{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T05:06:36Z","timestamp":1780981596709,"version":"3.54.1"},"reference-count":66,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2022,5,25]],"date-time":"2022-05-25T00:00:00Z","timestamp":1653436800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Institute for Data Valorisation (IVADO)"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>When a natural disaster occurs, humanitarian organizations need to be prompt, effective, and efficient to support people whose security is threatened. Satellite imagery offers rich and reliable information to support expert decision-making, yet its annotation remains labour-intensive and tedious. In this work, we evaluate the applicability of convolutional neural networks (CNN) in supporting building damage assessment in an emergency context. Despite data scarcity, we develop a deep learning workflow to support humanitarians in time-constrained emergency situations. To expedite decision-making and take advantage of the inevitable delay to receive post-disaster satellite images, we decouple building localization and damage classification tasks into two isolated models. Our contribution is to show the complexity of the damage classification task and use established transfer learning techniques to fine-tune the model learning and estimate the minimal number of annotated samples required for the model to be functional in operational situations.<\/jats:p>","DOI":"10.3390\/rs14112532","type":"journal-article","created":{"date-parts":[[2022,5,25]],"date-time":"2022-05-25T08:41:33Z","timestamp":1653468093000},"page":"2532","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":40,"title":["On Transfer Learning for Building Damage Assessment from Satellite Imagery in Emergency Contexts"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0598-1050","authenticated-orcid":false,"given":"Isabelle","family":"Bouchard","sequence":"first","affiliation":[{"name":"Polytechnique Montreal, Montreal, QC H3T 1J4, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marie-\u00c8ve","family":"Rancourt","sequence":"additional","affiliation":[{"name":"HEC Montreal, Montreal, QC H3T 2A7, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Aloise","sequence":"additional","affiliation":[{"name":"Polytechnique Montreal, Montreal, QC H3T 1J4, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1471-646X","authenticated-orcid":false,"given":"Freddie","family":"Kalaitzis","sequence":"additional","affiliation":[{"name":"Oxford University, Oxford OX1 2JD, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2022,5,25]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1126\/science.aad8728","article-title":"Global trends in satellite-based emergency mapping","volume":"353","author":"Voigt","year":"2016","journal-title":"Science"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"151","DOI":"10.1007\/s10994-009-5152-4","article-title":"A theory of learning from different domains","volume":"79","author":"Blitzer","year":"2010","journal-title":"Mach. Learn."},{"key":"ref_3","unstructured":"Rolnick, D., Donti, P.L., Kaack, L.H., Kochanski, K., Lacoste, A., Sankaran, K., Ross, A.S., Milojevic-Dupont, N., Jaques, N., and Waldman-Brown, A. (2019). Tackling climate change with machine learning. arXiv."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Rausch, L., Friesen, J., Altherr, L.C., Meck, M., and Pelz, P.F. (2018). A holistic concept to design optimal water supply infrastructures for informal settlements using remote sensing data. Remote Sens., 10.","DOI":"10.3390\/rs10020216"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Kogan, F. (2019). Remote Sensing for Food Security, Springer.","DOI":"10.1007\/978-3-319-96256-6"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.compenvurbsys.2015.02.002","article-title":"Remote sensing for urban planning and management: The use of window-independent context segmentation to extract urban features in Stockholm","volume":"52","author":"Nielsen","year":"2015","journal-title":"Comput. Environ. Urban Syst."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Filipponi, F. (2019). Exploitation of sentinel-2 time series to map burned areas at the national level: A case study on the 2017 italy wildfires. Remote Sens., 11.","DOI":"10.3390\/rs11060622"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"4035","DOI":"10.1080\/0143116031000103853","article-title":"Remote sensing of tropical forest environments: Towards the monitoring of environmental resources for sustainable development","volume":"24","author":"Foody","year":"2003","journal-title":"Int. J. Remote Sens."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Schumann, G.J., Brakenridge, G.R., Kettner, A.J., Kashif, R., and Niebuhr, E. (2018). Assisting flood disaster response with earth observation data and products: A critical assessment. Remote Sens., 10.","DOI":"10.3390\/rs10081230"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1016\/j.isprsjprs.2017.05.002","article-title":"Simultaneous extraction of roads and buildings in remote sensing imagery with convolutional neural networks","volume":"130","author":"Alshehhi","year":"2017","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Shrestha, S., and Vanneschi, L. (2018). Improved fully convolutional network with conditional random fields for building extraction. Remote Sens., 10.","DOI":"10.3390\/rs10071135"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1016\/j.isprsjprs.2019.02.019","article-title":"Automatic building extraction from high-resolution aerial images and LiDAR data using gated residual refinement network","volume":"151","author":"Huang","year":"2019","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2793","DOI":"10.1109\/TPAMI.2017.2750680","article-title":"Learning building extraction in aerial scenes with convolutional networks","volume":"40","author":"Yuan","year":"2017","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Liu, P., Liu, X., Liu, M., Shi, Q., Yang, J., Xu, X., and Zhang, Y. (2019). Building footprint extraction from high-resolution images via spatial residual inception convolutional neural network. Remote Sens., 11.","DOI":"10.3390\/rs11070830"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"128774","DOI":"10.1109\/ACCESS.2019.2940527","article-title":"Automatic building extraction on high-resolution remote sensing imagery using deep convolutional encoder-decoder with spatial pyramid pooling","volume":"7","author":"Liu","year":"2019","journal-title":"IEEE Access"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Ma, J., Wu, L., Tang, X., Liu, F., Zhang, X., and Jiao, L. (2020). Building extraction of aerial images by a global and multi-scale encoder-decoder network. Remote Sens., 12.","DOI":"10.3390\/rs12152350"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1842","DOI":"10.1109\/JSTARS.2020.2991391","article-title":"Refined extraction of building outlines from high-resolution remote sensing imagery based on a multifeature convolutional neural network and morphological filtering","volume":"13","author":"Xie","year":"2020","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"4287","DOI":"10.1109\/TGRS.2020.3014312","article-title":"Scene-driven multitask parallel attention network for building extraction in high-resolution remote sensing images","volume":"59","author":"Guo","year":"2020","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"112589","DOI":"10.1016\/j.rse.2021.112589","article-title":"Deep building footprint update network: A semi-supervised method for updating existing building footprint from bi-temporal remote sensing images","volume":"264","author":"Guo","year":"2021","journal-title":"Remote Sens. Environ."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Cooner, A.J., Shao, Y., and Campbell, J.B. (2016). Detection of urban damage using remote sensing and machine learning algorithms: Revisiting the 2010 Haiti earthquake. Remote Sens., 8.","DOI":"10.3390\/rs8100868"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Fujita, A., Sakurada, K., Imaizumi, T., Ito, R., Hikosaka, S., and Nakamura, R. (2017, January 8\u201312). Damage detection from aerial images via convolutional neural networks. Proceedings of the 2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA), Nagoya, Japan.","DOI":"10.23919\/MVA.2017.7986759"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Sublime, J., and Kalinicheva, E. (2019). Automatic post-disaster damage mapping using deep-learning techniques for change detection: Case study of the Tohoku tsunami. Remote Sens., 11.","DOI":"10.3390\/rs11091123"},{"key":"ref_23","unstructured":"Doshi, J., Basu, S., and Pang, G. (2018). From satellite imagery to disaster insights. arXiv."},{"key":"ref_24","unstructured":"Van Etten, A., Lindenbaum, D., and Bacastow, T.M. (2018). Spacenet: A remote sensing dataset and challenge series. arXiv."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Demir, I., Koperski, K., Lindenbaum, D., Pang, G., Huang, J., Basu, S., Hughes, F., Tuia, D., and Raskar, R. (2018, January 18\u201322). Deepglobe 2018: A challenge to parse the earth through satellite images. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Salt Lake City, UT, USA.","DOI":"10.1109\/CVPRW.2018.00031"},{"key":"ref_26","unstructured":"Gupta, R., Hosfelt, R., Sajeev, S., Patel, N., Goodman, B., Doshi, J., Heim, E., Choset, H., and Gaston, M. (2019). xbd: A dataset for assessing building damage from satellite imagery. arXiv."},{"key":"ref_27","unstructured":"Durnov, V. (2020, March 01). Github\u2014DIUx-xView\/xView2_first_place: 1st Place Solution for \u2019xView2: Assess Building Damage\u2019 Challenge. Available online: https:\/\/github.com\/DIUx-xView\/xView2_first_place."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Shao, J., Tang, L., Liu, M., Shao, G., Sun, L., and Qiu, Q. (2020). BDD-Net: A General Protocol for Mapping Buildings Damaged by a Wide Range of Disasters Based on Satellite Imagery. Remote Sens., 12.","DOI":"10.3390\/rs12101670"},{"key":"ref_29","unstructured":"Gupta, R., and Shah, M. (2020). Rescuenet: Joint building segmentation and damage assessment from satellite imagery. arXiv."},{"key":"ref_30","unstructured":"Weber, E., and Kan\u00e9, H. (2020). Building disaster damage assessment in satellite imagery with multi-temporal fusion. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Hao, H., Baireddy, S., Bartusiak, E.R., Konz, L., LaTourette, K., Gribbons, M., Chan, M., Comer, M.L., and Delp, E.J. (2020). An attention-based system for damage assessment using satellite imagery. arXiv.","DOI":"10.1109\/IGARSS47720.2021.9554054"},{"key":"ref_32","unstructured":"Shen, Y., Zhu, S., Yang, T., and Chen, C. (2020). Cross-directional Feature Fusion Network for Building Damage Assessment from Satellite Imagery. arXiv."},{"key":"ref_33","unstructured":"Boin, J.B., Roth, N., Doshi, J., Llueca, P., and Borensztein, N. (2020). Multi-class segmentation under severe class imbalance: A case study in roof damage assessment. arXiv."},{"key":"ref_34","unstructured":"Khvedchenya, E., and Gabruseva, T. (2021). Fully convolutional Siamese neural networks for buildings damage assessment from satellite images. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Xiao, H., Peng, Y., Tan, H., and Li, P. (2021, January 5\u20139). Dynamic Cross Fusion Network for Building-Based Damage Assessment. Proceedings of the 2021 IEEE International Conference on Multimedia and Expo (ICME), Shenzhen, China.","DOI":"10.1109\/ICME51207.2021.9428414"},{"key":"ref_36","first-page":"1","article-title":"Bdanet: Multiscale convolutional neural network with cross-directional attention for building damage assessment from satellite images","volume":"60","author":"Shen","year":"2021","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Calton, L., and Wei, Z. (2022). Using Artificial Neural Network Models to Assess Hurricane Damage through Transfer Learning. Appl. Sci., 12.","DOI":"10.3390\/app12031466"},{"key":"ref_38","unstructured":"Xu, J.Z., Lu, W., Li, Z., Khaitan, P., and Zaytseva, V. (2019). Building damage detection in satellite imagery using convolutional neural networks. arXiv."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Valentijn, T., Margutti, J., van den Homberg, M., and Laaksonen, J. (2020). Multi-hazard and spatial transferability of a cnn for automated building damage assessment. Remote Sens., 12.","DOI":"10.3390\/rs12172839"},{"key":"ref_40","unstructured":"Benson, V., and Ecker, A. (2020). Assessing out-of-domain generalization for robust building damage detection. arXiv."},{"key":"ref_41","unstructured":"Li, Y., Wang, N., Shi, J., Liu, J., and Hou, X. (2016). Revisiting batch normalization for practical domain adaptation. arXiv."},{"key":"ref_42","unstructured":"Athiwaratkun, B., Finzi, M., Izmailov, P., and Wilson, A.G. (2018). There are many consistent explanations of unlabeled data: Why you should average. arXiv."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Nex, F., Duarte, D., Tonolo, F.G., and Kerle, N. (2019). Structural building damage detection with deep learning: Assessment of a state-of-the-art CNN in operational conditions. Remote Sens., 11.","DOI":"10.3390\/rs11232765"},{"key":"ref_44","unstructured":"Lee, J., Xu, J.Z., Sohn, K., Lu, W., Berthelot, D., Gur, I., Khaitan, P., Huang, K.-W.(F.), Koupparis, K., and Kowatsch, B. (2020). Assessing Post-Disaster Damage from Satellite Imagery using Semi-Supervised Learning Techniques. arXiv."},{"key":"ref_45","unstructured":"Berthelot, D., Carlini, N., Goodfellow, I., Papernot, N., Oliver, A., and Raffel, C. (2019). Mixmatch: A holistic approach to semi-supervised learning. arXiv."},{"key":"ref_46","unstructured":"Sohn, K., Berthelot, D., Li, C.L., Zhang, Z., Carlini, N., Cubuk, E.D., Kurakin, A., Zhang, H., and Raffel, C. (2020). Fixmatch: Simplifying semi-supervised learning with consistency and confidence. arXiv."},{"key":"ref_47","unstructured":"Xia, J., Yokoya, N., and Adriano, B. (2021). Building Damage Mapping with Self-PositiveUnlabeled Learning. arXiv."},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Ismail, A., and Awad, M. (2022). Towards Cross-Disaster Building Damage Assessment with Graph Convolutional Networks. arXiv.","DOI":"10.1109\/IGARSS46834.2022.9883832"},{"key":"ref_49","unstructured":"Kuzin, D., Isupova, O., Simmons, B.D., and Reece, S. (2021). Disaster mapping from satellites: Damage detection with crowdsourced point labels. arXiv."},{"key":"ref_50","unstructured":"Anand, V., and Miura, Y. (2021). PREDISM: Pre-Disaster Modelling With CNN Ensembles for At-Risk Communities. arXiv."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Presa-Reyes, M., and Chen, S.C. (2021, January 8\u201310). Weakly-Supervised Damaged Building Localization and Assessment with Noise Regularization. Proceedings of the 2021 IEEE 4th International Conference on Multimedia Information Processing and Retrieval (MIPR), Virtual.","DOI":"10.1109\/MIPR51284.2021.00009"},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"101009","DOI":"10.1016\/j.aei.2019.101009","article-title":"Convolutional neural networks for object detection in aerial imagery for disaster response and recovery","volume":"43","author":"Pi","year":"2020","journal-title":"Adv. Eng. Inform."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"102994","DOI":"10.1016\/j.autcon.2019.102994","article-title":"Automated regional seismic damage assessment of buildings using an unmanned aerial vehicle and a convolutional neural network","volume":"109","author":"Xiong","year":"2020","journal-title":"Autom. Constr."},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Rudner, T.G.J., Ru\u00dfwurm, M., Fil, J., Pelich, R., Bischke, B., Kopackov\u00e1, V., and Bilinski, P. (2018, January 3\u20138). Rapid Computer Vision-Aided Disaster Response via Fusion of Multiresolution, Multisensor, and Multitemporal Satellite Imagery. Proceedings of the First Workshop on AI for Social Good. Neural Information Processing Systems (NIPS-2018), Montreal, QC, Canada.","DOI":"10.1609\/aaai.v33i01.3301702"},{"key":"ref_55","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1007\/s13278-019-0588-4","article-title":"Localizing and quantifying infrastructure damage using class activation mapping approaches","volume":"9","author":"Li","year":"2019","journal-title":"Soc. Netw. Anal. Min."},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"89","DOI":"10.5194\/isprs-annals-IV-2-89-2018","article-title":"Satellite image classification of building damages using airborne and satellite image samples in a deep learning approach","volume":"IV-2","author":"Duarte","year":"2018","journal-title":"ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci."},{"key":"ref_57","doi-asserted-by":"crossref","unstructured":"Weber, E., Papadopoulos, D.P., Lapedriza, A., Ofli, F., Imran, M., and Torralba, A. (2022). Incidents1M: A large-scale dataset of images with natural disasters, damage, and incidents. arXiv.","DOI":"10.1109\/TPAMI.2022.3191996"},{"key":"ref_58","unstructured":"Glasmachers, T. (2017, January 15\u201317). Limits of End-to-End Learning. Proceedings of the Asian Conference on Machine Learning, Seoul, Korea."},{"key":"ref_59","unstructured":"Oktay, O., Schlemper, J., Folgoc, L.L., Lee, M.C.H., Heinrich, M.P., Misawa, K., Mori, K., McDonagh, S.G., Hammerla, N.Y., and Kainz, B. (2018). Attention U-Net: Learning Where to Look for the Pancreas. arXiv."},{"key":"ref_60","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_61","unstructured":"Koch, G., Zemel, R., and Salakhutdinov, R. (2015, January 6\u201311). Siamese neural networks for one-shot image recognition. Proceedings of the ICML Deep Learning Workshop, Lille, France."},{"key":"ref_62","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2015). Deep Residual Learning for Image Recognition. arXiv.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_63","doi-asserted-by":"crossref","unstructured":"Chen, H., Nemni, E., Vallecorsa, S., Li, X., Wu, C., and Bromley, L. (2022). Dual-Tasks Siamese Transformer Framework for Building Damage Assessment. arXiv.","DOI":"10.1109\/IGARSS46834.2022.9883139"},{"key":"ref_64","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1016\/j.neucom.2020.07.005","article-title":"Unsupervised Domain Adaptation with Self-attention for Post-disaster Building Damage Detection","volume":"415","author":"Li","year":"2020","journal-title":"Neurocomputing"},{"key":"ref_65","doi-asserted-by":"crossref","unstructured":"Benjdira, B., Bazi, Y., Koubaa, A., and Ouni, K. (2019). Unsupervised Domain Adaptation Using Generative Adversarial Networks for Semantic Segmentation of Aerial Images. Remote Sens., 11.","DOI":"10.3390\/rs11111369"},{"key":"ref_66","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2020.3036452","article-title":"Class-Aware Domain Adaptation for Semantic Segmentation of Remote Sensing Images","volume":"60","author":"Xu","year":"2020","journal-title":"IEEE Trans. Geosci. Remote Sens."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/14\/11\/2532\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T23:18:35Z","timestamp":1760138315000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/14\/11\/2532"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,25]]},"references-count":66,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2022,6]]}},"alternative-id":["rs14112532"],"URL":"https:\/\/doi.org\/10.3390\/rs14112532","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,5,25]]}}}