{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T23:29:55Z","timestamp":1790033395818,"version":"4.0.1"},"reference-count":49,"publisher":"MDPI AG","issue":"22","license":[{"start":{"date-parts":[[2022,11,13]],"date-time":"2022-11-13T00:00:00Z","timestamp":1668297600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000104","name":"NASA Harvest Consortium","doi-asserted-by":"publisher","award":["80NSSC17K0652"],"award-info":[{"award-number":["80NSSC17K0652"]}],"id":[{"id":"10.13039\/100000104","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000104","name":"NASA Harvest Consortium","doi-asserted-by":"publisher","award":["54308-Z6059203"],"award-info":[{"award-number":["54308-Z6059203"]}],"id":[{"id":"10.13039\/100000104","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Stanford\u2019s Institute for Human-Centered Artificial Intelligence","award":["80NSSC17K0652"],"award-info":[{"award-number":["80NSSC17K0652"]}]},{"name":"Stanford\u2019s Institute for Human-Centered Artificial Intelligence","award":["54308-Z6059203"],"award-info":[{"award-number":["54308-Z6059203"]}]},{"name":"Ciriacy-Wantrup Postdoctoral Fellowship at the University of California, Berkeley","award":["80NSSC17K0652"],"award-info":[{"award-number":["80NSSC17K0652"]}]},{"name":"Ciriacy-Wantrup Postdoctoral Fellowship at the University of California, Berkeley","award":["54308-Z6059203"],"award-info":[{"award-number":["54308-Z6059203"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>Crop field boundaries aid in mapping crop types, predicting yields, and delivering field-scale analytics to farmers. Recent years have seen the successful application of deep learning to delineating field boundaries in industrial agricultural systems, but field boundary datasets remain missing in smallholder systems due to (1) small fields that require high resolution satellite imagery to delineate and (2) a lack of ground labels for model training and validation. In this work, we use newly-accessible high-resolution satellite imagery and combine transfer learning with weak supervision to address these challenges in India. Our best model uses 1.5 m resolution Airbus SPOT imagery as input, pre-trains a state-of-the-art neural network on France field boundaries, and fine-tunes on India labels to achieve a median Intersection over Union (mIoU) of 0.85 in India. When we decouple field delineation from cropland classification, a model trained in France and applied as-is to India Airbus SPOT imagery delineates fields with a mIoU of 0.74. If using 4.8 m resolution PlanetScope imagery instead, high average performance (mIoU &gt; 0.8) is only achievable for fields larger than 1 hectare. Experiments also show that pre-training in France reduces the number of India field labels needed to achieve a given performance level by as much as 10\u00d7 when datasets are small. These findings suggest our method is a scalable approach for delineating crop fields in regions of the world that currently lack field boundary datasets. We publicly release 10,000 Indian field boundary labels and our delineation model to facilitate the creation of field boundary maps and new methods by the community.<\/jats:p>","DOI":"10.3390\/rs14225738","type":"journal-article","created":{"date-parts":[[2022,11,14]],"date-time":"2022-11-14T04:24:10Z","timestamp":1668399850000},"page":"5738","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":61,"title":["Unlocking Large-Scale Crop Field Delineation in Smallholder Farming Systems with Transfer Learning and Weak Supervision"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4618-5675","authenticated-orcid":false,"given":"Sherrie","family":"Wang","sequence":"first","affiliation":[{"name":"Goldman School of Public Policy, University of California, Berkeley, 2607 Hearst Ave, Berkeley, CA 94720, USA"},{"name":"Institute for Computational and Mathematical Engineering, Stanford University, 475 Via Ortega, Stanford, CA 94305, USA"},{"name":"Department of Earth System Science, Stanford University, 473 Via Ortega, Stanford, CA 94305, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5599-7456","authenticated-orcid":false,"given":"Fran\u00e7ois","family":"Waldner","sequence":"additional","affiliation":[{"name":"European Commission Joint Research Centre, Via Enrico Fermi 2749, 21027 Ispra, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5969-3476","authenticated-orcid":false,"given":"David B.","family":"Lobell","sequence":"additional","affiliation":[{"name":"Department of Earth System Science, Stanford University, 473 Via Ortega, Stanford, CA 94305, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2022,11,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1980","DOI":"10.1111\/gcb.12838","article-title":"Mapping global cropland and field size","volume":"21","author":"Fritz","year":"2015","journal-title":"Glob. Chang. Biol."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1016\/j.rse.2015.10.034","article-title":"Conterminous United States crop field size quantification from multi-temporal Landsat data","volume":"172","author":"Yan","year":"2016","journal-title":"Remote Sens. Environ."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"4091","DOI":"10.1080\/01431160310001619580","article-title":"Efficiency and accuracy of per-field classification for operational crop mapping","volume":"25","author":"Wit","year":"2004","journal-title":"Int. J. Remote Sens."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1016\/j.rse.2018.02.045","article-title":"A high-performance and in-season classification system of field-level crop types using time-series Landsat data and a machine learning approach","volume":"210","author":"Cai","year":"2018","journal-title":"Remote Sens. Environ."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"303","DOI":"10.1016\/j.rse.2018.12.026","article-title":"Crop type mapping without field-level labels: Random forest transfer and unsupervised clustering techniques","volume":"222","author":"Wang","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"647","DOI":"10.1016\/j.rse.2018.06.036","article-title":"Estimating smallholder crops production at village level from Sentinel-2 time series in Mali\u2019s cotton belt","volume":"216","author":"Lambert","year":"2018","journal-title":"Remote Sens. Environ."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"324","DOI":"10.1016\/j.rse.2015.04.021","article-title":"A scalable satellite-based crop yield mapper","volume":"164","author":"Lobell","year":"2015","journal-title":"Remote Sens. Environ."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"106","DOI":"10.1016\/j.fcr.2018.01.028","article-title":"Predicting spatial patterns of within-field crop yield variability","volume":"219","author":"Maestrini","year":"2018","journal-title":"Field Crops Res."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"144","DOI":"10.1016\/j.rse.2019.04.005","article-title":"Field-level crop yield mapping with Landsat using a hierarchical data assimilation approach","volume":"228","author":"Kang","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1016\/j.fcr.2018.08.005","article-title":"Towards a national, remote-sensing-based model for predicting field-scale crop yield","volume":"227","author":"Donohue","year":"2018","journal-title":"Field Crops Res."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"107886","DOI":"10.1016\/j.agrformet.2019.107886","article-title":"Satellite-based soybean yield forecast: Integrating machine learning and weather data for improving crop yield prediction in southern Brazil","volume":"284","author":"Schwalbert","year":"2020","journal-title":"Agric. For. Meteorol."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Al-Gaadi, K.A., Hassaballa, A.A., Tola, E., Kayad, A.G., Madugundu, R., Alblewi, B., and Assiri, F. (2016). Prediction of Potato Crop Yield Using Precision Agriculture Techniques. PLoS ONE, 11.","DOI":"10.1371\/journal.pone.0162219"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1007\/s11119-018-9589-y","article-title":"Farmer attitudes to the use of sensors and automation in fertilizer decision-making: Nitrogen fertilization in the Australian grains sector","volume":"20","author":"Bramley","year":"2019","journal-title":"Precis. Agric."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"131","DOI":"10.1093\/oxfordjournals.oep.a041621","article-title":"Identification of the Inverse Relationship between Farm Size and Productivity: An Empirical Analysis of Peasant Agricultural Production","volume":"36","author":"Carter","year":"1984","journal-title":"Oxf. Econ. Pap."},{"key":"ref_15","first-page":"5","article-title":"Farm Size and Productivity: Understanding the Strengths of Smallholders and Improving Their Livelihoods","volume":"46","author":"Chand","year":"2011","journal-title":"Econ. Political Wkly."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1016\/j.foodpol.2018.03.015","article-title":"New perspectives on farm size and productivity","volume":"84","author":"Rada","year":"2019","journal-title":"Food Policy"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1016\/j.agee.2012.01.010","article-title":"Should increasing the field size of monocultural crops be expected to exacerbate pest damage?","volume":"150","author":"Segoli","year":"2012","journal-title":"Agric. Ecosyst. Environ."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1080\/07060660509507219","article-title":"Effect of previous crop, tillage, field size, adjacent crop, and sampling direction on airborne propagules of Gibberella zeae\/Fusarium graminearum, fusarium head blight severity, and deoxynivalenol accumulation in winter wheat","volume":"27","author":"Schaafsma","year":"2005","journal-title":"Can. J. Plant Pathol."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"219","DOI":"10.1016\/j.agee.2014.11.018","article-title":"Farmlands with smaller crop fields have higher within-field biodiversity","volume":"200","author":"Fahrig","year":"2015","journal-title":"Agric. Ecosyst. Environ."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1016\/j.ecolind.2018.03.001","article-title":"Bringing diversity back to agriculture: Smaller fields and non-crop elements enhance biodiversity in intensively managed arable farmlands","volume":"90","author":"Hula","year":"2018","journal-title":"Ecol. Indic."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"111253","DOI":"10.1016\/j.rse.2019.111253","article-title":"Delineation of agricultural fields in smallholder farms from satellite images using fully convolutional networks and combinatorial grouping","volume":"231","author":"Persello","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_22","unstructured":"Agence de Services et de Paiement (2021, July 01). Registre Parcellaire Graphique (RPG): Contours des Parcelles et \u00eelots Culturaux et Leur Groupe de Cultures Majoritaire, Available online: https:\/\/www.data.gouv.fr\/en\/datasets\/registre-parcellaire-graphique-rpg-contours-des-parcelles-et-ilots-culturaux-et-leur-groupe-de-cultures-majoritaire\/."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"111741","DOI":"10.1016\/j.rse.2020.111741","article-title":"Deep learning on edge: Extracting field boundaries from satellite images with a convolutional neural network","volume":"245","author":"Waldner","year":"2020","journal-title":"Remote Sens. Environ."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Vlachopoulos, O., Leblon, B., Wang, J., Haddadi, A., LaRocque, A., and Patterson, G. (2020). Delineation of Crop Field Areas and Boundaries from UAS Imagery Using PBIA and GEOBIA with Random Forest Classification. Remote Sens., 12.","DOI":"10.3390\/rs12162640"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"237","DOI":"10.1109\/JSTARS.2018.2884513","article-title":"Boundary Delineation of Agricultural Fields in Multitemporal Satellite Imagery","volume":"12","author":"North","year":"2019","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1016\/j.rse.2014.01.006","article-title":"Automated crop field extraction from multi-temporal Web Enabled Landsat Data","volume":"144","author":"Yan","year":"2014","journal-title":"Remote Sens. Environ."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Rahman, M.S., Di, L., Yu, Z., Yu, E.G., Tang, J., Lin, L., Zhang, C., and Gaigalas, J. (2019, January 16\u201319). Crop Field Boundary Delineation using Historical Crop Rotation Pattern. Proceedings of the 2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics), Istanbul, Turkey.","DOI":"10.1109\/Agro-Geoinformatics.2019.8820240"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Aung, H.L., Uzkent, B., Burke, M., Lobell, D., and Ermon, S. (2020, January 14\u201319). Farm Parcel Delineation Using Spatio-Temporal Convolutional Networks. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, Seattle, WA, USA.","DOI":"10.1109\/CVPRW50498.2020.00046"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Masoud, K.M., Persello, C., and Tolpekin, V.A. (2020). Delineation of Agricultural Field Boundaries from Sentinel-2 Images Using a Novel Super-Resolution Contour Detector Based on Fully Convolutional Networks. Remote Sens., 12.","DOI":"10.3390\/rs12010059"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"2514","DOI":"10.1109\/36.964989","article-title":"Integrated method for boundary delineation of agricultural fields in multispectral satellite images","volume":"39","author":"Rydberg","year":"2001","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Waldner, F., Diakogiannis, F.I., Batchelor, K., Ciccotosto-Camp, M., Cooper-Williams, E., Herrmann, C., Mata, G., and Toovey, A. (2021). Detect, Consolidate, Delineate: Scalable Mapping of Field Boundaries Using Satellite Images. Remote Sens., 13.","DOI":"10.3390\/rs13112197"},{"key":"ref_32","first-page":"102557","article-title":"Automated delineation of agricultural field boundaries from Sentinel-2 images using recurrent residual U-Net","volume":"105","author":"Zhang","year":"2021","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"744863","DOI":"10.3389\/frai.2021.744863","article-title":"High Resolution, Annual Maps of Field Boundaries for Smallholder-Dominated Croplands at National Scales","volume":"4","author":"Estes","year":"2022","journal-title":"Front. Artif. Intell."},{"key":"ref_34","unstructured":"O\u2019Shea, T. (2022, July 01). Universal access to satellite monitoring paves the way to protect the world\u2019s tropical forests. Available online: https:\/\/www.planet.com\/pulse\/universal-access-to-satellite-monitoring-paves-the-way-to-protect-the-worlds-tropical-forests."},{"key":"ref_35","unstructured":"Airbus (2022, January 03). Airbus OneAtlas Basemap. Available online: https:\/\/oneatlas.airbus.com\/service\/basemap."},{"key":"ref_36","unstructured":"Descartes Labs (2022, November 01). Airbus OneAtlas SPOT V2. Available online: https:\/\/descarteslabs.com\/datasources\/."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"170136","DOI":"10.1038\/sdata.2017.136","article-title":"A global reference database of crowdsourced cropland data collected using the Geo-Wiki platform","volume":"4","author":"Lesiv","year":"2017","journal-title":"Sci. Data"},{"key":"ref_38","unstructured":"Planet Labs, Inc. (2022, July 01). Visual Basemaps. Available online: https:\/\/developers.planet.com\/docs\/data\/visual-basemaps\/."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"94","DOI":"10.1016\/j.isprsjprs.2020.01.013","article-title":"ResUNet-a: A deep learning framework for semantic segmentation of remotely sensed data","volume":"162","author":"Diakogiannis","year":"2020","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Diakogiannis, F.I., Waldner, F., and Caccetta, P. (2021). Looking for Change? Roll the Dice and Demand Attention. Remote Sens., 13.","DOI":"10.3390\/rs13183707"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Navab, N., Hornegger, J., Wells, W.M., and Frangi, A.F. U-Net: Convolutional Networks for Biomedical Image Segmentation. Proceedings of the Medical Image Computing and Computer-Assisted Intervention\u2014MICCAI 2015, Available online: https:\/\/link.springer.com\/chapter\/10.1007\/978-3-319-24574-4_28.","DOI":"10.1007\/978-3-319-24571-3"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1093\/nsr\/nwx106","article-title":"A brief introduction to weakly supervised learning","volume":"5","author":"Zhou","year":"2017","journal-title":"Natl. Sci. Rev."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"207","DOI":"10.3390\/rs12020207","article-title":"Weakly Supervised Deep Learning for Segmentation of Remote Sensing Imagery","volume":"12","author":"Wang","year":"2020","journal-title":"Remote Sens."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Buchhorn, M., Lesiv, M., Tsendbazar, N.E., Herold, M., Bertels, L., and Smets, B. (2020). Copernicus Global Land Cover Layers\u2014Collection 2. Remote Sens., 12.","DOI":"10.3390\/rs12061044"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"294","DOI":"10.1016\/j.compag.2019.02.009","article-title":"A comparison of object-based image analysis approaches for field boundary delineation using multi-temporal Sentinel-2 imagery","volume":"158","author":"Watkins","year":"2019","journal-title":"Comput. Electron. Agric."},{"key":"ref_46","first-page":"135","article-title":"Playing with Kruskal: Algorithms for morphological trees in edge-weighted graphs","volume":"Volume 7883","author":"Hendriks","year":"2013","journal-title":"Proceedings of the International Symposium on Mathematical Morphology"},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Chicco, D., and Jurman, G. (2020). The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genom., 21.","DOI":"10.1186\/s12864-019-6413-7"},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Rezatofighi, H., Tsoi, N., Gwak, J., Sadeghian, A., Reid, I., and Savarese, S. (2019, January 16\u201320). Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00075"},{"key":"ref_49","unstructured":"Zanaga, D., Van De Kerchove, R., De Keersmaecker, W., Souverijns, N., Brockmann, C., Quast, R., Wevers, J., Grosu, A., Paccini, A., and Vergnaud, S. (2020, July 01). ESA WorldCover 10m 2020 v100. Available online: https:\/\/zenodo.org\/record\/5571936#.Y3w5nvdByUk."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/14\/22\/5738\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T01:17:20Z","timestamp":1760145440000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/14\/22\/5738"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,13]]},"references-count":49,"journal-issue":{"issue":"22","published-online":{"date-parts":[[2022,11]]}},"alternative-id":["rs14225738"],"URL":"https:\/\/doi.org\/10.3390\/rs14225738","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,11,13]]}}}