-ti-resolution remote sensing data have greatly enriched the CD data sources. Among them, high-resolution optical remote sensing images contain abundant spatial detail information, making it possible to interpret fine-grained scenes and greatly expand the application breadth and depth of CD. Generally, traditional optical remote sensing CD methods are cumbersome in steps and have a low level of automation. In con-trast, artificial intelligence (AI) based CD methods possess powerful feature extraction and non-linear modeling capabilities, thereby gaining advantages that traditional methods cannot match. As a result, they have become the mainstream approaches in the field of CD. This review article systematically summarizes the datasets, theories, and methods of CD for optical remote sensing image. It provides a comprehensive analysis of AI-based CD algorithms based on deep learning paradigms from the perspectives of algorithm granularity. In-depth analysis of the performance of typical algorithms are further conducted. Finally, we summarize the challenges and trends of the CD algorithms in the AI era, aiming to provide important guidelines and insights for relevant researchers.

Publicaiton trends of DLCD

BCD datasets of optical remote sensing images

Target of interest	Datasets	Year	Image Pairs	Image size	Resolution	Website
Building	TUE-CD[paper]	2024	1656	256×256	1.8m	link
	EGY-BCD[paper]	2023	6091	256×256	0.25m	link
	HRCUS-CD[paper]	2023	11388	256×256	0.5m	link
	SI-BU dataset[paper]	2023	4932	512×512	0.2m	link
	LEVIR-CC[paper]	2022	10077	1024×1024	0.5m	link
	SYSU-CD[paper]	2021	20000	256×256	0.5m	link
	CD_Data_GZ[paper]	2020	19	1006×1168-4936×5224	0.55m	link
	DSIFN Dataset[paper]	2020	3940	512×512	-	link
	LEVIR-CD[paper]	2020	637	1024×1024	0.5m	link
	LEVIR-CD+[paper]	2020	1970	1024×1024	0.5m	link
	WHU Building CD[paper]	2019	1	32507×15354	0.2m	link
	CDD Dataset[paper]	2018	16000	256×256	0.03-1m	link
	ABCD[paper]	2017	16950	128×128	0.4m	link
Land cover	TZ-CD[paper]	2023	1	30307×40620	1m	-
	BTCDD[paper]	2021	5281	256×256	-	-
	ZY3[paper]	2017	1	458×559	5.8m	link
	AICD[paper]	2011	1000	800×600	0.5m	-
Mine	MineNetCD[paper]	2024	71711	256×256	1.2m	link
Cropland	CLCD[paper]	2022	600	512×512	0.5-2m	link
Riverway	The River Data Set[paper]	2019	1	463×241	30m	link

BCD contest dataset for optical remote sensing images

Target of interest	Event	Year	Track	Image Pairs	Image size	Resolution	Website
Land cover	"Shengteng Cup" remote sensing image intelligent processing algorithm competition	2021	Remote sensing image change detection	4194	512×512	1-2m	link
	Artificial Intelligence Remote sensing Interpretation Competition	2020	change detection	4662	512×512	0.5-3m	link
	SN7: Multi-Temporal Urban Development Challenge	2020	Multi-Temporal Urban Development Challenge	12	1024×1024	4m	link
	Remote sensing image sparse characterization and intelligent Processing algorithm competition	2019	Remote sensing image change detection	103	960×960	-	link
Building	The 5th "Sino-Keke Star Map Cup" International high-resolution remote sensing image interpretation Competition	2021	Building survey and change detection in high resolution visible light images	2000	512×512	2m	link

BCD datasets of multi-modal remote sensing images

Target of interest	Multimodal types	Datasets	Year	Image Pairs	Image size	Resolution	Website
Flood	Opt-SAR	CAU-Flood dataset[paper]	2023	18302	256×256	10m	link
		dataset3_Wuhan[paper]	2022	1	11216×13693	3m	link
		Ombria dataset[paper]	2022	844	256×256	10m	link
Land for construction	Opt-Map	EVLab-CMCD[paper]	2024	5622	512×512	0.8m	link
Building	Opt-DSM	Hi-BCD[paper]	2023	1500	1000×1000	0.25m	link

SCD datasets of optical remote sensing images

Target of interest	Datasets	Year	Image Pairs	Image size	Resolution	Website
Land cover	CropSCD[paper]	2024	4141	512×512	0.5-2m	link
	Hi-CNA dataset[paper]	2024	6797	512×512	0.8m	link
	ChangNet[paper]	2023	31000	1900×1200	0.3m	link
	CNAM-CD[paper]	2023	2503	512×512	0.5m	link
	WUSU dataset[paper]	2023	3	6358×6382/7025×5500	1m	link
	DynamicEarthNet[paper]	2022	600	1024×1024	3m	link
	Landsat-SCD[paper]	2022	8468	416×416	30m	link
	S2MTCP[paper]	2021	1520	600×600	10m	link
	Hi-UCD[paper]	2020	1293	1024×1024	0.1m	link
	SECOND[paper]	2020	4662	512×512	-	link
	HRSCD[paper]	2019	291	10000×10000	0.5m	link
	Mts-WH[paper]	2019	1	7200×6000	1m	link
	HCCD[paper]	2018	3	390×200	30m	-
Building	BANDON[paper]	2023	2283	2048×2048	0.6m	link
	NanjingDataset[paper]	2022	2519	256×256	0.3m	link
	QFabric[paper]	2021	2520	120×120-12000×12000	0.31-0.7m	link
	S2Looking[paper]	2021	5000	1024×1024	0.5-0.8m	link
Landslide	GVLM[paper]	2023	17	1748×1748-10808×7424	0.59m	link

SCD contest dataset for optical remote sensing images

Target of interest	Event	Year	Track	Image Pairs	Image size	Resolution	Website
Land cover	2024 "Jilin-1" Cup Satel-lite Remote Sensing Application Youth Innovation and Entrepreneurship Competition	2024	High resolution remote sensing image total element change detection	5000	512×512	<0.75m	link
	"Jilin-1" Cup Satellite Remote Sensing Application Youth Innovation and Entrepreneurship Compe-tition	2023	Cultivated land change detection based on high-resolution satellite images	8000	256×256	<0.75m	link
	Guofeng East Eye Cup" remote sensing image intelligent processing algorithm competition	2023	Object level change detection	6000+	512×512	1-2m	link
	"Space Map Cup" remote sensing image intelligent processing algorithm competition	2022	Remote sensing image change detection
	National Artificial Intelligence Competition	2020	AI + Remote sensing track	-	256×256	-	link
Building	"Remote Sensing Image Intelligent Interpretation Technology Challenge	2021	Building change detection in remote sensing images	10000	512×512	-	link
	"Tianzhi Cup" artificial intelligence challenge	2021	Visible light image building intelligent change detection	5000	1024×1024	0.5-0.7m	link
	xView2 Challenge	2019	Building Damage Assessment	11034	1024×1024	<0.8m	link
Flood	SpaceNet8: Flood Detection Challenge	2022	Flood Detection Challenge Using Multiclass Segmentation	12	1300×1300	0.3-0.8m	link

CNN-based IB-DLCD methods

Category	Abbreviation	Title	Publication	Website
Early fusion	Res2-Unet[paper]	Res2-Unet, a new deep architecture for building detection from high spatial resolution images	JSTARS2022	link
	FC-EF-Res[paper]	Multitask learning for large-scale semantic change detection	CVIU2019	link
	UNet++_MSOF[paper]	End-to-end change detection for high resolution satellite images using improved UNet++	RS2019	link
	FC-EF[paper]	Fully convolutional siamese networks for change detection	ICIP2018	link
Middle fusion	DESNet[paper]	A Difference Enhanced Neural Network for Semantic Change Detection of Remote Sensing Images	GRSL2023	-
	EGPNet[paper]	Edge-Guided Parallel Network for VHR Remote Sensing Image Change Detection	JSTARS2023	link
	Bi-SRNet[paper]	Bi-temporal semantic reasoning for the semantic change detection in HR remote sensing images	TGRS2022	link
	P2V-CD[paper]	Transition is a process: Pair-to-video change detection networks for very high resolution remote sensing images	TIP2022	link
	SMD-Net[paper]	SMD-Net: Siamese multi-scale difference-enhancement network for change detection in remote sensing	RS2022	-
	BASNet[paper]	BASNet: A Boundary-Aware Siamese Network for Accurate Remote-Sensing Change Detection	GRSL2021	-
	DSNet[paper]	Deep Siamese Networks Based Change Detection with Remote Sensing Images	RS2021	-
	SNUNet-CD[paper]	SNUNet-CD: A densely connected Siamese network for change detection of VHR images	GRSL2021	link
	FC-Siam-conc[paper]	Fully convolutional siamese networks for change detection	ICIP2018	link
	FC-Siam-diff[paper]
Late fusion	RaSRNet[paper]	RaSRNet: An end-to-end Relation-aware Semantic Reasoning Network for Change Detection in Optical Remote Sensing Images	TIM2023	-
	SGSLN[paper]	Exchanging Dual-Encoder–Decoder: A New Strategy for Change Detection With Semantic Guidance and Spatial Localization	TGRS2023	link
	FCCDN[paper]	FCCDN: Feature constraint network for VHR image change detection	JPRS2022	link

Attention-based IB-DLCD methods

Abbreviation	Title	Publication	Website
A2Net[paper]	Lightweight Remote Sensing Change Detection With Progressive Feature Aggregation and Supervised Attention	TGRS2023	link
DMINet[paper]	Change detection on remote sensing images using dual-branch multilevel intertemporal network	TGRS2023	link
GAS-Net[paper]	Global-aware siamese network for change detection on remote sensing images	JPRS2023	link
USSFC-Net[paper]	Ultralightweight Spatial–Spectral Feature Cooperation Network for Change Detection in Remote Sensing Images	TGRS2023	link
SNAFF[paper]	High‐resolution optical remote sensing image change detection based on dense connection and attention feature fusion network	PHOR2023	_
DMSANet[paper]	Building Change Detection in Remote Sensing Images Based on Dual Multi-Scale Attention	RS2022	_
ISNet[paper]	ISNet: Towards Improving Separability for Remote Sensing Image Change Detection	TGRS2022	link
AGCDetNet[paper]	AGCDetNet: An attention-guided network for building change detection in high-resolution remote sensing images	JSTARS2021	_
ADS-Net[paper]	ADS-Net: An Attention-Based deeply supervised network for remote sensing image change detection	JAG2021	_
SSA-SiamNet[paper]	SSA-SiamNet: Spectral–spatial-wise attention-based Siamese network for hyperspectral image change detection	TGRS2021	_
MSPSNet[paper]	Deep Multiscale Siamese Network With Parallel Convolutional Structure and Self-Attention for Change Detection	TGRS2021	_
DTCDSCN[paper]	Building change detection for remote sensing images using a dual-task constrained deep siamese convolutional network model	GRSL2020
IFN[paper]	A deeply supervised image fusion network for change detection in high resolution bi-temporal remote sensing images	JPRS2020	link
STANet[paper]	A spatial-temporal attention-based method and a new dataset for remote sensing image change detection	RS2020	link

Transformer-based IB-DLCD methods

Abbreviation	Title	Publication	Website
BiFA[paper]	Bifa: Remote sensing image change detection with bitemporal feature alignment	TGRS2024	link
CDMamba[paper]	CDMamba: Remote Sensing Image Change Detection with Mamba	ArXiv2024	link
CDMask[paper]	Rethinking Remote Sensing Change Detection With A Mask View	ArXiv2024	link
Changemamba[paper]	Changemamba: Remote sensing change detection with spatio-temporal state space model	ArXiv2024	link
MaskCD[paper]	MaskCD: A Remote Sensing Change Detection Network Based on Mask Classification	ArXiv2024	link
M-CD[paper]	A Mamba-based Siamese Network for Remote Sensing Change Detection	ArXiv2024	link
SBA-PN[paper]	Siamese Bi-Attention Pooling Network for Change Detection in Remote Sensing	JSTARS2024	_
ScratchFormer[paper]	Remote sensing change detection with transformers trained from scratch	TGRS2024	link
MDAFormer[paper]	MDAFormer: Multi-level difference aggregation transformer for change detection of VHR optical imagery	JAG2023	_
TransY-Net[paper]	TransY-Net: Learning Fully Transformer Networks for Change Detection of Remote Sensing Images	TGRS2023	link
ChangeFormer[paper]	A transformer-based siamese network for change detection	IGARSS2022	link
FTN[paper]	Fully transformer network for change detection of remote sensing images	ACVV2022	link
Pyramid-SCDFormer[paper]	A transformer-based Siamese network and an open optical dataset for semantic change detection of remote sensing images	JDE2022	_
SST-Former[paper]	Spectral–spatial–temporal transformers for hyperspectral image change detection	TGRS2022	link
SwinSUNet[paper]	SwinSUNet: Pure transformer network for remote sensing image change detection	TGRS2022	_
TMFF[paper]	Transformer-based multi-scale feature fusion network for remote sensing change detection	JRS2022	_
TransUNetCD[paper]	TransUNetCD: A hybrid transformer network for change detection in optical remote-sensing images.	TGRS2022	_

CNN-Transformer hybrid-based IB-DLCD methods

Category	Abbreviation	Title	Publication	Website
Sequential integration	GCFormer[paper]	GCFormer: Global Context-aware Transformer for Remote Sensing Image Change Detection	TGRS2024	link
	MSCANet[paper]	A CNN-transformer network with multiscale context aggregation for fine-grained cropland change detection	JSTARS2022	link
	UVACD[paper]	A network combining a transformer and a convolutional neural network for remote sensing image change detection	RS2022	-
	BIT[paper]	Remote sensing image change detection with transformers	TGRS2021	link
	MTCNet[paper]	A CBAM based multiscale transformer fusion approach for remote sensing image change detection	TGRS2021	-
Multi-path CNN integration	SCanNet[paper]	Joint spatio-temporal modeling for semantic change detection in remote sensing images	TGRS2024	link
	DMATNet[paper]	Multi-path CNN integration**|**Remote sensing image change detection transformer network based on dual-feature mixed attention	TGRS2022	-
Parallel integration	WNet[paper]	Wnet: W-shaped hierarchical network for remote sensing image change detection	TGRS2023	link
	ACABFNet[paper]	Axial cross attention meets CNN: Bibranch fusion network for change detection	JSTARS2022	link
	ICIF-Net[paper]	ICIF-Net: Intra-scale cross-interaction and inter-scale feature fusion network for bitemporal remote sensing images change detection	TGRS2022	link
Unilateral mixing	ConvTransNet[paper]	ConvTransNet: A CNN-Transformer Network for Change Detection with Multi-Scale Global-Local Representations	TGRS2023	-
	CTD-Former[paper]	Relation Changes Matter: Cross-Temporal Difference Transformer for Change Detection in Remote Sensing Images	TGRS2023	link
	DAHT-Net[paper]	DAHT-Net: Deformable Attention-Guided Hierarchical Transformer Network Based on Remote Sensing Image Change Detection	ACCESS2023	-
	GateFormer[paper]	GateFormer: Gate Attention UNet With Transformer for Change Detection of Remote Sensing Images	JSTARS2023	-
	GeoFormer[paper]	GeoFormer: A Geometric Representation Transformer for Change Detection	TGRS2023	link
	SMNet[paper]	SMNet: symmetric multi-task network for semantic change detection in remote sensing images based on CNN and transformer	RS2023	-
	SUT[paper]	A Full-Scale Connected CNN–Transformer Network for Remote Sensing Image Change Detection	RS2023	-
Bilateral mixing	ACAHNet[paper]	Asymmetric cross-attention hierarchical network based on CNN and transformer for bitemporal remote sensing images change detection	TGRS2023	-
	DMMSTNet[paper]	Remote Sensing Image Change Detection Based on Deep Multi-Scale Multi-Attention Siamese Transformer Network	RS2023	-
	TChange[paper]	TChange: A Hybrid Transformer-CNN Change Detection Network	RS2023	-
	H-TransCD[paper]	Hybrid-transcd: A hybrid transformer remote sensing image change detection network via token aggregation	JPRS2022	-

Label-efficient learning-based IB-DLCD methods

Method	Category	Abbreviation	Title	Publication	Website
Semi-supervised	Adversarial learning	SemiCDNet[paper]	SemiCDNet: A semisupervised convolutional neural network for change detection in high resolution remote-sensing images	TGRS2020	-
		GDCN[paper]	Adversarial learning**|**A generative discriminatory classified network for change detection in multispectral imagery	JSTARS2019	-
	Consistency constraints	C2F-SemiCD[paper]	C2F-SemiCD: A Coarse-to-Fine Semi-Supervised Change Detection Method Based on Consistency Regularization in High-Resolution Remote-Sensing Images	TGRS2024	link
		FPA[paper]	Semisupervised Change Detection With Feature-Prediction Alignment	TGRS2023	link
		Semi-LCD[paper]	Consistency-guided lightweight network for semi-supervised binary change detection of buildings in remote sensing images	GRS2023	-
		SaDL[paper]	Semantic-aware dense representation learning for remote sensing image change detection	TGRS2022	link
		SemiCD[paper]	Revisiting consistency regularization for semi-supervised change detection in remote sensing images	ArXiv2022	link
	Pseudo-label learning	DCSS[paper]	Dynamically updated semi-supervised change detection network combining cross-supervision and screening algorithms	TGRS2024	-
		ECPS[paper]	ECPS: Cross Pseudo Supervision Based on Ensemble Learning for Semi-Supervised Remote Sensing Change Detection	TGRS2024	link
		STCRNet[paper]	STCRNet: A Semi-Supervised Network Based on Self-Training and Consistency Regularization for Change Detection in VHR Remote Sensing Images	JSTARS2023	link
		ST-RCL[paper]	Joint Self-training and Rebalanced Consistency Learning for Semi-supervised Change Detection	TGRS2023	-
		SemiBuildingChange[paper]	SemiBuildingChange: A Semi-supervised High-Resolution Remote Sensing Image Building Change Detection Method With a Pseudo Bi-Temporal Data Generator	TGRS2023	-
		SSCD[paper]	Pseudo-label learning**|**A New Semi-Supervised Method for Detecting Semantic Changes in Remote Sensing Images	GRSL2023	-
		RCL[paper]	Reliable contrastive learning for semi-supervised change detection in remote sensing images	TGRS2022	link
		SemiSANet[paper]	SemiSANet: A semi-supervised high-resolution remote sensing image change detection model using Siamese networks with graph attention	RS2022	-
		IAug_CDNet[paper]	Adversarial instance augmentation for building change detection in remote sensing images	TGRS2021	link
Active learning		DALCD[paper]	Deep active learning in remote sensing for data efficient change detection	ArXiv2020	-
		CFSS-CD[paper]	A coarse-to-fine semi-supervised change detection for multispectral images	TGRS2018	-
Weakly-supervised	Point-level labels	CARGNet[paper]	Point Label Meets Remote Sensing Change Detection: A Consistency-Aligned Regional Growth Network	TGRS2023	link
	Patch-level labels	MS-Former[paper]	MS-Former: Memory-Supported Transformer for Weakly Supervised Change Detection with Patch-Level Annotations	ArXiv2023	link
		SDCDNetpaper]	SDCDNet: A Semi-Dual Change Detection Network Framework with Super-Weak Lable for Remote Sensing Image	TGRS2023	-
	Image-level labels	WSLCD[paper]	Beyond Pixel-Level Annotation: Exploring Self-Supervised Learning for Change Detection With Image-Level Supervision	TGRS2024	link
		BGMix[paper]	Background-mixed augmentation for weakly supervised change detection	AAAI2023	link
		FCD-GAN[paper]	Fully convolutional change detection framework with generative adversarial network for unsupervised, weakly supervised and regional supervised change detection	TPAMI2023	link
		TransWCD-DL[paper]	Exploring Effective Priors and Efficient Models for Weakly-Supervised Change Detection	ArXiv2023	link
		WSCD[paper]	A Siamese Network Combining Multi-Scale Joint Supervision and Improved Consistency Regularization for Weakly Supervised Building Change Detection	JSTARS2023	-
	Low-resolution labels	RFWSCD[paper]	Weakly supervised semantic change detection via label refinement framework	IGARSS2021	-

Citation

Please cite our paper if you find it is useful for your research.

@article{PENG2025104282,
  title={Deep learning change detection techniques for optical remote sensing imagery: Status, perspectives and challenges},
  author={Peng, Daifeng, Liu, Xuelian, Zhang, Yongjun,  Guan, Haiyan,  Li, Yansheng and Bruzzone, Lorenzo},
  journal={International Journal of Applied Earth Observation and Geoinformation},
  volume={136},
  pages={104282},
  year={2025},
  publisher={Elsevier},
  issn = {1569-8432},
  doi = {https://doi.org/10.1016/j.jag.2024.104282},
  url = {https://www.sciencedirect.com/science/article/pii/S1569843224006381}
}

Reference

• wenhwu/awesome-remote-sensing-change-detection