Deep Transfer Learning for Remote Sensing

Dear Colleagues,

Recently, deep learning (DL) for remote sensing (RS) image processing has gradually become a hot topic. Many deep learning models, including ResNet, AlexNet, as well as the newly proposed capsule network, have all been proven to have decent performance on RS images with enough prior knowledge for training. One existing problem is the limitation of label information for newly collected RS data, and this phenomenon will make it even more difficult for the DL models to process the RS images. With the development of modern satellite sensors and easy access to new RS data, the problem of processing such a large amount of data becomes even more serious and urgent. A straightforward consideration is to resort to existing labeled RS data to help with the unknown new data. To achieve this purpose, deep transfer learning-based frameworks that can overcome the semantic gap between different datasets have become a research frontier in RS data processing. The deep information of existing labeled data is exploited to predict the label of newly collected RS data.

This Special Issue is devoted to exploring the potential of deep transfer learning framework in RS image processing. Due to different acquisition conditions and sensors, the spectra observed on a new scene can be quite different from the existing scene even if they represent the same types of objects. This spectral difference brings huge semantic disparity among different RS datasets. Therefore, how to select, construct, and correlate the deep networks by transfer learning for different RS datasets will be the major concern of this Special Issue.

Topics of interest include, but are not limited to:

• Theories for domain adaptation and generalization;
• Auto-encoder-based transfer learning for remote sensing;
• CNN-based transfer learning for remote sensing;
• RNN-based transfer learning for remote sensing;
• Capsule network-based transfer learning for remote sensing;
• Domain generalization algorithms for visual problems;
• Deep representation learning for domain adaptation and generalization.

Dr. Jianzhe Lin
Dr. Zhiyu Jiang
Dr. Sarbjit Sarkaria
Dr. Dandan Ma
Dr. Yang Zhao
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.