Deep Learning for Remote Sensing and Geodata

Dear Colleagues,

With the volume and the variety of remote sensing data increasing exponentially, it has become possible to observe the same location on the Earth multiple times per day. The richness of remote sensing data enables land-use observation, change and anomaly detection, water quality monitoring and observation of greenhouse gases. While the satellite and aerial data are readily available, one obstacle for the deployment of global-scale Artificial Intelligence (AI) models is the lack of labeled data used for training machine learning algorithms and the lack of benchmark datasets that allow for side-by-side comparison of different models.

In the last decades, AI has evolved from classical machine learning models (Random Forest, Support Vector Machines, etc.) to deep learning (Convolutional Neural Networks, etc.) and then to foundational models. Foundational models based on self-supervised learning trained on massive heterogeneous datasets can be tuned to various downstream tasks on the backbone of the same model. While the foundational models are exposed to different types of data (multispectral, radar, LiDAR, crowdsource data), there is ongoing interest in understanding the performance and generalizability of these models for multiple applications. Of ongoing interest is the comparison of foundational models with classical deep learning models and benchmarking the accuracy and reproducibility of these modes.

We encourage submissions of original manuscripts that focus on scalable AI methodologies and benchmark dataset creation to develop foundational models and to characterize network architecture and applications based on remote sensing data. Topics can include, but are not limited to:

• Self-supervised neural networks applied to remote sensing data;
• Development of new algorithms and applications using heterogeneous data sources;
• Automated data label generation for AI models;
• Remote sensing of land, water, and air using multiscale deep learning models;
• Integration of physical and statistical constrain to model continental scale systems;
• Multi-sensor fusion and deep learning algorithms for detection and anomaly identification;
• Application of AI models in agriculture, forestry, water quality, food security, and infrastructure monitoring.

Dr. Levente Klein
Dr. Conrad Albrecht
Prof. Dr. Martin Werner
Dr. Salman Khan
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.

• foundational models
• deep learning
• artificial intelligence
• satellite and aerial remote sensing
• LiDAR and point cloud processing
• earth digital twin