A comprehensive and up-to-date compilation of datasets, tools, methods (including foundation models, diffusion models, transformers, and CNNs), review papers, and competitions for remote sensing change detection.
- Datasets
- Tools
- Methods
- Review Papers
- Competitions
- Satellite Data Resources for Disaster Response
- More Resources
- Citation
- SCD: Semantic Change Detection, BCD: Binary Change Detection, DDA: Disaster Damage Assessment, BDA: Building Damage Assessment
| Year | Task | Target | Dataset | Publication | Source | Image Pairs | Image Size | Resolution | Location | Class |
|---|---|---|---|---|---|---|---|---|---|---|
| 2025 | SCD | Land cover | LevirSCD | arXiv2025 | GF-1, Google Earth | 3,225 | 256×256 | 1-2 | Beijing, China | 16 |
| 2025 | BCD | Land cover | JL1-CD | arXiv2025 | Jilin-1 | 5,000 | 512×512 | 0.5-0.75m | Multiple provinces in China | 2 |
| 2025 | SCD | Building | EBD | JRS2025 | WorldView-3 | >18,000 | 512×512 | 0.3-0.5m | Global | 7 |
| 2024 | BCD | Mine | MineNetCD | TGRS2024 | Google Earth | 71,711 | 256×256 | 1.2m | Global | 2 |
| 2024 | BCD | Building | TUE-CD | TGRS2024 | WorldView-2 | 1,656 | 256×256 | 1.8m | Turkey | 2 |
| 2024 | SCD | Cropland | CropSCD | TGRS2024 | - | 4,141 | 512×512 | 0.5-2m | Guangdong, China | 9 |
| 2024 | SCD | Cropland | Hi-CNA | ISPRS P&RS 2024 | GF-2 | 6,797 | 512×512 | 0.8m | China (Hebei, Shanxi, Shandong, and Hubei) | 5 |
| 2024 | SCD | Land cover | ChangNet | ICASSP2024 | WayBack | 31,000 | 1,900×1,200 | 0.3m | 100 Cities in China | 6 |
| 2023 | SCD | Cropland | JL1-Cropland-CD | - | Jilin-1 | 8,000 | 256×256 | <0.75m | - | 9 |
| 2023 | BCD | Building | EGY-BCD | GRSL2023 | Google Earth | 6,091 | 256×256 | 0.25m | Egypt | 2 |
| 2023 | BCD | Building | HRCUS-CD | TGRS2023 | - | 11,388 | 256×256 | 0.5m | Zhuhai, China | 2 |
| 2023 | BCD | Building | SI-BU | ISPRS P&RS 2023 | Google Earth | 4,932 | 512×512 | 0.2m | Guiyang, China | 2 |
| 2023 | SCD | Land cover | CNAM-CD | RS2023 | Google Earth | 2,503 | 512×512 | 0.5m | 12 State-level New Areas in China | 6 |
| 2023 | SCD | Land cover | WUSU | IJDE2023 | GF-2 | 3 | 6,358×6,382 / 7,025×5,500 | 1m | Wuhan, China | 12 |
| 2023 | BCD | Landslide | GVLM | ISPRS P&RS 2023 | Google Earth | 17 | 1,748×1,748-10,808×7,424 | 0.59m | Global | 2 |
| 2023 | SCD | Building | BANDON | Sci. China Inf. Sci. 2023 | Google Earth, Microsoft Virtual Earth, and ArcGIS | 2,283 | 2,048×2,048 | 0.6m | China (Beijing, Shanghai, Wuhan, Shenzhen, Hong Kong, and Jinan) | 6 |
| 2023 | SCD | Land cover | DynamicEarthNet | CVPR2022 | PlanetFusion | 54,750 | 1,024×1,024 | 3m | Global | 7 |
| 2022 | BCD | Cropland | CLCD | JSTARS2022 | GF-2 | 600 | 512×512 | 0.5-2m | Guangdong, China | 2 |
| 2022 | RSICC | Building | LEVIR-CC | TGRS2022 | Google Earth | 10,077 | 1,024×1,024 | 0.5m | Texas, USA | 2 |
| 2022 | BCD | Land cover | SYSU-CD | TGRS2021 | - | 20,000 | 256×256 | 0.5m | Hong Kong, China | 2 |
| 2022 | SCD | Building | S2Looking | RS2021 | GF, SuperView, BJ-2 | 5,000 | 1,024×1,024 | 0.5-0.8m | Global | 2 |
| 2022 | BCD | Building | LEVIR-CD+ | RS2021 | Google Earth | 985 | 1,024×1,024 | 0.5m | Texas, USA | 2 |
| 2022 | SCD | Land cover | Landsat-SCD | IJDE2022 | Landsat | 8,468 | 416×416 | 30m | Xinjiang, China | 10 |
| 2022 | SCD | Building | NanjingDataset | ISPRS P&RS 2022 | Google Earth | 2,519 | 256×256 | 0.3m | Nanjing, China | 3 |
| 2022 | RSICC | Urban | Dubai-CC | TGRS2022 | Landsat 7 | 500 | 50×50 | 30m | Dubai | 6 |
| 2022 | SCD | Flood | SpaceNet 8 | CVPR2022W | Maxar | 12 | 1,300×1,300 | 0.3-0.8m | Germany, and Louisiana | 4 |
| 2021 | SCD | Land cover | MSD | JSTARS2022 | NAIP, Landsat-8, and NLCD | 2,250 | - | 1m, 30m | Maryland, USA | 16 |
| 2021 | SCD | Land cover | S2MTCP | ICPR2021 | Sentinel-2 | 1,520 | 600×600 | 10m | Global | - |
| 2021 | BCD | Urban | HTCD | RS2021 | Google Earth, Open Aerial Map | 3,772 | 256×256, 2,048×2,048 | 0.5971m, 0.07465m | Chisinau, Moldova | 2 |
| 2020 | BCD | Building | GZ-CD (or CD_Data_GZ) | TGRS2020 | Google Earth | 19 | 1,006×1,168-4,936×5,224 | 0.55m | Guangzhou, China | 2 |
| 2020 | BCD | Building | DSIFN (or DSIFN-CD) | ISPRS P&RS 2020 | Google Earth | 3,940 | 512×512 | - | China (Beijing, Chengdu, Shenzhen, Chongqing, Wuhan, and Xian) | 2 |
| 2020 | BCD | Building | LEVIR-CD | RS2020 | Google Earth | 637 | 1,024×1,024 | 0.5m | Texas, USA | 2 |
| 2020 | SCD | Land cover | Hi-UCD | arXiv2020 | Aerial Images | 1,293 | 1,024×1,024 | 0.1m | Tallinn, Estonia | 9 |
| 2020 | SCD | Land cover | SECOND | TGRS2021 | Aerial Images | 4,662 | 512×512 | - | China (Hangzhou, Chengdu, and Shanghai) | 6 |
| 2020 | BCD | Building | MUDS (or SpaceNet 7) | CVPR2021 | - | - | 1,024×1,024 | 4m | Global | 2 |
| 2019 | BDA | Building | xBD | arXiv2019 | Maxar | 11,034 | 1,024×1,024 | <0.8m | Global | 4 |
| 2019 | SCD | Land cover | HRSCD | CVIU2019 | IGN | 291 | 10,000×10,000 | 0.5m | France (Rennes, and Caen) | 5 |
| 2018 | BCD | Building | WHU-CD | TGRS2018 | Aerial Image | 1 | 32,507×15,354 | 0.2m | Christchurch, New Zealand | 2 |
| 2018 | BCD | Building | CDD (or SVCD) | Int. Arch. Photogramm. Remote Sens. Spatial Inf. 2018 | Google Earth | 1,6000 | 256×256 | 0.03-1m | - | 2 |
| 2018 | BCD | Riverway | The River Data Set | TGRS2018 | EO-1 Hyperion | 1 | 463×241 | 30m | Jiangsu, China | 2 |
| 2018 | BCD | Land cover | OSCD | IGARSS2018 | Sentinel-2 | 24 | 600×600 | 10-60m | Global | 2 |
| 2008 | BCD | Land cover | SZTAKI | TGRS2009 | Aerial Images | 13 | 952x640 | 1.5m | - |
| Year | Task | Target | Dataset | Publication | Source | Image Pairs | Image Size | Resolution | Location | Class |
|---|---|---|---|---|---|---|---|---|---|---|
| 2025 | DDA | Disaster | DisasterM3 | NeurIPS2025 | Optical-SAR-Instruction | - | - | - | Global | - |
| 2025 | SCD | Building | BRIGHT | arXiv2025 | Optical and SAR | 4,538 | 1,024×1,024 | 0.3-1m | Global | 4 |
| 2024 | SCD | Building | Hi-BCD | Information Fusion 2023 | Aerial Images, DSMs | 1,500 | 1,000×1,000 | 0.25m | Netherlands (Amsterdam, Rotterdam, and Utrecht) | 3 |
| 2024 | SCD | Flood | UrbanSARFloods | CVPR2024W | Sentinel-1 | 8,879 | 512×512 | 20m | Global | 5 |
| 2024 | SCD | Land use | EVLab-CMCD | ISPSR P&RS 2024 | GF-2, BJ-2, Historical land use maps | 5,622 | 512×512 | 0.8m | 10 Cities in China | 13 |
| 2023 | BCD | Flood | CAU-Flood | JAG2023 | Sentinel-1, Sentinel-2 | 18,302 | 256×256 | 10m | Global | 2 |
| 2023 | SCD | Flood | Kuro Siwo | NeurIPS2024 | Sentinel-1, DEM | 67,490 | 224×224 | 10m | Global | 3 |
| 2023 | SCD | Urban | SMARS | ISPRS P&RS 2023 | Simulated Orthoimages and DSMs | - | 512×512 | 0.3m, 0.5m | Simulated Paris and Venice | 3 |
| 2023 | BCD | Urban | 3DCD | ISPRS P&RS 2023 | Aerial Images, DSMs | 472 | 400×400, 200×200 | 0.5m, 1m | Valladolid, Spain | 2 |
| 2023 | SCD | Urban | Urb3DCD–V2 | ISPRS P&RS 2023 | ALS, Multi-Sensor | - | - | - | Simulated | 7 |
| 2022 | BCD | Flood | Wuhan | JAG2022 | Sentinel-2, COSMO-SkyMed | 1 | 11,216×13,693 | 3m | Wuhan, China | 2 |
| 2022 | BCD | Flood | Ombria | JSTARS2022 | Sentinel-1, Sentinel-2 | 1,688 | 256×256 | 10m | Global | 2 |
| 2021 | BCD | Land cover | MultiModalOSCD | ISPRS. XXIV ISPRS Congress 2021 | Sentinel-1, Sentinel-2 | 24 | 600×600 | 10-60m | Global | 2 |
| Year | Abbreviation | Description | Other |
|---|---|---|---|
| 2024 | rschange | An open-source toolbox dedicated to reproducing and developing advanced methods (e.g., DDLNet, CDMask) for change detection in remote sensing images. |   |
| 2024 | torchange | A benchmark library providing out-of-box, straightforward implementations of contemporary spatiotemporal change detection models (e.g., ChangeStar, Changen, AnyChange), metrics, and datasets to promote reproducibility in remote sensing research. |   |
| 2022 | Open-CD | The most comprehensive open-source toolbox for change detection, offering a unified platform with diverse methods, training/inference tools, data analysis scripts, and benchmarks to support research and development in the field. Paper: arXiv2024. |   |
| 2022 | PaddleRS | A remote sensing toolkit based on PaddlePaddle that supports change detection among other tasks, providing dedicated models (e.g., BIT, FarSeg), large-image processing capabilities, and practical tutorials for analyzing temporal land cover differences. The PyTorch version is called CDLab. |   |
| 2020 | Change Detection Repository | It provides Python implementations of selected traditional change detection methods (e.g., CVA, SFA, MAD) and deep learning-based approaches (e.g., SiamCRNN, DSFA, and FCN-based methods). |   |
| 2019 | ChangeDetectionToolbox | This MATLAB toolbox provides a modular, end-to-end framework for remote sensing change detection, implementing key methods such as CVA , MAD , and IRMAD to generate difference images and evaluate change maps. |   |
| Year | Abbreviation | Title | Publication | Keywords | Experiment Datasets | Other |
|---|---|---|---|---|---|---|
| 2025 | SA-CDNet | Detect Changes Like Humans: Incorporating Semantic Priors for Improved Change Detection | TGRS2025 | dual-stream decoder, multiscale feature, visual foundation model | AIRS, INRIA-Building, and WHU-Building; DLCCC, and LoveDA; WHU-CD, LEVIR-CD, LEVIR-CD+, S2Looking, WHU Cultivate Land Dataset |   |
| 2025 | DynamicEarth | DynamicEarth: How Far are We from Open-Vocabulary Change Detection? | arXiv2025 | Open-Vocabulary Change Detection, SAM, DINO | WHU-CD, LEVIR-CD, SECOND, S2Looking, and BANDON |   |
| 2025 | Change3D | Change3D: Revisiting Change Detection and Captioning from A Video Modeling Perspective | CVPR2025 | Perception Feature Extraction, Change Decoder, Caption Decoder | WHU-CD, HRSCD, xBD, LEVIR-CD, CLCD, SECOND, LEVIR-CC, and DUBAI-CC |   |
| 2025 | DisasterM3 | DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response | NeurIPS2025 | Multi-hazard, Multi-sensor, Multi-task | DisasterM3 |   |
| 2024 | AnyChange | Segment Any Change | NeurIPS2024 | Zero-shot change detection, SAM, bitemporal latent matching | xBD, LEVIR-CD, S2Looking, SECOND | |
| 2024 | SemiCD-VL | SemiCD-VL: Visual-Language Model Guidance Makes Better Semi-Supervised Change Detector | TGRS2024 | Visual-language model, semi-supervised learning, foundation model | WHU-CD, LEVIR-CD |   |
| 2024 | ChangeCLIP | ChangeCLIP: Remote sensing change detection with multimodal vision-language representation learning | ISPRS P&RS 2024 | Multimodal, vision-Language Representation Learning | WHU-CD, CDD, LEVIR-CD, LEVIR-CD+, and SYSU-CD |   |
| 2023 | BAN | A New Learning Paradigm for Foundation Model-Based Remote-Sensing Change Detection | TGRS2024 | Foundation Model, visual tuning | WHU-CD, LEVIR-CD, S2Looking, Landsat-SCD, and BANDON |   |
| 2023 | SAM-CD | Adapting Segment Anything Model for Change Detection in VHR Remote Sensing Images | TGRS2024 | SAM, vision foundation models | WHU-CD, LEVIR-CD, CLCD, S2Looking |   |
| Year | Abbreviation | Title | Publication | Keywords | Experiment Datasets | Other |
|---|---|---|---|---|---|---|
| 2025 | NeDS | Neural disaster simulation for transferable building damage assessment | RSE2025 | Synthetic data fine-tuning, deep generative models, conditional latent diffusion model | xBD, Los Angeles Wildfire (2025), and Nigeria Flooding (2025) | |
| 2024 | Changen2 | Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Model | TPAMI2024 | Synthetic data pre-training, generative model, foundation model | WHU-CD, xBD, LEVIR-CD, S2Looking, SECOND | |
| 2023 | Changen | Scalable Multi-Temporal Remote Sensing Change Data Generation via Simulating Stochastic Change Process | ICCV2023 | Deep generative model, change event simulation, semantic change synthesis | WHU-CD, LEVIR-CD, S2Looking |   |
| 2022 | DDPM-CD | DDPM-CD: Denoising Diffusion Probabilistic Models as Feature Extractors for Remote Sensing Change Detection | WACV2025 | Image synthesis, Denoising Diffusion Probabilistic Models | WHU-CD, CDD, DSIFN-CD, and LEVIR-CD |   |
| Year | Abbreviation | Title | Publication | Keywords | Experiment Datasets | Other |
|---|---|---|---|---|---|---|
| 2025 | FoBa | FoBa: A Foreground-Background co-Guided Method and New Benchmark for Remote Sensing Semantic Change Detection | arXiv2025 | foreground background co-guided, bi-temporal interaction, mamba, new benchmark | SECOND, JL1, and the proposed LevirSCD |   |
| 2025 | GSTM-SCD | GSTM-SCD: Graph-enhanced spatio-temporal state space model for semantic change detection in multi-temporal remote sensing images | ISPRS P&RS 2025 | State space model, Graph optimization, Spatio-temporal modeling | SECOND, Landsat-SCD, WUSU and DynamicEarthNet |   |
| 2024 | STMNet | STMNet: Single-Temporal Mask-Based Network for Self-Supervised Hyperspectral Change Detection | TGRS2024 | hyperspectral image, multiscale feature, single temporal, mask | Farmland dataset, Hermiston dataset, Bay dataset |   |
| 2024 | STCA | Towards transferable building damage assessment via unsupervised single-temporal change adaptation | RSE2024 | Unsupervised adaptation, single-temporal learning, semantic change detection | xBD, Turkey–Syria earthquake (2023), Kalehe DRC flooding (2023), Maui Hawaii fire (2023) | - |
| 2024 | ChangeSparse | Unifying Remote Sensing Change Detection via Deep Probabilistic Change Models: from Principles, Models to Applications | ISPRS P&RS 2024 | Probabilistic change model, sparsity of change, sparse change transformer | CDD, S2Looking, California Flood dataset, xBD, SECOND, DynamicEarthNet | |
| 2024 | ChangeStar2 | Single-Temporal Supervised Learning for Universal Remote Sensing Change Detection | IJCV2024 | Universal change detection, single-temporal supervised learning | WHU-CD, CDD, xBD, LEVIR-CD, S2Looking, SpaceNet8, DynamicEarthNet, SECOND | |
| 2024 | BiFA | BiFA: Remote Sensing Image Change Detection With Bitemporal Feature Alignment | TGRS2024 | Bitemporal interaction, feature alignment, flow field | WHU-CD, LEVIR-CD, LEVIR-CD+, SYSU-CD, DSIFN-CD, and CLCD |   |
| 2024 | CDMamba | CDMamba: Incorporating Local Clues Into Mamba for Remote Sensing Image Binary Change Detection | TGRS2025 | Mamba, bitemporal interaction, state space model | WHU-CD, CDD, LEVIR-CD, LEVIR-CD+, and CLCD |   |
| 2024 | CDMask | Rethinking Remote Sensing Change Detection With A Mask View | arXiv2024 | Mask view, mask-level Classification, MaskFormer | WHU-CD, LEVIR-CD, SYSU-CD, DSIFN-CD, and CLCD | |
| 2024 | ChangeMamba | ChangeMamba: Remote Sensing Change Detection With Spatiotemporal State Space Model | TGRS2024 | Mamba, spatiotemporal relationship, state space model | WHU-CD, xBD, SECOND, LEVIR-CD+, and SYSU-CD |   |
| 2024 | MaskCD | MaskCD: A Remote Sensing Change Detection Network Based on Mask Classification | TGRS2024 | Deformable attention, mask classification, masked cross-attention | LEVIR-CD, CLCD, SYSU-CD, EGY-BCD, and GVLM-CD |   |
| 2024 | M-CD | A Mamba-based Siamese Network for Remote Sensing Change Detection | arXiv2024 | Mamba, state space model, difference Module | WHU-CD, CDD, DSIFN-CD, and LEVIR-CD |   |
| 2024 | ScratchFormer | Remote Sensing Change Detection With Transformers Trained From Scratch | TGRS2024 | Trained from scratch, shuffled sparse-attention operation, change-enhanced feature fusion, | WHU-CD, OSCD, CDD, DSIFN-CD, and LEVIR-CD |   |
| 2024 | SitsSCD | Satellite Image Time Series Semantic Change Detection: Novel Architecture and Analysis of Domain Shift | arXiv2024 | Temporal attention, Temporal shift, Spatial shift | DynamicEarthNet, MUDS |   |
| 2023 | 3DCD | Inferring 3D change detection from bitemporal optical images | ISPRS P&RS 2023 | 3D Change Detection, Elevation change detection | 3DCD |   |
| 2023 | Siamese KPConv | Siamese KPConv: 3D multiple change detection from raw point clouds using deep learning | ISPRS P&RS 2023 | 3D Change Detection, Siamese network, 3D Kernel Point Convolution | Urb3DCD–V2, AHN-CD, Change3D |   |
| 2023 | MapFormer | MapFormer: Boosting Change Detection by Using Pre-change Information | ICCV2023 | Conditional Change Detection, multi-modal feature fusion, cross-modal contrastive loss | HRSCD, DynamicEarthNet |   |
| 2023 | CACo | Change-Aware Sampling and Contrastive Learning for Satellite Images | CVPR2023 | Self-supervised learning, Change-Aware Contrastive Loss | OSCD, DynamicEarthNet, EuroSat, and BigEarthNet |   |
| 2023 | Self-Pair | Self-Pair: Synthesizing Changes from Single Source for Object Change Detection in Remote Sensing Imagery | WACV2023 | Synthetic data, single-temporal supervision, visual similarity in unchanged area | WHU-CD, SpaceNet2, xBD, LEVIR-CD |   |
| 2022 | Changer | Changer: Feature Interaction is What You Need for Change Detection | TGRS2023 | Feature Interaction | LEVIR-CD, S2Looking | |
| 2022 | ChangeMask | ChangeMask: Deep Multi-task Encoder-Transformer-Decoder Architecture for Semantic Change Detection | ISPRS P&RS 2022 | Multi-task learning, temporal symmetry, multi-temporal | SECOND, Hi-UCD | |
| 2022 | FHD | Feature Hierarchical Differentiation for Remote Sensing Image Change Detection | GRSL2022 | Hierarchical differentiation, time-specific features | DSIFN, LEVIR-CD, LEVIR-CD+, S2Looking |   |
| 2022 | SST-Former | Spectral–spatial–temporal transformers for hyperspectral image change detection | TGRS2022 | Hyperspectral, cross-attention, self-attention | Farmland CD dataset, Barbara CD dataset, and Bay Area CD dataset |   |
| 2022 | CDViT | A Divided Spatial and Temporal Context Network for Remote Sensing Change Detection | JSTARS2022 | Self-attention, spatial-temporal transformer | WHU-CD, LEVIR-CD |   |
| 2022 | ChangeFormer | A Transformer-Based Siamese Network for Change Detection | IGARSS2022 | Transformer Siamese network, attention mechanism | DSIFN-CD, and LEVIR-CD |   |
| 2021 | BIT | Remote Sensing Image Change Detection with Transformers | TGRS2021 | Transformer | WHU-CD, DSIFN-CD, LEVIR-CD |   |
| Year | Abbreviation | Title | Publication | Keywords | Experiment Datasets | Other |
|---|---|---|---|---|---|---|
| 2025 | PRO-HRSCD | Rethinking Semantic Change Detection From a Semantic Alignment Perspective | TGRS2025 | Feature space alignment, multitask learning, prototype learning | SECOND, and Landsat-SCD |   |
| 2024 | ClearSCD | The ClearSCD model: Comprehensively leveraging semantics and change relationships for semantic change detection in high spatial resolution remote sensing imagery | ISPRS P&RS 2024 | Multi-task learning, contrastive learning, change vector analysis | Hi-UCD, LsSCD |   |
| 2024 | SSLChange | SSLChange: A Self-Supervised Change Detection Framework Based on Domain Adaptation | TGRS2024 | Domain adaption, hierarchical features, image contrastive learning | CDD, LEVIR-CD |   |
| 2024 | U-Net, U-Net SiamDiff, and U-Net SiamConc | A Change Detection Reality Check | ICLR2024W | Reality Check, Benchmarking | WHU-CD, LEVIR-CD |   |
| 2023 | I3PE | Exchange means change: An unsupervised single-temporal change detection framework based on intra-and inter-image patch exchange | ISPRS P&RS 2023 | Single-temporal change detection, image patch exchange, adaptive clustering | SYSU-CD, SECOND, Wuhan dataset |   |
| 2023 | A2Net | Lightweight Remote Sensing Change Detection With Progressive Feature Aggregation and Supervised Attention | TGRS2023 | Lightweight, progressive feature aggregation, supervised Attention | WHU-CD, LEVIR-CD and SYSU-CD |   |
| 2023 | DMINet | Change Detection on Remote Sensing Images Using Dual-Branch Multilevel Intertemporal Network | TGRS2023 | Dual-branch difference acquisition, intertemporal joint-attention, multilevel aggregation | WHU-CD, GZ-CD, LEVIR-CD, and SYSU-CD |   |
| 2023 | AFCF3D-Net | Adjacent-level feature cross-fusion with 3D CNN for remote sensing image change detection | TGRS2023 | 3D CNN, feature cross-fusion, full-scale connection | WHU-CD, LEVIR-CD, SYSU-CD |   |
| 2023 | LightCDNet | LightCDNet: Lightweight Change Detection Network Based on VHR Images | GRSL2023 | Early fusion, lightweight, deep supervised fusion | LEVIR-CD |   |
| 2023 | USSFC-Net | Ultralightweight Spatial–Spectral Feature Cooperation Network for Change Detection in Remote Sensing Images | TGRS2023 | Ultralightweight, multiscale feature extraction, spatial–spectral feature cooperation | CDD, DSIFN-CD, LEVIR-CD |   |
| 2023 | SAR-CD | Improved Difference Images for Change Detection Classifiers in SAR Imagery Using Deep Learning | TGRS2023 | Mapping transformation function, SAR, U-Net | SCDD |   |
| 2022 | RDPNet | RDP-Net: Region detail preserving network for change detection | TGRS2022 | Training strategy, edge loss, lightweight backbone | CDD,LEVIR-CD |   |
| 2022 | FFCTL | A full-level fused cross-task transfer learning method for building change detection using noise-robust pretrained networks on crowdsourced labels | RSE2022 | Transfer learning, crowdsourced label,pseudo label | ZY-3 building and change detection dataset |   |
| 2022 | SaDL_CD | Semantic-aware Dense Representation Learning for Remote Sensing Image Change Detection | TGRS2022 | Self-supervised learning, semantic-aware representation learning | WHU-CD, GZ-CD, LEVIR-CD |   |
| 2022 | TinyCD | TINYCD: A (Not So) Deep Learning Model For Change Detection | Neural Comput & Applic 2022 | Lightweight, tiny Model, siamese U-Net architecture, feature interaction | WHU-CD, LEVIR-CD |   |
| 2022 | SDACD | An End-to-end Supervised Domain Adaptation Framework for Cross-Domain Change Detection | PR2022 | Supervised Domain Adaptation, Image Adaptation, Feature Adaptation | CDD, and WHU-CD |   |
| 2022 | Bi-SRNet | Bi-temporal semantic reasoning for the semantic change detection in HR remote sensing images | TGRS2022 | Triple-branch, semantic correlations | SECOND |   |
| 2022 | SemiCD | Revisiting Consistency Regularization for Semi-supervised Change Detection in Remote Sensing Images | arXiv2022 | Semi-supervised, Consistency Regularization | WHU-CD, LEVIR-CD |   |
| 2022 | FCCDN | FCCDN: Feature Constraint Network for VHR Image Change Detection | ISPRS P&RS 2022 | Self-supervised learning, non-local feature pyramid network, dual encoder-decoder backbone | WHU-CD, LEVIR-CD, SECOND |   |
| 2021 | ChangeStar | Change is Everywhere: Single-Temporal Supervised Object Change Detection in Remote Sensing Imagery | ICCV2021 | Single-temporal supervision, temporal symmetry | xBD, SpaceNet2, WHU-CD, LEVIR-CD |   |
| 2021 | ChangeOS | Building damage assessment for rapid disaster response with a deep object-based semantic change detection framework: from natural disasters to man-made disasters | RSE2021 | Semantic change detection, disaster response, OBIA | xBD, The Beirut port explosion (2020), The Bata military barracks explosion (2021) |   |
| 2021 | Optical-SAR-CD | Self-supervised multisensor change detection | TGRS2021 | Self-supervised, Multisensor | OSCD (Sentinel-2 and Sentinel-1) | - |
| 2021 | CEECNet | Looking for change? Roll the Dice and demand Attention | RS2021 | Dice similarity, attention module | WHU-CD, LEVIR-CD |   |
| 2021 | ESCNet | ESCNet: An End-to-End Superpixel-Enhanced Change Detection Network for Very-High-Resolution Remote Sensing Images | TNNLS2021 | Superpixel segmentation, adaptive superpixel merging | SZTAKI, CDD |   |
| 2021 | SeCo | Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data | ICCV2021 | Self-supervised learning | BigEarthNet, EuroSAT, OSCD |   |
| 2021 | SRCDNet | Super-resolution-based Change Detection Network with Stacked Attention Module for Images with Different Resolutions | TGRS2021 | Super-resolution, metric learning | BCDD, CDD, GZ-CD |   |
| 2021 | IAug-CDNet | Adversarial Instance Augmentation for Building Change Detection in Remote Sensing Images | TGRS2021 | Adversarial instance augmentation, synthetic data | WHU-CD, LEVIR-CD |   |
| 2021 | SNUNet-CD | SNUNet-CD: A Densely Connected Siamese Network for Change Detection of VHR Images | GRSL2021 | Fully convolutional siamese network | CDD | |
| 2021 | DDNet | Change Detection in Synthetic Aperture Radar Images Using a Dual-Domain Network | GRSL2021 | SAR, frequency domain | Ottawa dataset, Sulzberger dataset, Yellow River dataset |   |
| 2021 | ACDA | Hyperspectral anomaly change detection based on autoencoder | JSTARS2021 | Hyperspectral, Anomaly Change Detection, Autoencoder | Viareggio 2013 |   |
| 2018 | FC-EF, FC-Siam-diff, FC-Siam-conc | Fully convolutional siamese networks for change detection | ICIP2018 | Fully Convolutional Siamese Networks | SZTAKI, OSCD |   |
| Year | Abbreviation | Title | Publication |
|---|---|---|---|
| 2013 | SFA | Slow Feature Analysis for Change Detection in Multispectral Imagery | TGRS2013 |
| 2009 | PCA-Kmeans | Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and k-Means Clustering | GRSL2009 |
| 2007 | IR-MAD | The Regularized Iteratively Reweighted Multivariate Alteration Detection | TIP2007 |
| 1998 | MAD | Multivariate Alteration Detection (MAD) and MAF Postprocessing in Multispectral, Bitemporal Image Data: New Approaches to Change Detection Studies | RSE1998 |
| 1980 | CVA | Change Vector Analysis: An Approach for Detecting Forest Changes with Landsat | LARS symposia 1980 |
| Year | Title | Publication | Description |
|---|---|---|---|
| 2025 | 深度学习遥感变化检测研究进展：像素-对象-场景 | 遥感技术与应用2025 | 本文从像素级、对象级和场景级三个层次系统总结深度学习在遥感变化检测中的研究进展，结合典型案例分析其实际应用，并展望其未来发展趋势。 |
| 2025 | On the use of Graphs for Satellite Image Time Series | arXiv2025 | Explores the integration of graph-based techniques for spatio-temporal analysis of satellite image time series, focusing on the construction of spatio-temporal graphs and their applications in tasks such as land cover mapping and water resource forecasting, along with future research perspectives. |
| 2025 | A Survey of Sample-Efficient Deep Learning for Change Detection in Remote Sensing: Tasks, Strategies, and Challenges | GRSM2025 | Summarizes literature on deep learning-based change detection methods for different tasks and strategies in sample-limited scenarios, discussing recent advances in image generation, self-supervision, and visual foundation models to address data scarcity. |
| 2025 | Deep learning change detection techniques for optical remote sensing imagery: Status, perspectives and challenges | JAG2025 | Systematically summarizes datasets, theories, and methods of change detection for optical remote sensing imagery, analyzing AI-based algorithms from the perspective of algorithm granularity and discussing challenges and trends in the AI era. Updates are available at daifeng2016/Awesome-Optical-Remote-Sensing-Datasets-and-Methods. |
| 2024 | Deep-Learning for Change Detection Using Multi-Modal Fusion of Remote Sensing Images: A Review | RS2024 | Explores the application of deep learning for change detection in remote sensing imagery using heterogeneous data sources, including multispectral, hyperspectral, radar, and multitemporal imagery, and discusses public datasets, models, challenges, and future trends. |
| 2024 | Deep Learning for Satellite Image Time-Series Analysis: A review | GRSM2024 | Summarizes state-of-the-art methods for modeling environmental and agricultural variables from satellite image time series (SITS) using deep learning, addressing the complexity of SITS data and its applications in land and natural resource management. |
| 2024 | Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review through Various Learning Paradigms | RS2024 | Comprehensively examines deep learning-based change detection in remote sensing, covering key architectures, learning paradigms (supervised, semi-supervised, weakly supervised, and unsupervised), benchmark datasets, and emerging opportunities such as self-supervised learning, foundation models, and multimodal data fusion, while highlighting current challenges and promising future research directions to advance the field. |
| 2024 | Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review | RS2024 | Presents a comprehensive survey of deep learning-based change detection in remote sensing over the past decade, offering a systematic taxonomy from perspectives of algorithm granularity, supervision modes, and frameworks, while reviewing key datasets, evaluation metrics, state-of-the-art performance, and identifying promising future research directions to guide and inspire the community. |
| 2023 | 深度学习的遥感变化检测综述：文献计量与分析 | 遥感学报2023 | 本文综述了基于深度学习的遥感变化检测研究进展，从像素、对象和场景三个粒度系统梳理方法体系，指出对象与场景级方法更具优势，并强调未来需突破多模态异质数据融合、非理想样本处理及多元变化信息提取等挑战，以推动其在多领域更广泛、智能化的应用。 |
| 2023 | 人工智能时代的遥感变化检测技术：继承、发展与挑战 | 遥感学报2023 | 本文系统梳理了人工智能时代下光学遥感影像变化检测技术从传统方法向数据—模型—知识联合驱动的智能化转型历程，分析了无监督、监督与弱监督三类方法的发展趋势，并指出未来需重点突破模型可解释性、泛化迁移能力及跨场景跨领域应用等关键瓶颈问题。相关讲解视频详见：【前沿进展】变化检测与深度学习。 |
| 2023 | 3D urban object change detection from aerial and terrestrial point clouds: A review | JAG2023 | Reviews developments in 3D change detection for urban objects using point cloud data, analyzing buildings, street scenes, urban trees, and construction sites, and discusses data sources, methods, and future challenges. |
| 2023 | Change detection of urban objects using 3D point clouds: A review | ISPRS P&RS 2023 | Provides a comprehensive review of point-cloud-based 3D change detection for urban objects, covering data registration, variance estimation, change analysis, and applications in land cover monitoring, vegetation surveys, and construction automation. |
| 2022 | Deep learning for change detection in remote sensing: a review | GSIS2022 | Analyzes why deep learning enhances remote sensing change detection by examining its improved information representation, methodological advances, and performance gains across spectral, spatial, temporal, and multi-sensor dimensions, while also identifying key limitations and future directions for deep learning change detection development. |
| 2022 | Land Cover Change Detection Techniques: Very-high-resolution optical images: A review | GRSM2022 | Reviews land cover change detection techniques using very-high-resolution remote sensing images, focusing on the ability to capture detailed changes and discussing various methods and applications. |
| 2022 | A Survey on Deep Learning-Based Change Detection from High-Resolution Remote Sensing Images | RS2022 | Reviews deep learning-based change detection methods for high-resolution remote sensing images, categorizing algorithms by network architecture, and discusses datasets, evaluation metrics, challenges, and future research directions. |
| 2022 | A review of multi-class change detection for satellite remote sensing imagery | GSIS2022 | Provides a comprehensive review of Multi-class Change Detection (MCD) in remote sensing, covering its background, key challenges, benchmark datasets, methodological categories, real-world applications, and future research directions, aiming to fill the gap in existing literature and serve as a foundational reference for advancing fine-grained land change analysis beyond traditional binary detection. |
| 2021 | Change Detection From Very-High-Spatial-Resolution Optical Remote Sensing Images: Methods, applications, and future directions | GRSM2021 | Provides a comprehensive overview of change detection in very-high-spatial-resolution (≤5 m) remote sensing images, systematically examining current methods, real-world applications, and future research directions to address challenges such as limited spectral information, spectral variability, and geometric distortions. |
| 2020 | A survey of change detection methods based on remote sensing images for multi-source and multi-objective scenarios | RS2020 | Surveys change detection methods for multi-source remote sensing images and multi-objective scenarios, summarizing a general framework including change information extraction, data fusion, and analysis, and discusses future directions. |
| 2020 | Change Detection Based on Artificial Intelligence: State-of-the-Art and Challenges | RS2020 | Reviews the state-of-the-art methods, applications, and challenges of AI for change detection, covering data sources, deep learning frameworks, and unsupervised schemes, and discusses issues like heterogeneous data processing and AI reliability. Updates are available at MinZHANG-WHU/Change-Detection-Review. |
| 2019 | A Review of Change Detection in Multitemporal Hyperspectral Images: Current Techniques, Applications, and Challenges | GRSM2019 | Presents a comprehensive review of change detection in hyperspectral remote sensing images, covering fundamental concepts, methodological categories, current techniques, and key challenges, while demonstrating state-of-the-art approaches through experimental results to highlight the unique potential and complexity of exploiting high spectral resolution for fine-scale land-cover change monitoring. |
| 2018 | 多时相遥感影像变化检测方法综述 | 武汉大学学报 (信息科学版) 2018 | 本文系统回顾了多时相遥感影像变化检测技术的发展历程，从预处理、方法分类到精度评价全面梳理研究进展，指出当前尚无普适性通用方法，并分析核心难点与应对策略，旨在推动该领域向更深入、更系统方向发展。 |
| 2017 | 多时相遥感影像变化检测的现状与展望 | 测绘学报2017 | 本文围绕多时相遥感影像变化检测的基本流程，从预处理、方法、阈值分割到精度评价系统梳理最新研究进展，总结其在生态环境监测与城市发展等领域的应用，并展望高光谱与高分辨率影像驱动下的未来发展方向。 |
| 2017 | Change detection using landsat time series: A review of frequencies, preprocessing, algorithms, and applications | ISPRS P&RS 2017 | Reviews change detection studies based on Landsat time series, covering frequencies, preprocessing steps, algorithms, and applications, and discusses the impact of free access to Landsat data on change detection methodologies. |
| 2016 | Optical remotely sensed time series data for land cover classification: A review | ISPRS P&RS 2016 | Reviews the use of optical remote sensing time series data for land cover classification, discussing issues and opportunities in generating annual land cover products and methods for incorporating time series information. |
| 2016 | SAR影像变化检测研究进展 | 计算机研究与发展2015 | 本文系统梳理了SAR影像变化检测的经典流程与传统方法，重点综述近年来在差异图生成及阈值、聚类、图切、水平集等分析方法上的新兴算法改进，并通过两组数据集定量验证其性能，最后展望了该领域仍需深入研究的关键方向。 |
| 2015 | A critical synthesis of remotely sensed optical image change detection techniques | RSE2015 | Provides a critical synthesis of remote sensing change detection techniques, organizing the literature by unit of analysis and comparison method to reduce conceptual overlap and guide future research. |
| 2013 | Change detection from remotely sensed images: From pixel-based to object-based approaches | ISPRS P&RS 2013 | Reviews change detection methodologies from pixel-based to object-based approaches, discussing the potential of object-based methods and data mining techniques with the advent of very-high-resolution imagery. |
| 2012 | Object-based change detection | IJRS2012 | Discusses object-based change detection (OBCD) using high-spatial-resolution imagery, comparing it with pixel-based approaches and reviewing algorithms and applications for detailed change information extraction. |
| 2012 | A review of large area monitoring of land cover change using Landsat data | RSE2012 | Reviews methods for large area monitoring of land cover change using Landsat data, focusing on forest cover change, and discusses radiometric correction, temporal updating, and the impact of free access to terrain-corrected data. |
| 2011 | 多时相遥感影像变化检测综述 | 地理信息世界2011 | 本文系统回顾多时相遥感影像变化检测的发展现状，从环境变化特性出发，围绕预处理、方法分类、精度评估等四大方面梳理技术演进，并提出融合多源数据、集成处理与智能方法的综合解决方案，同时指出当前挑战与应对策略，以推动该领域深入发展。 |
| 2005 | Image change detection algorithms: a systematic survey | TIP2004 | Provides a systematic survey of image change detection algorithms, covering common processing steps and core decision rules, and discusses preprocessing methods, consistency enforcement, and performance evaluation principles. |
| 2004 | Digital change detection methods in ecosystem monitoring: a review | IJRS2004 | Reviews digital change detection methods in ecosystem monitoring, covering multi-temporal, multi-spectral data techniques, preprocessing routines, and change detection algorithms, and highlights the complementarity between different methods. |
| 2004 | Change detection techniques | IJRS2004 | Summarizes and reviews change detection techniques using remote sensing data, highlighting image differencing, principal component analysis, and post-classification comparison as common methods, and discusses emerging techniques like spectral mixture analysis and neural networks. |
| 2003 | 利用遥感影像进行变化检测 | 武汉大学学报 (信息科学版) 2003 | 本文针对遥感影像变化检测的紧迫需求与技术难点，提出影像配准与变化检测同步求解的新思路，并探讨其拓展至三维变化检测的可行性，系统比较七类主流方法，最后指明未来重点研究方向。 |
| Year | Target | Contest | Track | Image Pairs | Image Size | Resolution | Other |
|---|---|---|---|---|---|---|---|
| 2025 | Building | AI for Earthquake Response | Detect damaged vs. undamaged buildings by analyzing high-resolution pre- and post-event satellite imagery | - | - | - | - |
| 2024 | Land cover | ISPRS第一技术委员会多模态遥感应用算法智能解译大赛 | 基于高分辨率可见光图像的感兴趣区域内部变化智能检测 | 4,000 | 512×512 | 2m | - |
| 2024 | Land cover | “吉林一号”杯卫星遥感应用青年创新创业大赛 | 高分辨率遥感影像全要素变化检测研究 | 5,000 | 512×512 | <0.75m | - |
| 2023 | Cropland | “吉林一号”杯卫星遥感应用青年创新创业大赛 | 基于高分辨率卫星影像的耕地变化检测 | 8,000 | 256×256 | <0.75m | - |
| 2023 | Land cover | “国丰东方慧眼杯”遥感影像智能处理算法大赛 | 对象级变化检测 | >6,000 | 512×512 | 1-2m | - |
| 2022 | Land cover | “航天宏图杯”遥感影像智能处理算法大赛 | 遥感影像变化检测 | >6,000 | 512×512 | 1-2m | - |
| 2022 | Flood | SpaceNet8: Flood Detection Challenge | Flood Detection Challenge Using Multiclass Segmentation | 12 | 1,300×1,300 | 0.3-0.8m | Dataset Paper, Solution Paper |
| 2021 | Land cover | IEEE GRSS Data Fusion Contest | Multitemporal Semantic Change Detection | 2,250 | - | - | Outcome Paper |
| 2021 | Land cover | DynamicEarthNet Challenge | Weakly-Supervised Unsupervised Binary Land Cover Change Detection, Multi-Class Change Detection | 54,750 | 1,024x1,024 | 3.0 | Top1 Solution, Dataset Paper |
| 2021 | Land cover | “昇腾杯”遥感影像智能处理算法大赛 | 耕地建筑物变化检测 | >6,000 | 512×512 | 1-2m | Top4 Solution, Top5 Solution |
| 2021 | Building | 遥感图像智能解译技术挑战赛 | 遥感图像建筑物变化检测 | 10,000 | 512×512 | - | - |
| 2021 | Building | 慧眼“天智杯”人工智能挑战赛 | 可见光建筑智能变化检测 | 5,000 | 1,024×1,024 | 0.5-0.7m | - |
| 2020 | Land cover | 商汤科技首届AI遥感解译大赛 | 变化检测 | 4,662 | 512×512 | 0.5-3m | Top1 Solution |
| 2020 | Land cover | SpaceNet 7: Multi-Temporal Urban Development Challenge | Multi-Temporal Urban Development Challenge | - | 1,024×1,024 | 4m | Solutions, Dataset Paper |
| 2019 | Building | xView2 Challenge (or xBD) | Building Damage Assessment | 11,034 | 1,024×1,024 | - | Dataset Paper |
| Name | Description |
|---|---|
| Maxar Open Data Program | The Maxar Open Data Program provides pre- and post-event satellite imagery (from WorldView-3 and other sensors) for select sudden-onset major crises, along with crowdsourced damage assessments. |
| 吉林一号资源库 | 提供高分辨率卫星影像和专题数据，支持自然灾害监测、农业估产、生态环境保护、水利管理及应急响应等多领域应用。部分数据集仅限教育认证用户。 |
| Planet Disaster Datasets | Planet makes available select imagery for major disaster events, including major earthquakes, floods, storms, wildfires, and human-made disasters. To download the data, users must complete a form for access qualification. |
| The International Charter: Space And Major Disasters | Disaster mapping results and analyses are available for various global hazards, but the underlying satellite imagery is not directly provided. |
- daifeng2016/Awesome-Optical-Remote-Sensing-Datasets-and-Methods
- chrieke/awesome-satellite-imagery-datasets
- MinZHANG-WHU/Change-Detection-Review
If you find our project useful in your research, please consider citing:
@misc{awesome_rscd_2019,
title={Awesome Remote Sensing Change Detection},
author={Awesome RSCD Contributors},
howpublished = {\url{https://github.com/wenhwu/awesome-remote-sensing-change-detection}},
year={2019}
}