{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T18:45:50Z","timestamp":1776278750586,"version":"3.50.1"},"reference-count":51,"publisher":"MDPI AG","issue":"21","license":[{"start":{"date-parts":[[2023,10,27]],"date-time":"2023-10-27T00:00:00Z","timestamp":1698364800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"European Space Agency","award":["4000135015\/21\/I-NB"],"award-info":[{"award-number":["4000135015\/21\/I-NB"]}]},{"name":"European Space Agency","award":["62001229"],"award-info":[{"award-number":["62001229"]}]},{"name":"European Space Agency","award":["62101264"],"award-info":[{"award-number":["62101264"]}]},{"name":"European Space Agency","award":["62101260"],"award-info":[{"award-number":["62101260"]}]},{"name":"National Natural Science Foundation of China","award":["4000135015\/21\/I-NB"],"award-info":[{"award-number":["4000135015\/21\/I-NB"]}]},{"name":"National Natural Science Foundation of China","award":["62001229"],"award-info":[{"award-number":["62001229"]}]},{"name":"National Natural Science Foundation of China","award":["62101264"],"award-info":[{"award-number":["62101264"]}]},{"name":"National Natural Science Foundation of China","award":["62101260"],"award-info":[{"award-number":["62101260"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>Deep learning (DL) models are gaining popularity in forest variable prediction using Earth observation (EO) images. However, in practical forest inventories, reference datasets are often represented by plot- or stand-level measurements, while high-quality representative wall-to-wall reference data for end-to-end training of DL models are rarely available. Transfer learning facilitates expansion of the use of deep learning models into areas with sub-optimal training data by allowing pretraining of the model in areas where high-quality teaching data are available. In this study, we perform a \u201cmodel transfer\u201d (or domain adaptation) of a pretrained DL model into a target area using plot-level measurements and compare performance versus other machine learning models. We use an earlier developed UNet based model (SeUNet) to demonstrate the approach on two distinct taiga sites with varying forest structure and composition. The examined SeUNet model uses multi-source EO data to predict forest height. Here, EO data are represented by a combination of Copernicus Sentinel-1 C-band SAR and Sentinel-2 multispectral images, ALOS-2 PALSAR-2 SAR mosaics and TanDEM-X bistatic interferometric radar data. The training study site is located in Finnish Lapland, while the target site is located in Southern Finland. By leveraging transfer learning, the SeUNet prediction achieved root mean squared error (RMSE) of 2.70 m and R2 of 0.882, considerably more accurate than traditional benchmark methods. We expect such forest-specific DL model transfer can be suitable also for other forest variables and other EO data sources that are sensitive to forest structure.<\/jats:p>","DOI":"10.3390\/rs15215152","type":"journal-article","created":{"date-parts":[[2023,10,27]],"date-time":"2023-10-27T11:57:56Z","timestamp":1698407876000},"page":"5152","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["Deep Learning Model Transfer in Forest Mapping Using Multi-Source Satellite SAR and Optical Images"],"prefix":"10.3390","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9655-1207","authenticated-orcid":false,"given":"Shaojia","family":"Ge","sequence":"first","affiliation":[{"name":"School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8576-404X","authenticated-orcid":false,"given":"Oleg","family":"Antropov","sequence":"additional","affiliation":[{"name":"VTT Technical Research Centre of Finland, 00076 Espoo, Finland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1446-010X","authenticated-orcid":false,"given":"Tuomas","family":"H\u00e4me","sequence":"additional","affiliation":[{"name":"VTT Technical Research Centre of Finland, 00076 Espoo, Finland"}]},{"given":"Ronald E.","family":"McRoberts","sequence":"additional","affiliation":[{"name":"Department of Forest Resources, University of Minnesota, St. Paul, MN 55108, USA"}]},{"given":"Jukka","family":"Miettinen","sequence":"additional","affiliation":[{"name":"VTT Technical Research Centre of Finland, 00076 Espoo, Finland"}]}],"member":"1968","published-online":{"date-parts":[[2023,10,27]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"757","DOI":"10.1007\/s10712-019-09510-6","article-title":"The role and need for space-based forest biomass-related measurements in environmental management and policy","volume":"40","author":"Herold","year":"2019","journal-title":"Surv. Geophys."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"412","DOI":"10.1016\/j.rse.2006.09.034","article-title":"Remote sensing support for national forest inventories","volume":"110","author":"McRoberts","year":"2007","journal-title":"Remote Sens. Environ."},{"key":"ref_3","unstructured":"GFOI (2014). Integrating Remote-Sensing and Ground-Based Observations for Estimation of Emissions and Removals of Greenhouse Gases in Forests: Methods and Guidance from the Global Forest Observations Initiative, Group on Earth Observations."},{"key":"ref_4","first-page":"53","article-title":"Forest biomass retrieval approaches from Earth observation in different biomes","volume":"77","author":"Quegan","year":"2019","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1016\/j.rse.2019.01.019","article-title":"Comparison of Sentinel-2 and Landsat 8 imagery for forest variable prediction in boreal region","volume":"223","author":"Astola","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Kuenzer, C., Dech, S., and Wagner, W. (2015). Remote Sensing Time Series: Revealing Land Surface Dynamics, Springer International Publishing.","DOI":"10.1007\/978-3-319-15967-6"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Antropov, O., Rauste, Y., H\u00e4me, T., and Praks, J. (2017). Polarimetric ALOS PALSAR time series in mapping biomass of boreal forests. Remote Sens., 9.","DOI":"10.3390\/rs9100999"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Olesk, A., Praks, J., Antropov, O., Zalite, K., Arum\u00e4e, T., and Voormansik, K. (2016). Interferometric SAR coherence models for characterization of hemiboreal forests using TanDEM-X data. Remote Sens., 8.","DOI":"10.3390\/rs8090700"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"5294","DOI":"10.1109\/TGRS.2015.2420996","article-title":"Forest height estimation by means of Pol-InSAR data inversion: The role of the vertical wavenumber","volume":"53","author":"Kugler","year":"2015","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1109\/MGRS.2021.3136100","article-title":"Deep learning and Earth Observation to support the sustainable development goals: Current approaches, open challenges, and future opportunities","volume":"10","author":"Persello","year":"2022","journal-title":"IEEE Geosci. Remote Sens. Mag."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1109\/MGRS.2020.3046356","article-title":"Deep learning meets SAR: Concepts, models, pitfalls, and perspectives","volume":"9","author":"Zhu","year":"2021","journal-title":"IEEE Geosci. Remote Sens. Mag."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1109\/MGRS.2017.2762307","article-title":"Deep learning in remote sensing: A comprehensive review and list of resources","volume":"5","author":"Zhu","year":"2017","journal-title":"IEEE Geosci. Remote Sens. Mag."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Astola, H., Seitsonen, L., Halme, E., Molinier, M., and L\u00f6nnqvist, A. (2021). Deep neural networks with transfer learning for forest variable estimation using Sentinel-2 imagery in boreal forest. Remote Sens., 13.","DOI":"10.3390\/rs13122392"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Illarionova, S., Trekin, A., Ignatiev, V., and Oseledets, I. (2021). Tree species mapping on Sentinel-2 satellite imagery with weakly supervised classification and object-wise sampling. Forests, 12.","DOI":"10.3390\/f12101413"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"5776","DOI":"10.1109\/JSTARS.2022.3188201","article-title":"Improved semisupervised UNet deep learning model for forest height mapping with satellite SAR and optical Data","volume":"15","author":"Ge","year":"2022","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Ge, S., Su, W., Gu, H., Rauste, Y., Praks, J., and Antropov, O. (2022). Improved LSTM model for boreal forest height mapping using Sentinel-1 time series. Remote Sens., 14.","DOI":"10.1101\/2022.09.18.508417"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"113205","DOI":"10.1016\/j.rse.2022.113205","article-title":"Mapping tree species proportions from satellite imagery using spectral\u2013spatial deep learning","volume":"280","author":"Bolyn","year":"2022","journal-title":"Remote Sens. Environ."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Wang, S., Chen, W., Xie, S.M., Azzari, G., and Lobell, D.B. (2020). Weakly supervised deep learning for segmentation of remote sensing imagery. Remote Sens., 12.","DOI":"10.3390\/rs12020207"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"34116","DOI":"10.1109\/ACCESS.2022.3161568","article-title":"Estimation of the canopy height model from multispectral satellite imagery with convolutional neural networks","volume":"10","author":"Illarionova","year":"2022","journal-title":"IEEE Access"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"111347","DOI":"10.1016\/j.rse.2019.111347","article-title":"Country-wide high-resolution vegetation height mapping with Sentinel-2","volume":"233","author":"Lang","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Bueso-Bello, J.-L., Carcereri, D., Martone, M., Gonz\u00e1lez, C., Posovszky, P., and Rizzoli, P. (2022). Deep learning for mapping tropical forests with TanDEM-X bistatic InSAR data. Remote Sens., 14.","DOI":"10.3390\/rs14163981"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"4612","DOI":"10.1109\/JSTARS.2022.3179819","article-title":"On the potential of sequential and nonsequential regression models for Sentinel-1-based biomass prediction in Tanzanian Miombo forests","volume":"15","author":"Bjork","year":"2022","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_23","first-page":"103358","article-title":"Cross-city Landuse classification of remote sensing images via deep transfer learning","volume":"122","author":"Zhao","year":"2023","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"109402","DOI":"10.1016\/j.patcog.2023.109402","article-title":"Cross-task and cross-domain SAR target recognition: A meta-transfer learning approach","volume":"138","author":"Zhang","year":"2023","journal-title":"Pattern Recognit."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Javed, A., Kim, T., Lee, C., Oh, J., and Han, Y. (2023). Deep learning-based detection of urban forest cover change along with overall urban changes using very-high-resolution satellite images. Remote Sens., 15.","DOI":"10.3390\/rs15174285"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"110362","DOI":"10.1016\/j.asoc.2023.110362","article-title":"Detection of forest fire using deep convolutional neural networks with transfer learning approach","volume":"143","author":"Reis","year":"2023","journal-title":"Appl. Soft Comput."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1109\/JPROC.2020.3004555","article-title":"A comprehensive survey on transfer learning","volume":"109","author":"Zhuang","year":"2020","journal-title":"Proc. IEEE"},{"key":"ref_28","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015, January 5\u20139). U-net: Convolutional networks for biomedical image segmentation. Proceedings of the Medical Image Computing and Computer-Assisted Intervention\u2013MICCAI 2015: 18th International Conference, Munich, Germany. Proceedings, Part III 18."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"10357","DOI":"10.1109\/JSTARS.2021.3116094","article-title":"Wide-area land cover mapping with Sentinel-1 imagery using deep learning semantic segmentation models","volume":"14","author":"Antropov","year":"2021","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"100093","DOI":"10.1016\/j.srs.2023.100093","article-title":"High-resolution mapping of forest structure from integrated SAR and optical images using an enhanced U-net method","volume":"8","author":"Gazzea","year":"2023","journal-title":"Sci. Remote Sens."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Huang, Z., Pan, Z., and Lei, B. (2017). Transfer learning with deep convolutional neural network for SAR target classification with limited labeled data. Remote Sens., 9.","DOI":"10.3390\/rs9090907"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1016\/j.isprsjprs.2019.02.006","article-title":"Semantic segmentation of slums in satellite images using transfer learning on fully convolutional neural networks","volume":"150","author":"Wurm","year":"2019","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"298","DOI":"10.1109\/JSTARS.2011.2176720","article-title":"Modeling aboveground biomass in tropical forests using multi-frequency SAR data\u2014A comparison of methods","volume":"5","author":"Englhart","year":"2012","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Ge, S., Tomppo, E., Rauste, Y., McRoberts, R.E., Praks, J., Gu, H., Su, W., and Antropov, O. (2023). Sentinel-1 time series for predicting growing stock volume of boreal forest: Multitemporal analysis and feature selection. Remote Sens., 15.","DOI":"10.3390\/rs15143489"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Antropov, O., Miettinen, J., H\u00e4me, T., Yrj\u00f6, R., Seitsonen, L., McRoberts, R.E., Santoro, M., Cartus, O., Duran, N.M., and Herold, M. (2022, January 17\u201322). Intercomparison of Earth Observation data and methods for forest mapping in the context of forest carbon monitoring. Proceedings of the IGARSS 2022\u20142022 IEEE International Geoscience and Remote Sensing Symposium, Kuala Lumpur, Malaysia.","DOI":"10.1109\/IGARSS46834.2022.9884618"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Berninger, A., Lohberger, S., St\u00e4ngel, M., and Siegert, F. (2018). SAR-based estimation of above-ground biomass and its changes in tropical forests of Kalimantan using L- and C-band. Remote Sens., 10.","DOI":"10.3390\/rs10060831"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1982","DOI":"10.1016\/j.rse.2007.03.032","article-title":"Combining national forest inventory field plots and remote sensing data for forest databases","volume":"112","author":"Tomppo","year":"2008","journal-title":"Remote Sens. Environ."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"9467","DOI":"10.1080\/01431161.2021.1998715","article-title":"Demonstration of large area forest volume and primary production estimation approach based on Sentinel-2 imagery and process based ecosystem modelling","volume":"42","author":"Miettinen","year":"2021","journal-title":"Int. J. Remote Sens."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Rauste, Y., Lonnqvist, A., Molinier, M., Henry, J.B., and Hame, T. (2007, January 23\u201328). Ortho-rectification and terrain correction of polarimetric SAR data applied in the ALOS\/Palsar context. Proceedings of the 2007 IEEE International Geoscience and Remote Sensing Symposium, Barcelona, Spain.","DOI":"10.1109\/IGARSS.2007.4423123"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"3081","DOI":"10.1109\/TGRS.2011.2120616","article-title":"Flattening gamma: Radiometric terrain correction for SAR imagery","volume":"49","author":"Small","year":"2011","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"100006","DOI":"10.1016\/j.fecs.2022.100006","article-title":"Forest height mapping using inventory and multi-source satellite data over Hunan Province in southern China","volume":"9","author":"Huang","year":"2022","journal-title":"For. Ecosyst."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Luo, Y., Qi, S., Liao, K., Zhang, S., Hu, B., and Tian, Y. (2023). Mapping the forest height by fusion of ICESat-2 and multi-source remote sensing imagery and topographic information: A case study in Jiangxi province, China. Forests, 14.","DOI":"10.3390\/f14030454"},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Zhang, N., Chen, M., Yang, F., Yang, C., Yang, P., Gao, Y., Shang, Y., and Peng, D. (2022). Forest height mapping using feature selection and machine learning by integrating multi-source satellite data in Baoding city, North China. Remote Sens., 14.","DOI":"10.3390\/rs14184434"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Praks, J., Hallikainen, M., Antropov, O., and Molina, D. (2012, January 22\u201327). Boreal forest tree height estimation from interferometric TanDEM-X images. Proceedings of the 2012 IEEE International Geoscience and Remote Sensing Symposium, Munich, Germany.","DOI":"10.1109\/IGARSS.2012.6351309"},{"key":"ref_45","first-page":"102163","article-title":"High-resolution mapping of forest canopy height using machine learning by coupling ICESat-2 LiDAR with Sentinel-1, Sentinel-2 and Landsat-8 data","volume":"92","author":"Li","year":"2020","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"269","DOI":"10.1016\/j.isprsjprs.2022.11.011","article-title":"Country-wide retrieval of forest structure from optical and SAR satellite imagery with deep ensembles","volume":"195","author":"Becker","year":"2023","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"3177","DOI":"10.1109\/JSTARS.2016.2582722","article-title":"Forest canopy height estimation using Tandem-X coherence data","volume":"9","author":"Chen","year":"2016","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"5565","DOI":"10.1109\/JSTARS.2015.2501648","article-title":"Seasonal differences in forest height estimation from interferometric TanDEM-X coherence data","volume":"8","author":"Olesk","year":"2015","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_49","doi-asserted-by":"crossref","unstructured":"Praks, J., Antropov, O., Olesk, A., and Voormansik, K. (2018, January 22\u201327). Forest height estimation from TanDEM-X images with semi-empirical coherence models. Proceedings of the IGARSS 2018\u20142018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain.","DOI":"10.1109\/IGARSS.2018.8519569"},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"3443","DOI":"10.1109\/JSTARS.2018.2866059","article-title":"Radar forest height estimation in mountainous terrain using Tandem-X coherence data","volume":"11","author":"Chen","year":"2018","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_51","first-page":"101904","article-title":"Canopy height estimation with TanDEM-X in temperate and boreal forests","volume":"82","author":"Schlund","year":"2019","journal-title":"Int. J. Appl. Earth Obs. Geoinf."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/15\/21\/5152\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T21:13:12Z","timestamp":1760130792000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/15\/21\/5152"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,10,27]]},"references-count":51,"journal-issue":{"issue":"21","published-online":{"date-parts":[[2023,11]]}},"alternative-id":["rs15215152"],"URL":"https:\/\/doi.org\/10.3390\/rs15215152","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,10,27]]}}}