{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T15:30:41Z","timestamp":1772206241324,"version":"3.50.1"},"reference-count":57,"publisher":"MDPI AG","issue":"13","license":[{"start":{"date-parts":[[2019,6,28]],"date-time":"2019-06-28T00:00:00Z","timestamp":1561680000000},"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>Speckle filtering is an unavoidable step when dealing with applications that involve amplitude or intensity images acquired by coherent systems, such as Synthetic Aperture Radar (SAR). Speckle is a target-dependent phenomenon; thus, its estimation and reduction require the individuation of specific properties of the image features. Speckle filtering is one of the most prominent topics in the SAR image processing research community, who has first tackled this issue using handcrafted feature-based filters. Even if classical algorithms have slowly and progressively achieved better and better performance, the more recent Convolutional-Neural-Networks (CNNs) have proven to be a promising alternative, in the light of the outstanding capabilities in efficiently learning task-specific filters. Currently, only simplistic CNN architectures have been exploited for the speckle filtering task. While these architectures outperform classical algorithms, they still show some weakness in the texture preservation. In this work, a deep encoder\u2013decoder CNN architecture, focused in the specific context of SAR images, is proposed in order to enhance speckle filtering capabilities alongside texture preservation. This objective has been addressed through the adaptation of the U-Net CNN, which has been modified and optimized accordingly. This architecture allows for the extraction of features at different scales, and it is capable of producing detailed reconstructions through its system of skip connections. In this work, a two-phase learning strategy is adopted, by first pre-training the model on a synthetic dataset and by adapting the learned network to the real SAR image domain through a fast fine-tuning procedure. During the fine-tuning phase, a modified version of the total variation (TV) regularization was introduced to improve the network performance when dealing with real SAR data. Finally, experiments were carried out on simulated and real data to compare the performance of the proposed method with respect to the state-of-the-art methodologies.<\/jats:p>","DOI":"10.3390\/rs11131532","type":"journal-article","created":{"date-parts":[[2019,6,28]],"date-time":"2019-06-28T11:20:26Z","timestamp":1561720826000},"page":"1532","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":131,"title":["Deep Learning for SAR Image Despeckling"],"prefix":"10.3390","volume":"11","author":[{"given":"Francesco","family":"Lattari","sequence":"first","affiliation":[{"name":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milano, Italy"}]},{"given":"Borja","family":"Gonzalez Leon","sequence":"additional","affiliation":[{"name":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milano, Italy"}]},{"given":"Francesco","family":"Asaro","sequence":"additional","affiliation":[{"name":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milano, Italy"}]},{"given":"Alessio","family":"Rucci","sequence":"additional","affiliation":[{"name":"TRE ALTAMIRA s.r.l., 20143 Milano, Italy"}]},{"given":"Claudio","family":"Prati","sequence":"additional","affiliation":[{"name":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milano, Italy"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8306-6739","authenticated-orcid":false,"given":"Matteo","family":"Matteucci","sequence":"additional","affiliation":[{"name":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milano, Italy"}]}],"member":"1968","published-online":{"date-parts":[[2019,6,28]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"R1","DOI":"10.1088\/0266-5611\/14\/4\/001","article-title":"Synthetic aperture radar interferometry","volume":"14","author":"Bamler","year":"1998","journal-title":"Inverse Probl."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"165","DOI":"10.1109\/TPAMI.1980.4766994","article-title":"Digital Image Enhancement and Noise Filtering by Use of Local Statistics","volume":"PAMI-2","author":"Lee","year":"1980","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"165","DOI":"10.1109\/TPAMI.1985.4767641","article-title":"Adaptive Noise Smoothing Filter for Images with Signal-Dependent Noise","volume":"PAMI-7","author":"Kuan","year":"1985","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1109\/TPAMI.1982.4767223","article-title":"A Model for Radar Images and Its Application to Adaptive Digital Filtering of Multiplicative Noise","volume":"PAMI-4","author":"Frost","year":"1982","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_5","unstructured":"Shi, Z., and Fung, K.B. (1994, January 8\u201312). A comparison of digital speckle filters. Proceedings of the IGARSS\u201994\u20141994 IEEE International Geoscience and Remote Sensing Symposium, Pasadena, CA, USA."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"992","DOI":"10.1109\/36.62623","article-title":"Adaptive speckle filters and scene heterogeneity","volume":"28","author":"Lopes","year":"1990","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"2196","DOI":"10.1109\/TGRS.2002.802473","article-title":"SAR speckle reduction using wavelet denoising and Markov random field modeling","volume":"40","author":"Xie","year":"2002","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2001","DOI":"10.1109\/TGRS.2011.2169679","article-title":"Evaluation of Bayesian Despeckling and Texture Extraction Methods Based on Gauss\u2013Markov and Auto-Binomial Gibbs Random Fields: Application to TerraSAR-X Data","volume":"50","author":"Gleich","year":"2012","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"485","DOI":"10.1080\/07038992.2017.1381549","article-title":"The Effect of PolSAR Image De-speckling on Wetland Classification: Introducing a New Adaptive Method","volume":"43","author":"Mahdianpari","year":"2017","journal-title":"Can. J. Remote Sens."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"2363","DOI":"10.1109\/TGRS.2002.805083","article-title":"Speckle removal from SAR images in the undecimated wavelet domain","volume":"40","author":"Argenti","year":"2002","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"711","DOI":"10.1109\/TGRS.2003.821885","article-title":"Homomorphic wavelet-based statistical despeckling of SAR images","volume":"42","author":"Solbo","year":"2004","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_12","unstructured":"Lopes, A., Nezry, E., Touzi, R., and Laur, H. (1990, January 20\u201324). Maximum A Posteriori Speckle Filtering And First Order Texture Models In Sar Images. Proceedings of the 10th Annual International Symposium on Geoscience and Remote Sensing, College Park, MD, USA."},{"key":"ref_13","unstructured":"Buades, A., Coll, B., and Morel, J. (2005, January 20\u201325). A non-local algorithm for image denoising. Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR\u201905), San Diego, CA, USA."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2661","DOI":"10.1109\/TIP.2009.2029593","article-title":"Iterative Weighted Maximum Likelihood Denoising With Probabilistic Patch-Based Weights","volume":"18","author":"Deledalle","year":"2009","journal-title":"IEEE Trans. Image Process."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2080","DOI":"10.1109\/TIP.2007.901238","article-title":"Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering","volume":"16","author":"Dabov","year":"2007","journal-title":"IEEE Trans. Image Process."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"606","DOI":"10.1109\/TGRS.2011.2161586","article-title":"A Nonlocal SAR Image Denoising Algorithm Based on LLMMSE Wavelet Shrinkage","volume":"50","author":"Parrilli","year":"2012","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1016\/0167-2789(92)90242-F","article-title":"Nonlinear total variation based noise removal algorithms","volume":"60","author":"Rudin","year":"1992","journal-title":"Phys. D Nonlinear Phenom."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"925","DOI":"10.1137\/060671814","article-title":"A Variational Approach to Removing Multiplicative Noise","volume":"68","author":"Aubert","year":"2008","journal-title":"SIAM J. Appl. Math."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"168","DOI":"10.1007\/s10851-009-0179-5","article-title":"Removing Multiplicative Noise by Douglas-Rachford Splitting Methods","volume":"36","author":"Steidl","year":"2010","journal-title":"J. Math. Imaging Vis."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1720","DOI":"10.1109\/TIP.2010.2045029","article-title":"Multiplicative Noise Removal Using Variable Splitting and Constrained Optimization","volume":"19","author":"Figueiredo","year":"2010","journal-title":"IEEE Trans. Image Process."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"294","DOI":"10.1137\/070689954","article-title":"A Nonlinear Inverse Scale Space Method for a Convex Multiplicative Noise Model","volume":"1","author":"Shi","year":"2008","journal-title":"SIAM J. Img. Sci."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"2765","DOI":"10.1109\/TGRS.2014.2364525","article-title":"Adaptive Total Variation Regularization Based SAR Image Despeckling and Despeckling Evaluation Index","volume":"53","author":"Zhao","year":"2015","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Palsson, F., Sveinsson, J.R., Ulfarsson, M.O., and Benediktsson, J.A. (2012, January 22\u201327). SAR image denoising using total variation based regularization with sure-based optimization of the regularization parameter. Proceedings of the 2012 IEEE International Geoscience and Remote Sensing Symposium, Munich, Germany.","DOI":"10.1109\/IGARSS.2012.6351075"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"3460","DOI":"10.1109\/TGRS.2011.2124465","article-title":"A New Algorithm for Processing Interferometric Data-Stacks: SqueeSAR","volume":"49","author":"Ferretti","year":"2011","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"5467","DOI":"10.1109\/TGRS.2017.2707806","article-title":"Multitemporal SAR Image Despeckling Based on Block-Matching and Collaborative Filtering","volume":"55","author":"Chierchia","year":"2017","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Zhao, W., Deledalle, C.A., Denis, L., Ma\u00eetre, H., Nicolas, J.M., and Tupin, F. (2018). Ratio-based multi-temporal SAR images denoising. IEEE Trans. Geosci. Remote. Sens.","DOI":"10.1109\/TGRS.2018.2885683"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015, January 5\u20139). U-net: Convolutional networks for biomedical image segmentation. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Munich, Germany.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_28","unstructured":"Simonyan, K., and Zisserman, A. (2014). Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., and Rabinovich, A. (2015, January 7\u201312). Going deeper with convolutions. Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA.","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"ref_30","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_31","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1016\/j.isprsjprs.2010.11.001","article-title":"Support vector machines in remote sensing: A review","volume":"66","author":"Mountrakis","year":"2011","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Guo, J., Zhang, J., Zhang, Y., and Cao, Y. (July, January 30). Study on the comparison of the land cover classification for multitemporal MODIS images. Proceedings of the 2008 International Workshop on Earth Observation and Remote Sensing Applications, Beijing, China.","DOI":"10.1109\/EORSA.2008.4620305"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Redmon, J., and Farhadi, A. (2016). YOLO9000: Better, Faster, Stronger. arXiv.","DOI":"10.1109\/CVPR.2017.690"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"He, K., Gkioxari, G., Doll\u00e1r, P., and Girshick, R.B. (2017). Mask R-CNN. arXiv.","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref_35","unstructured":"Badrinarayanan, V., Kendall, A., and Cipolla, R. (2015). SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. arXiv."},{"key":"ref_36","unstructured":"Chen, L., Papandreou, G., Kokkinos, I., Murphy, K., and Yuille, A.L. (2016). DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. arXiv."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Lin, G., Milan, A., Shen, C., and Reid, I.D. (2016). RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation. arXiv.","DOI":"10.1109\/CVPR.2017.549"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1109\/MGRS.2016.2540798","article-title":"Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art","volume":"4","author":"Zhang","year":"2016","journal-title":"IEEE Geosci. Remote Sens. Mag."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"e1264","DOI":"10.1002\/widm.1264","article-title":"Deep learning for remote sensing image classification: A survey","volume":"8","author":"Li","year":"2018","journal-title":"Wiley Interdiscip. Rev. Data Min. Knowl. Discov."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"1915","DOI":"10.5194\/isprs-archives-XLII-3-1915-2018","article-title":"Rapid Target Detection in High Resolution Remote Sensing Images Using Yolo Model","volume":"XLII-3","author":"Wu","year":"2018","journal-title":"Int. Arch. Photogramm. Remote. Sens. Spat. Inf. Sci."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Zhen, Y., Liu, H., Li, J., Hu, C., and Pan, J. (2017, January 3\u20135). Remote sensing image object recognition based on convolutional neural network. Proceedings of the 2017 First International Conference on Electronics Instrumentation Information Systems (EIIS), Harbin, China.","DOI":"10.1109\/EIIS.2017.8298722"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Zhang, K., Zuo, W., Chen, Y., Meng, D., and Zhang, L. (2016). Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising. arXiv.","DOI":"10.1109\/TIP.2017.2662206"},{"key":"ref_43","unstructured":"Ioffe, S., and Szegedy, C. (2015). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Chierchia, G., Cozzolino, D., Poggi, G., and Verdoliva, L. (2017, January 23\u201328). SAR image despeckling through convolutional neural networks. Proceedings of the 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Fort Worth, TX, USA.","DOI":"10.1109\/IGARSS.2017.8128234"},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Wang, P., Zhang, H., and Patel, V.M. (2017). SAR Image Despeckling Using a Convolutional Neural Network. arXiv.","DOI":"10.1109\/LSP.2017.2758203"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Zhang, Q., Yang, Z., Yuan, Q., Li, J., Ma, X., Shen, H., and Zhang, L. (2017). Learning a Dilated Residual Network for SAR Image Despeckling. arXiv.","DOI":"10.3390\/rs10020196"},{"key":"ref_47","unstructured":"Yu, F., and Koltun, V. (2015). Multi-Scale Context Aggregation by Dilated Convolutions. arXiv."},{"key":"ref_48","unstructured":"Luo, W., Li, Y., Urtasun, R., and Zemel, R.S. (2017). Understanding the Effective Receptive Field in Deep Convolutional Neural Networks. arXiv."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"1145","DOI":"10.1364\/JOSA.66.001145","article-title":"Some fundamental properties of speckle*","volume":"66","author":"Goodman","year":"1976","journal-title":"J. Opt. Soc. Am."},{"key":"ref_50","unstructured":"Ulaby, F.T., and Dobson, M.C. (1989). Handbook of radar scattering statistics for terrain (Artech House Remote Sensing Library), Artech House."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Vincent, P., Larochelle, H., Bengio, Y., and Manzagol, P.A. (2008, January 5\u20139). Extracting and Composing Robust Features with Denoising Autoencoders. Proceedings of the 25th International Conference on Machine Learning, ICML\u201908, Helsinki, Finland.","DOI":"10.1145\/1390156.1390294"},{"key":"ref_52","first-page":"3371","article-title":"Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion","volume":"11","author":"Vincent","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref_53","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","volume":"Volume 9","author":"Teh","year":"2010","journal-title":"Machine Learning Research, Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 2010"},{"key":"ref_54","unstructured":"Dumoulin, V., and Visin, F. (2016). A guide to convolution arithmetic for deep learning. arXiv."},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Yang, Y., and Newsam, S. (2010, January 2\u20135). Bag-of-visual-words and Spatial Extensions for Land-use Classification. Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems, GIS\u201910, San Jose, CA, USA.","DOI":"10.1145\/1869790.1869829"},{"key":"ref_56","doi-asserted-by":"crossref","unstructured":"Zhou, W., Newsam, S.D., Li, C., and Shao, Z. (2017). PatternNet: A Benchmark Dataset for Performance Evaluation of Remote Sensing Image Retrieval. arXiv.","DOI":"10.1016\/j.isprsjprs.2018.01.004"},{"key":"ref_57","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A Method for Stochastic Optimization. arXiv."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/11\/13\/1532\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T13:02:00Z","timestamp":1760187720000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/11\/13\/1532"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,6,28]]},"references-count":57,"journal-issue":{"issue":"13","published-online":{"date-parts":[[2019,7]]}},"alternative-id":["rs11131532"],"URL":"https:\/\/doi.org\/10.3390\/rs11131532","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,6,28]]}}}