Deep Transfer Learning for Remote Sensing II
A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Remote Sensing Image Processing".
Deadline for manuscript submissions: 30 November 2024 | Viewed by 2658
Special Issue Editors
Interests: artificial intelligence; machine learning; pattern recognition; data mining
Interests: blind estimation of degradation characteristics (noise, PSF); blind restoration of multicomponent images; multimodal image correction; multicomponent image compression; multi-channel adaptive processing of signals and images; unsupervised machine learning and deep learning; multi-mode remote sensing data processing; remote sensing
Special Issues, Collections and Topics in MDPI journals
Interests: graph machine learning; deep learning; computer vision
Interests: image/video representations and analysis; semi-supervised/unsupervised data modeling
Special Issues, Collections and Topics in MDPI journals
Interests: machine learning; computer vision; artificial intelligence
Special Issue Information
Dear Colleagues,
Recently, deep learning (DL) for remote sensing (RS) image processing has emerged as a hot topic. Many deep learning models have been shown to perform well on RS images when trained with sufficient data. However, a limitation of DL for RS is that newly collected RS data usually have limited label information, which makes the performance of DL models processing RS images unsatisfactory. A straightforward solution is to resort to existing labeled RS data to help process the new data, which falls under the scope of transfer learning (also known as domain adaptation).
Transfer learning attempts to reduce the high demand for labeled data on a target task by reusing knowledge obtained from one or more source tasks. Nowadays, transfer learning with deep neural networks, known as deep transfer learning, is a mainstream approach due to its powerful representation learning ability. In RS data processing, deep transfer learning that can overcome the semantic gap between different datasets has become a research frontier and it can utilize the information contained in existing labeled data to help make predictions for newly collected RS data.
This Special Issue is dedicated to exploring the potential of deep transfer learning in RS image processing. Due to differences in acquisition conditions and sensors, the spectra observed in a new scene may be very different from existing scenes, even if they represent the same types of objects. Such spectral differences introduce significant semantic differences between different RS datasets. Therefore, how to build deep transfer learning models for different RS datasets will be the main focus of this Special Issue.
Topics of interest include, but are not limited to:
- Unsupervised domain adaptation for remote sensing under various settings (e.g., closed-set, open-set, partial, and universal domain adaptation);
- Semi-supervised transfer learning for remote sensing;
- Meta transfer learning for remote sensing;
- Multi-task learning for remote sensing;
- Domain generalization for remote sensing;
- Transfer learning for remote sensing with various architectures (e.g., vision transformers, CNNs, RNNs, and capsule networks);
- Representation learning for transfer learning in remote sensing;
- Utilizing pre-trained large vision/NLP models in remote sensing;
- Deep generative models for transfer learning in remote sensing;
- Theories of transfer learning, domain adaptation, and domain generalization.
Dr. Yu Zhang
Dr. Benoit Vozel
Dr. Yuheng Jia
Dr. Junhui Hou
Prof. Dr. Deyu Meng
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 100 words) can be sent to the Editorial Office for announcement on this website.
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.
Keywords
- transfer learning
- remote sensing images
- domain adaptation
- domain generalization
- CNN
Benefits of Publishing in a Special Issue
- Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
- Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
- Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
- External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
- e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.
Further information on MDPI's Special Issue polices can be found here.
Related Special Issue
- Deep Transfer Learning for Remote Sensing in Remote Sensing (7 articles)
