{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T14:28:30Z","timestamp":1775744910107,"version":"3.50.1"},"reference-count":36,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2019,6,8]],"date-time":"2019-06-08T00:00:00Z","timestamp":1559952000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>The reemergence of Deep Neural Networks (DNNs) has lead to high-performance supervised learning algorithms for the Electro-Optical (EO) domain classification and detection problems. This success is because generating huge labeled datasets has become possible using modern crowdsourcing labeling platforms such as Amazon\u2019s Mechanical Turk that recruit ordinary people to label data. Unlike the EO domain, labeling the Synthetic Aperture Radar (SAR) domain data can be much more challenging, and for various reasons, using crowdsourcing platforms is not feasible for labeling the SAR domain data. As a result, training deep networks using supervised learning is more challenging in the SAR domain. In the paper, we present a new framework to train a deep neural network for classifying Synthetic Aperture Radar (SAR) images by eliminating the need for a huge labeled dataset. Our idea is based on transferring knowledge from a related EO domain problem, where labeled data are easy to obtain. We transfer knowledge from the EO domain through learning a shared invariant cross-domain embedding space that is also discriminative for classification. To this end, we train two deep encoders that are coupled through their last year to map data points from the EO and the SAR domains to the shared embedding space such that the distance between the distributions of the two domains is minimized in the latent embedding space. We use the Sliced Wasserstein Distance (SWD) to measure and minimize the distance between these two distributions and use a limited number of SAR label data points to match the distributions class-conditionally. As a result of this training procedure, a classifier trained from the embedding space to the label space using mostly the EO data would generalize well on the SAR domain. We provide a theoretical analysis to demonstrate why our approach is effective and validate our algorithm on the problem of ship classification in the SAR domain by comparing against several other competing learning approaches.<\/jats:p>","DOI":"10.3390\/rs11111374","type":"journal-article","created":{"date-parts":[[2019,6,10]],"date-time":"2019-06-10T03:16:51Z","timestamp":1560136611000},"page":"1374","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":142,"title":["Deep Transfer Learning for Few-Shot SAR Image Classification"],"prefix":"10.3390","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7873-1004","authenticated-orcid":false,"given":"Mohammad","family":"Rostami","sequence":"first","affiliation":[{"name":"HRL Laboratories, Malibu, CA 90265-4797, USA"},{"name":"Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA 19104, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8495-5362","authenticated-orcid":false,"given":"Soheil","family":"Kolouri","sequence":"additional","affiliation":[{"name":"HRL Laboratories, Malibu, CA 90265-4797, USA"}]},{"given":"Eric","family":"Eaton","sequence":"additional","affiliation":[{"name":"Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA 19104, USA"}]},{"given":"Kyungnam","family":"Kim","sequence":"additional","affiliation":[{"name":"HRL Laboratories, Malibu, CA 90265-4797, USA"}]}],"member":"1968","published-online":{"date-parts":[[2019,6,8]]},"reference":[{"key":"ref_1","unstructured":"Rostami, M., Kolouri, S., Eaton, E., and Kim, K. (2019, January 16\u201320). Explaining distributed neural activations via unsupervised learning. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA, USA."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"245","DOI":"10.2528\/PIER11092604","article-title":"A new unmanned aerial vehicle synthetic aperture radar for environmental monitoring","volume":"122","author":"Koo","year":"2012","journal-title":"Prog. Electromagn. Res."},{"key":"ref_3","unstructured":"Maitre, H. (2010). Processing of Synthetic Aperture Radar (SAR) Images, Wiley."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Li, F.-F. (2009, January 20\u201325). Imagenet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Rostami, M., Huber, D., and Lu, T.C. (2018, January 2\u20137). A crowdsourcing triage algorithm for geopolitical event forecasting. Proceedings of the 12th ACM Conference on Recommender Systems, Vancouver, BC, Canada.","DOI":"10.1145\/3240323.3240385"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1484","DOI":"10.1109\/LGRS.2017.2717486","article-title":"Improving SAR automatic target recognition models with transfer learning from simulated data","volume":"14","author":"Kusk","year":"2017","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Schwegmann, C., Kleynhans, W., Salmon, B., Mdakane, L., and Meyer, R. (2016, January 10\u201315). Very deep learning for ship discrimination in synthetic aperture radar imagery. Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Beijing, China.","DOI":"10.1109\/IGARSS.2016.7729017"},{"key":"ref_8","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_9","doi-asserted-by":"crossref","first-page":"4806","DOI":"10.1109\/TGRS.2016.2551720","article-title":"Target classification using the deep convolutional networks for SAR images","volume":"54","author":"Chen","year":"2016","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"2834","DOI":"10.1109\/JSTARS.2018.2836909","article-title":"SAR Targets Classification Based on Deep Memory Convolution Neural Networks and Transfer Parameters","volume":"11","author":"Shang","year":"2018","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"072001","DOI":"10.1088\/1757-899X\/322\/7\/072001","article-title":"Transfer Learning with Convolutional Neural Networks for SAR Ship Recognition","volume":"322","author":"Zhang","year":"2018","journal-title":"IOP Conf. Ser. Mater. Sci. Eng."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"150","DOI":"10.1109\/LGRS.2018.2867242","article-title":"SAR Target Detection Based on SSD With Data Augmentation and Transfer Learning","volume":"16","author":"Wang","year":"2019","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"439","DOI":"10.1109\/LGRS.2018.2792683","article-title":"Ship classification in SAR images improved by AIS knowledge transfer","volume":"15","author":"Lang","year":"2018","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_15","unstructured":"Motiian, S., Jones, Q., Iranmanesh, S., and Doretto, G. (2017). Few-shot adversarial domain adaptation. Adv. Neural Inf. Process. Syst., 6670\u20136680."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Redko, I., Habrard, A., and Sebban, M. (2017). Theoretical analysis of domain adaptation with optimal transport. Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Springer.","DOI":"10.1007\/978-3-319-71246-8_45"},{"key":"ref_17","unstructured":"Villani, C. (2008). Optimal Transport: Old and New, Springer Science & Business Media."},{"key":"ref_18","unstructured":"Gretton, A., Smola, A., Huang, J., Schmittfull, M., Borgwardt, K., and Sch\u00f6lkopf, B. (2009). Covariate shift by kernel mean matching. Dataset Shift in Machine Learning, The MIT Press."},{"key":"ref_19","unstructured":"Rabin, J., Peyr\u00e9, G., Delon, J., and Bernot, M. (2011). Wasserstein barycenter and its application to texture mixing. International Conference on Scale Space and Variational Methods in Computer Vision, Springer."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Kolouri, S., Rohde, G.K., and Hoffman, H. (2018, January 18\u201322). Sliced Wasserstein Distance for Learning Gaussian Mixture Models. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.","DOI":"10.1109\/CVPR.2018.00361"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Kodirov, E., Xiang, T., Fu, Z., and Gong, S. (2015, January 7\u201313). Unsupervised domain adaptation for zero-shot learning. Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile.","DOI":"10.1109\/ICCV.2015.282"},{"key":"ref_22","unstructured":"Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2014, January 8\u201313). Generative adversarial nets. Proceedings of the Advances in Neural Information Processing Systems 27 (NIPS 2014), Montreal, QC, Canada."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"1853","DOI":"10.1109\/TPAMI.2016.2615921","article-title":"Optimal transport for domain adaptation","volume":"39","author":"Courty","year":"2017","journal-title":"IEEE TPAMI"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1613\/jair.1872","article-title":"Domain adaptation for statistical classifiers","volume":"26","author":"Daume","year":"2006","journal-title":"J. Artif. Intell. Res."},{"key":"ref_25","unstructured":"Kolouri, S., Pope, P.E., Martin, C.E., and Rohde, G.K. (2019, January 6\u20139). Sliced-Wasserstein Auto-Encoders. Proceedings of the International Conference on Learning Representation (ICLR), New Orleans, LA, USA."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1007\/s10851-014-0506-3","article-title":"Sliced and Radon Wasserstein barycenters of measures","volume":"51","author":"Bonneel","year":"2015","journal-title":"J. Math. Imaging Vis."},{"key":"ref_27","unstructured":"Carriere, M., Cuturi, M., and Oudot, S. (2017). Sliced wasserstein kernel for persistence diagrams. arXiv."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Long, M., Wang, J., Ding, G., Sun, J., and Yu, P.S. (2014, January 23\u201328). Transfer joint matching for unsupervised domain adaptation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Portland, OR, USA.","DOI":"10.1109\/CVPR.2014.183"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Damodaran, B., Kellenberger, B., Flamary, R., Tuia, D., and Courty, N. (2018). DeepJDOT: Deep Joint distribution optimal transport for unsupervised domain adaptation. arXiv.","DOI":"10.1007\/978-3-030-01225-0_28"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1109\/MSP.2017.2695801","article-title":"Optimal Mass Transport: Signal processing and machine-learning applications","volume":"34","author":"Kolouri","year":"2017","journal-title":"IEEE Signal Process. Mag."},{"key":"ref_31","unstructured":"Bonnotte, N. (2013). Unidimensional and Evolution Methods for Optimal Transportation. [Ph.D. Thesis, Scuola Normale Superiore]."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Santambrogio, F. (2015). Optimal Transport for Applied Mathematicians, Birk\u00e4user.","DOI":"10.1007\/978-3-319-20828-2"},{"key":"ref_33","unstructured":"Schwegmann, C.P., Kleynhans, W., Salmon, B.P., Mdakane, L.W., and Meyer, R.G.V. (2019, February 01). A SAR Ship Dataset for Detection, Discrimination and Analysis. Available online: https:\/\/ieee-dataport.org\/documents\/sar-ship-dataset-detection-discrimination-and-analysis."},{"key":"ref_34","unstructured":"Hammell, R. (2019, February 01). Ships in Satellite Imagery. Available online: https:\/\/www.kaggle.com\/rhammell\/ships-in-satellite-imagery."},{"key":"ref_35","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv."},{"key":"ref_36","unstructured":"McInnes, L., Healy, J., and Melville, J. (2018). Umap: Uniform manifold approximation and projection for dimension reduction. arXiv."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/11\/11\/1374\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T12:57:02Z","timestamp":1760187422000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/11\/11\/1374"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,6,8]]},"references-count":36,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2019,6]]}},"alternative-id":["rs11111374"],"URL":"https:\/\/doi.org\/10.3390\/rs11111374","relation":{"has-preprint":[{"id-type":"doi","id":"10.20944\/preprints201905.0030.v1","asserted-by":"object"}]},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,6,8]]}}}