{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T09:46:15Z","timestamp":1775209575569,"version":"3.50.1"},"reference-count":53,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T00:00:00Z","timestamp":1775174400000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["42201449"],"award-info":[{"award-number":["42201449"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Zhejiang Provincial Natural Science Foundation of China","award":["LQ23D010005"],"award-info":[{"award-number":["LQ23D010005"]}]},{"name":"Ningbo Science and Technology Innovation Project","award":["2022Z075"],"award-info":[{"award-number":["2022Z075"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Change detection (CD) in remote sensing images has long been of great interest to researchers. Due to the substantial time and effort required for data annotation in remote sensing, semi-supervised methods have attracted extensive attention as a promising solution for achieving satisfactory performance under limited samples. However, existing semi-supervised approaches often encounter significant challenges, most notably class imbalance and subtle changes in feature distributions, making it difficult to distinguish changes from unchanged regions. Here, we propose a novel strategy, named NF-SemiCD, which incorporates normalizing flows in a semi-supervised architecture. We leverage normalizing flows to characterize the feature distribution of unchanged regions and derive a probability distribution model. Its sensitivity to changes in probability allows us to get a probability feature map, which provides useful information on deep difference features. To this end, we devise a three-stage training scheme: (1) training an encoder-decoder network with labeled data, (2) training a normalizing flow decoder on labeled data, and (3) training the encoder-decoder network with all data. Experiments on three benchmark datasets demonstrate that NF-SemiCD outperforms existing state-of-the-art methods, highlighting its potential for improved change detection under semi-supervised settings.<\/jats:p>","DOI":"10.7717\/peerj-cs.3751","type":"journal-article","created":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T08:48:26Z","timestamp":1775206106000},"page":"e3751","source":"Crossref","is-referenced-by-count":0,"title":["NF-SemiCD: semi-supervised remote sensing change detection with normalizing flows"],"prefix":"10.7717","volume":"12","author":[{"given":"Guitao","family":"Yu","sequence":"first","affiliation":[{"name":"School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan, China"},{"name":"Intelligent Kitchen Engineering Research Center of Zhejiang Province, Intelligent Kitchen Engineering Research Center of Zhejiang Province, Ningbo, Zhejiang, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dongjie","family":"Zhang","sequence":"additional","affiliation":[{"name":"Ningbo University, Electrical Engineering and Computer Science, Ningbo, Zhejiang, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ye","family":"Zheng","sequence":"additional","affiliation":[{"name":"Ningbo University, Electrical Engineering and Computer Science, Ningbo, Zhejiang, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Baochen","family":"Yao","sequence":"additional","affiliation":[{"name":"Ningbo University, Electrical Engineering and Computer Science, Ningbo, Zhejiang, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chengbin","family":"Peng","sequence":"additional","affiliation":[{"name":"Ningbo University, Electrical Engineering and Computer Science, Ningbo, Zhejiang, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"4443","published-online":{"date-parts":[[2026,4,3]]},"reference":[{"issue":"1","key":"10.7717\/peerj-cs.3751\/ref-1","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1007\/bf03325962","article-title":"Coastline change detection using remote sensing","volume":"4","author":"Alesheikh","year":"2007","journal-title":"International Journal of Environmental Science & Technology"},{"key":"10.7717\/peerj-cs.3751\/ref-2","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2204.08454","article-title":"Revisiting consistency regularization for semi-supervised change detection in remote sensing images","author":"Bandara","year":"2022a"},{"key":"10.7717\/peerj-cs.3751\/ref-3","first-page":"207","article-title":"A transformer-based siamese network for change detection","author":"Bandara","year":"2022b"},{"key":"10.7717\/peerj-cs.3751\/ref-4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2022.3161337","article-title":"UCDNet: a deep learning model for urban change detection from bi-temporal multispectral sentinel-2 satellite images","volume":"60","author":"Basavaraju","year":"2022","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-5","doi-asserted-by":"crossref","first-page":"325","DOI":"10.1117\/12.2533812","article-title":"Unsupervised change-detection based on convolutional-autoencoder feature extraction","volume-title":"Image and Signal Processing for Remote Sensing XXV","volume":"11155","author":"Bergamasco","year":"2019"},{"key":"10.7717\/peerj-cs.3751\/ref-6","doi-asserted-by":"publisher","DOI":"10.1109\/IGARSS53475.2024.10640593","article-title":"Time travelling pixels: bitemporal features integration with foundation model for remote sensing image change detection","author":"Chen","year":"2023b"},{"key":"10.7717\/peerj-cs.3751\/ref-7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2021.3095166","article-title":"Remote sensing image change detection with transformers","volume":"60","author":"Chen","year":"2022","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"issue":"10","key":"10.7717\/peerj-cs.3751\/ref-8","doi-asserted-by":"publisher","first-page":"1662","DOI":"10.3390\/rs12101662","article-title":"A spatial-temporal attention-based method and a new dataset for remote sensing image change detection","volume":"12","author":"Chen","year":"2020","journal-title":"Remote Sensing"},{"issue":"11","key":"10.7717\/peerj-cs.3751\/ref-9","doi-asserted-by":"publisher","first-page":"6217","DOI":"10.5194\/essd-17-6217-2025","article-title":"Bright: a globally distributed multimodal building damage assessment dataset with very-high-resolution for all-weather disaster response","volume":"17","author":"Chen","year":"2025","journal-title":"Earth System Science Data"},{"key":"10.7717\/peerj-cs.3751\/ref-10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2024.3417253","article-title":"ChangeMamba: remote sensing change detection with spatiotemporal state space model","volume":"62","author":"Chen","year":"2024","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-11","doi-asserted-by":"publisher","DOI":"10.1016\/j.isprsjprs.2023.11.004","article-title":"Exchange means change: an unsupervised single-temporal change detection framework based on intra- and inter-image patch exchange","author":"Chen","year":"2023a"},{"key":"10.7717\/peerj-cs.3751\/ref-12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW50498.2020.00359","article-title":"RandAugment: practical automated data augmentation with a reduced search space","author":"Cubuk","year":"2019"},{"key":"10.7717\/peerj-cs.3751\/ref-13","first-page":"4063","article-title":"Fully convolutional siamese networks for change detection","author":"Daudt","year":"2018"},{"issue":"12","key":"10.7717\/peerj-cs.3751\/ref-14","doi-asserted-by":"publisher","first-page":"9976","DOI":"10.1109\/tgrs.2019.2930682","article-title":"Unsupervised deep slow feature analysis for change detection in multi-temporal remote sensing images","volume":"57","author":"Du","year":"2019","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"issue":"2","key":"10.7717\/peerj-cs.3751\/ref-15","doi-asserted-by":"publisher","first-page":"708","DOI":"10.1109\/jstars.2015.2477431","article-title":"Sparse unmixing-based change detection for multitemporal hyperspectral images","volume":"9","author":"Ert\u00fcrk","year":"2016","journal-title":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-16","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/lgrs.2021.3056416","article-title":"SNUNet-CD: a densely connected siamese network for change detection of VHR images","volume":"19","author":"Fang","year":"2022","journal-title":"IEEE Geoscience and Remote Sensing Letters"},{"issue":"8","key":"10.7717\/peerj-cs.3751\/ref-17","doi-asserted-by":"publisher","first-page":"1240","DOI":"10.1109\/lgrs.2019.2895656","article-title":"Sea ice change detection in SAR images based on convolutional-wavelet neural networks","volume":"16","author":"Gao","year":"2019","journal-title":"IEEE Geoscience and Remote Sensing Letters"},{"issue":"4","key":"10.7717\/peerj-cs.3751\/ref-18","doi-asserted-by":"publisher","first-page":"2417","DOI":"10.1109\/tgrs.2012.2210901","article-title":"A change detection approach to flood mapping in urban areas using TerraSAR-X","volume":"51","author":"Giustarini","year":"2012","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-19","doi-asserted-by":"publisher","DOI":"10.1109\/WACV51458.2022.00188","article-title":"CFLOW-AD: real-time unsupervised anomaly detection with localization via conditional normalizing flows","author":"Gudovskiy","year":"2021"},{"key":"10.7717\/peerj-cs.3751\/ref-20","article-title":"xBD: a dataset for assessing building damage from satellite imagery","author":"Gupta","year":"2019"},{"key":"10.7717\/peerj-cs.3751\/ref-21","doi-asserted-by":"publisher","first-page":"3867","DOI":"10.1109\/jstars.2023.3264802","article-title":"Hanet: A hierarchical attention network for change detection with bitemporal very-high-resolution remote sensing images","volume":"16","author":"Han","year":"2023","journal-title":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2024.3370568","article-title":"C2F-Semicd: a coarse-to-fine semi-supervised change detection method based on consistency regularization in high-resolution remote sensing images","volume":"62","author":"Han","year":"2024","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"issue":"7","key":"10.7717\/peerj-cs.3751\/ref-23","doi-asserted-by":"publisher","first-page":"5481","DOI":"10.1007\/s00521-024-10891-y","article-title":"An efficient and scalable semi-supervised framework for semantic segmentation","volume":"37","author":"Hao","year":"2025","journal-title":"Neural Computing and Applications"},{"issue":"1","key":"10.7717\/peerj-cs.3751\/ref-24","doi-asserted-by":"publisher","first-page":"117","DOI":"10.1016\/j.ijsbe.2015.02.005","article-title":"Monitoring urban growth and land use change detection with GIS and remote sensing techniques in Daqahlia governorate Egypt","volume":"4","author":"Hegazy","year":"2015","journal-title":"International Journal of Sustainable Built Environment"},{"key":"10.7717\/peerj-cs.3751\/ref-25","doi-asserted-by":"publisher","first-page":"108886","DOI":"10.1016\/j.engappai.2024.108886","article-title":"A multi-view consistency framework with semi-supervised domain adaptation","volume":"136","author":"Hong","year":"2024","journal-title":"Engineering Applications of Artificial Intelligence"},{"issue":"1","key":"10.7717\/peerj-cs.3751\/ref-26","doi-asserted-by":"publisher","first-page":"574","DOI":"10.1109\/tgrs.2018.2858817","article-title":"Fully convolutional networks for multisource building extraction from an open aerial and satellite imagery data set","volume":"57","author":"Ji","year":"2019","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-27","doi-asserted-by":"publisher","first-page":"565","DOI":"10.5194\/isprs-archives-xlii-2-565-2018","article-title":"Change detection in remote sensing images using conditional adversarial networks","volume":"XLII-2","author":"Lebedev","year":"2018","journal-title":"The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences"},{"key":"10.7717\/peerj-cs.3751\/ref-28","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2024.3365825","article-title":"A new learning paradigm for foundation model-based remote sensing change detection","author":"Li","year":"2024"},{"issue":"23","key":"10.7717\/peerj-cs.3751\/ref-29","doi-asserted-by":"publisher","first-page":"2844","DOI":"10.3390\/rs11232844","article-title":"The temporal dynamics of slums employing a CNN-based change detection approach","volume":"11","author":"Liu","year":"2019","journal-title":"Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-30","doi-asserted-by":"crossref","DOI":"10.1109\/ICASSP49357.2023.10097089","article-title":"Semi-supervised remote sensing image change detection using mean teacher model for constructing pseudo-labels","author":"Mao","year":"2023"},{"issue":"4","key":"10.7717\/peerj-cs.3751\/ref-31","doi-asserted-by":"publisher","first-page":"1369","DOI":"10.1109\/TPAMI.2019.2960224","article-title":"Semi-supervised semantic segmentation with high-and low-level consistency","volume":"43","author":"Mittal","year":"2019","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"issue":"6\u20137","key":"10.7717\/peerj-cs.3751\/ref-32","doi-asserted-by":"publisher","first-page":"1159","DOI":"10.1080\/014311600210128","article-title":"Land cover characterization and change detection for environmental monitoring of pan-Europe","volume":"21","author":"Mucher","year":"2000","journal-title":"International Journal of Remote Sensing"},{"issue":"2","key":"10.7717\/peerj-cs.3751\/ref-33","doi-asserted-by":"publisher","first-page":"125","DOI":"10.1016\/j.isprsjprs.2006.09.004","article-title":"Multiple support vector machines for land cover change detection: an application for mapping urban extensions","volume":"61","author":"Nemmour","year":"2006","journal-title":"ISPRS Journal of Photogrammetry and Remote Sensing"},{"issue":"57","key":"10.7717\/peerj-cs.3751\/ref-34","first-page":"1","article-title":"Normalizing flows for probabilistic modeling and inference","volume":"22","author":"Papamakarios","year":"2021","journal-title":"Journal of Machine Learning Research"},{"issue":"7","key":"10.7717\/peerj-cs.3751\/ref-35","doi-asserted-by":"publisher","first-page":"5891","DOI":"10.1109\/tgrs.2020.3011913","article-title":"SemiCDNet: a semisupervised convolutional neural network for change detection in high resolution remote-sensing images","volume":"59","author":"Peng","year":"2021","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"issue":"4","key":"10.7717\/peerj-cs.3751\/ref-36","doi-asserted-by":"publisher","first-page":"744","DOI":"10.3390\/rs13040744","article-title":"Deep learning of sea surface temperature patterns to identify ocean extremes","volume":"13","author":"Prochaska","year":"2021","journal-title":"Remote Sensing"},{"issue":"12","key":"10.7717\/peerj-cs.3751\/ref-37","doi-asserted-by":"publisher","first-page":"10047","DOI":"10.1109\/tgrs.2020.3043766","article-title":"Unsupervised change detection in satellite images with generative adversarial network","volume":"59","author":"Ren","year":"2021","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"issue":"6","key":"10.7717\/peerj-cs.3751\/ref-38","doi-asserted-by":"publisher","first-page":"3677","DOI":"10.1109\/tgrs.2018.2886643","article-title":"Unsupervised deep change vector analysis for multiple-change detection in VHR images","volume":"57","author":"Saha","year":"2019","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-39","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2021.3106381","article-title":"An unsupervised remote sensing change detection method based on multiscale graph convolutional network and metric learning","volume":"60","author":"Tang","year":"2022","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-40","first-page":"2517","article-title":"ADVENT: adversarial entropy minimization for domain adaptation in semantic segmentation","author":"Vu","year":"2019"},{"key":"10.7717\/peerj-cs.3751\/ref-41","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2022.3228016","article-title":"Reliable contrastive learning for semi-supervised change detection in remote sensing images","volume":"60","author":"Wang","year":"2022b","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-42","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2024.3357085","article-title":"CF-GCN: graph convolutional network for change detection in remote sensing images","volume":"62","author":"Wang","year":"2024","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-43","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2022.3176603","article-title":"An empirical study of remote sensing pretraining","volume":"61","author":"Wang","year":"2022a","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-44","doi-asserted-by":"publisher","first-page":"2272","DOI":"10.1109\/jstars.2023.3345017","article-title":"STCRNet: a semi-supervised network based on self-training and consistency regularization for change detection in VHR remote sensing images","volume":"17","author":"Wang","year":"2023","journal-title":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-45","article-title":"Blind super-resolution for remote sensing images via conditional stochastic normalizing flows","author":"Wu","year":"2022"},{"key":"10.7717\/peerj-cs.3751\/ref-46","doi-asserted-by":"crossref","DOI":"10.1109\/JURSE57346.2023.10144178","article-title":"UCDFormer: unsupervised change detection using real-time transformers","volume-title":"2023 Joint Urban Remote Sensing Event (JURSE)","author":"Xu","year":"2023b"},{"key":"10.7717\/peerj-cs.3751\/ref-47","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2022.110166","article-title":"Semi-supervised learning with pseudo-negative labels for image classification","author":"Xu","year":"2023a"},{"issue":"12","key":"10.7717\/peerj-cs.3751\/ref-48","doi-asserted-by":"publisher","first-page":"114171","DOI":"10.1016\/j.knosys.2025.114171","article-title":"Active learning with joint probabilistic modeling for point cloud semantic segmentation","volume":"328","author":"Yao","year":"2025","journal-title":"Knowledge-Based Systems"},{"issue":"13","key":"10.7717\/peerj-cs.3751\/ref-49","doi-asserted-by":"publisher","first-page":"4898","DOI":"10.1080\/01431161.2021.1906982","article-title":"NestNet: a multiscale convolutional neural network for remote sensing image change detection","volume":"42","author":"Yu","year":"2021","journal-title":"International Journal of Remote Sensing"},{"issue":"10","key":"10.7717\/peerj-cs.3751\/ref-50","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/lgrs.2024.3467260","article-title":"C\u00b2F\u00b2: cross-task cross-domain feature fusion for semi-supervised change detection","volume":"21","author":"Zhang","year":"2024","journal-title":"IEEE Geoscience and Remote Sensing Letters"},{"key":"10.7717\/peerj-cs.3751\/ref-51","doi-asserted-by":"publisher","first-page":"5406613","DOI":"10.1109\/tgrs.2023.3314452","article-title":"Joint self-training and rebalanced consistency learning for semi-supervised change detection","volume":"61","author":"Zhang","year":"2023a","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3751\/ref-52","doi-asserted-by":"publisher","first-page":"5401016","DOI":"10.1109\/tgrs.2023.3247605","article-title":"Semisupervised change detection with feature-prediction alignment","volume":"61","author":"Zhang","year":"2023b","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"issue":"2","key":"10.7717\/peerj-cs.3751\/ref-53","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.isprsjprs.2021.01.002","article-title":"Characterizing urban land changes of 30 global megacities using nighttime light time series stacks","volume":"173","author":"Zheng","year":"2021","journal-title":"ISPRS Journal of Photogrammetry and Remote Sensing"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3751.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3751.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3751.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3751.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T08:48:30Z","timestamp":1775206110000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3751"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,3]]},"references-count":53,"alternative-id":["10.7717\/peerj-cs.3751"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3751","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,3]]},"article-number":"e3751"}}