Deep Learning and Transformers for Classification, Prediction and Image Fusion

Dear Colleagues,

Deep learning- and transformer-based architectures are transforming how images are processed, interpreted, and integrated across scientific and industrial domains.

Learning robust, transferable representations from visual data is the main methodological thread shared by classification, prediction and image fusion, and attention mechanisms are key to learning such representations that capture local detail and long-range or cross-modal dependencies. These forms of representations can be used for classification and prediction tasks, while image fusion combines complementary information from multiple sensors or modalities in the same representation space, making the Vision Transformers and hybrid CNN–Transformer models central to all three tasks. However, there are still a lot of problems to be solved in order to maintain accuracy and computation efficiency in real-world applications, make the system robust and reliable in heterogeneous and noisy data conditions, and combine multimodal data without losing interpretation ability.

For the Special Issue “Deep Learning and Transformers for Classification, Prediction and Image Fusion”, we invite research papers that explore the above-mentioned challenges. The main focus of this collection is deep learning and Transformer-based representation learning methodologies and their applications in healthcare sector, remote sensing, and industrial inspections. Submissions should offer a clear methodological contribution (deep learning or Transformer-based models).

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

1. Deep learning architectures for image classification and semantic segmentation;
2. Vision Transformers and hybrid CNN–Transformer models for image analysis;
3. Multi-modal and multi-sensor image fusion techniques;
4. Deep learning for medical image diagnosis and prediction;
5. Remote sensing image classification and land-cover/land-use mapping;
6. Lightweight and resource-efficient deep learning models for edge deployment;
7. Explainable and interpretable AI for image-based decision support;
8. Self-supervised and few-shot learning for image classification and fusion;
9. Attention mechanisms and feature fusion strategies for spatial–spectral data;
10. Benchmarking, datasets, and evaluation methodologies for image processing tasks;
11. Generative models (GANs, diffusion models) for image synthesis, enhancement, and fusion;
12. Domain adaptation and transfer learning for cross-domain image classification;
13. Transformer-based time series and spatiotemporal prediction from image sequences;
14. Uncertainty quantification and robustness in deep learning-based image prediction;
15. Edge AI and real-time deployment of transformer models for image-based applications.

The articles in this Special Issue will highlight state-of-the-art innovative methodological developments and applications, providing a forum to share novel ideas for new research synergies between deep learning and transformer models and image processing research communities. We will be grateful to receive your submissions and hope to continue to develop and contribute to the theory and practice of this quickly changing field.

Prof. Dr. Ali Ahmed
Dr. Ashraf Osman Ibrahim
Guest Editors

Dr. Abdolraheem Khader
Guest Editor Assistant

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-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Information is an international peer-reviewed open access monthly 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 1800 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.