Advances in Deep Learning Change Detection Based on High-Resolution Remote Sensing Imagery

Dear Colleagues,

High-resolution remote sensing (RS) imagery change detection (CD) plays an important role in resources surveying, urban sprawl monitoring, environmental assessment, and rapid disaster response. With the development of remote sensing big data, high-performance computing resources and powerful network architectures (such as CNN, Transformer, Mamba, etc.), deep learning techniques have been widely introduced to address CD tasks, pushing the automation and intelligence of CD to an amazing level. However, how to effectively capture real changes while suppressing the interface of noise remains an ongoing challenge. In addition, large amounts of labeled data are often needed to train such CD networks, which is usually costly and time-consuming to acquire, especially for multi-temporal high-resolution RS images. In such a context, many label-efficient learning techniques have been proposed to tackle these issues, such as semi-supervised learning, weak-supervised learning, active learning, and self-supervised learning. Particularly, the emergence of remote sensing large foundation models (such as SatMAE, ScaleMAE, SkySense, etc.), zero-shot foundation models (such as CLIP, DINO, SAM, SAM2, etc.), and powerful generative models (such as Stable Diffusion, ControlNet, etc.) has opened up new opportunities to address CD tasks in a more robust way. Motivated by the rapid development of these amazing techniques, we are excited to invite you to submit a research paper to this Special Issue, entitled "Advances in Deep Learning Change Detection Based on High-Resolution Remote Sensing Imagery". The aim of this Special Issue is to highlight state-of-the-art research that addresses various issues of deep learning CD for high-resolution remote sensing imagery with the development of model architectures, label-efficient learning methods, and large foundation models, among others. In addition, we also aim to highlight research work focused on smart deep learning CD applications in various areas, including geohazard identification, natural resources investigation, environmental monitoring, and rapid disaster response.

Topics of interest include, but are not limited to, the following:

• High-accuracy change detection techniques for high-resolution remote sensing imagery;
• Semi-supervised change detection techniques for high-resolution remote sensing imagery;
• Weak-supervised change detection techniques for high-resolution remote sensing imagery;
• Active learning techniques for high-resolution remote sensing imagery change detection;
• Self-supervised change detection techniques for high-resolution remote sensing imagery;
• Foundation model-assisted change detection techniques for high-resolution remote sensing imagery;
• Diffusion model-assisted change detection techniques for high-resolution remote sensing imagery;
• High-quality change sample generation using diffusion models;
• Domain adaption change detection techniques for high-resolution remote sensing imagery;
• Zero-shot change detection techniques for high-resolution remote sensing imagery;
• Open-vocabulary change detection techniques for high-resolution remote sensing imagery.

Dr. Daifeng Peng
Dr. Sicong Liu
Dr. Yuxing Chen
Dr. Zhi Zheng
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.

• high-resolution remote sensing imagery
• change detection
• deep learning
• foundation model
• vision-language model
• label-efficient learning
• domain adaptation
• zero-shot learning