Deep Learning and Generative Artificial Intelligence for Hyperspectral Remote Sensing

Dear Colleagues,

Deep learning and generative artificial intelligence (AI) have become transformative forces in hyperspectral image (HSI) analysis, offering new ways to leverage the rich spectral–spatial information contained in HSI data. Hyperspectral imagery provides hundreds of contiguous spectral bands, enabling unparalleled material characterization and supporting diverse applications such as environmental monitoring, precision agriculture, urban analysis, earth observation and mineral exploration. However, the high dimensionality of HSI data, limited availability of labeled samples, and computational challenges pose significant obstacles to effective analysis.

Recent advances in deep learning have demonstrated remarkable success in extracting complex spectral–spatial features, achieving state-of-the-art performance in tasks such as classification, segmentation, unmixing, and anomaly detection. In parallel, generative AI techniques, including generative adversarial networks (GANs), diffusion models, and large-scale generative architectures, are opening new avenues for data augmentation, super-resolution, denoising, and realistic HSI synthesis. This convergence of deep learning and generative AI represents a paradigm shift, enabling more robust, scalable, and interpretable solutions for managing the complexity and growing volume of hyperspectral data.

This Special Issue will showcase cutting-edge research at the intersection of deep learning, generative AI, and hyperspectral remote sensing, fully aligned with the commitment of Remote Sensing to advancing innovative methodologies and impactful applications in Earth observation. Our goal is to gather contributions that achieve the following:
-  Advance the development of novel deep learning and generative AI models tailored to hyperspectral data;
-  Address current limitations such as high dimensionality, data scarcity, and model interpretability;
-  Demonstrate real-world applications across diverse domains where hyperspectral imaging plays a critical role.

Through both theoretical innovations and practical case studies, this Special Issue will highlight how deep learning and generative AI can unlock the full potential of HSI for accurate, efficient, and scalable analysis.

We invite original research articles, reviews, and application-focused studies on the following themes:  
- Advanced Deep Learning Architectures for HSI: Development of 1D/2D/3D CNNs, transformer-based models, graph neural networks, and hybrid spectral–spatial architectures optimized for hyperspectral data;
- Generative AI for Hyperspectral Imaging: GANs, diffusion models, and other generative approaches for HSI synthesis, data augmentation, anomaly generation, and reconstruction;
- Spectral–Spatial Feature Extraction and Fusion: Innovative techniques for integrating hyperspectral data with other modalities (e.g., LiDAR, SAR, multispectral) to improve classification and detection;
- Data Enhancement and Dimensionality Reduction: Super-resolution, denoising, compression, band selection, and feature reduction methods leveraging deep or generative models;
- Advanced Analysis Tasks: Pixel- and sub-pixel-level classification, anomaly and target detection, segmentation, spectral unmixing, and clustering using deep or generative models;
- Interpretability, Robustness, and Efficiency: Explainable AI for hyperspectral models, robustness to noise and adversarial attacks, and lightweight models for onboard or real-time processing.

Dr. Yaser Esmaeili Salehani
Dr. Linlin Xu
Dr. Yasser Maghsoudi
Guest Editors

Dr. Nasehe Jamshidpour
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 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.

• deep learning (DL)
• hyperspectral imaging (HSI)
• hyperspectral image classification and unmixing
• spectral–spatial analysis
• convolutional neural networks (CNNs)
• self supervised and generative models
• graph convolutional networks (GCNs)
• transformer models
• dimensionality reduction and band selection
• transfer learning and domain adaptation