Emanuele Dalsasso

PhD at télécom paris · emanuele.dalsasso@telecom-paris.fr

Currently pursuing a PhD degree on "Deep Learning for SAR imagery: from denoising to scene understanding".


Education

University of Trento

Master of Science
Information and Communication Engineering

Final project: “SAR image denoising through Convolutional Neural Networks” developed at the IMAGES group.

110 marks out of 110 with honour

September 2016 - October 2018

Technical University of Denmark

Master of Science
Erasmus+ scholarship.

One semester exchange program.

August 2017 - January 2018

University of Trento

Bachelor of Science
Electronics and Telecommunication Engineering

Final project: “Soil Moisture Estimation through Support Vector Regression using SAR data” developed at the RSLab.

September 2013 - July 2016

Teaching


Research

  • PhD main project: "Deep Learning for SAR imagery: from denoising to scene understanding"
  • Student projects supervision
    • FFDNet for SAR despeckling ()
    • SAR River Segmentation through deep learning ()
    • Route Extraction from SAR images: an image segmentation problem
  • Co-organising the Deep Learning Working Group of the IMAGES team
    • Give a look to our webpage:
  • Some articles I've been reading: under construction

Publications

A-M. Ilisei, M. Khodadadzadeh, E. Dalsasso, L. Bruzzone, “Automatic detection of subglacial lakes in radar sounder data acquired in Antarctica”, Proc. SPIE 10427, Image and Signal Processing for Remote Sensing XXIII, 1042718 (4 October 2017)

E. Dalsasso, X. Yang, L. Denis, F. Tupin, W. Yang, “SAR Image Despeckling by Deep Neural Networks: from a pre-trained model to an end-to-end training strategy” arXiv:2006.15559, 2020. PDF and code

E. Dalsasso, L. Denis, and F. Tupin, “How to handle spatial correlations in SAR despeckling? Resampling strategies and deep learning approaches” hal-02538046, 2020. PDF

E. Dalsasso, L. Denis, and F. Tupin, “SAR2SAR: a self-supervised despeckling algorithm for SAR images” arXiv:2006.15037, 2020. PDF and code