Dear Colleagues,

Hyperspectral images capture the spectral data for each pixel and provide very detailed characteristics of the materials within a scene. Hence, exploiting such detailed spectral information can open new possibilities in various domains. With the current sensor advances, we are facing exciting challenges concerned with efficient analysis of the highly dimensional image data in a plethora of real-life use cases, ranging from remote sensing, precision agriculture, chemistry, and biology to forensic applications, just to mention a few.

Classification and segmentation of hyperspectral imagery have been attracting research attention due to their wide practical applicability. By classification, we mean assigning class labels to specific hyperspectral pixels, while by segmentation, we mean finding the boundaries of the same-class objects in the entire input hyperspectral scene. Hence, segmentation involves classification of separate pixels in this case. Additionally, we can classify the full hyperspectral images (or patches) and assign specific class labels to them. Although these tasks are practically never the final ones in the hyperspectral processing chain, they clearly affect any further analysis steps. Therefore, improving the performance of hyperspectral classification and segmentation techniques is an extremely important research topic. We have been witnessing unprecedented success of deep learning in the field; however, there are still open issues to be carefully addressed in emerging applications—often related to the lack of ground-truth data (or its limited availability). Classical image analysis and machine learning algorithms are still of very high research interest as well.

The aim of this Special Issue is to gather and present recent advances in hyperspectral image classification and segmentation. The core themes of this topic cover all steps of the data processing pipeline, from its acquisition to final analysis and understanding, with special emphasis put on classification and segmentation. The themes of the Special Issue include but are not limited to:

• Hyperspectral data reduction in classification and segmentation tasks;
• Feature extraction and learning for accurate classification and segmentation;
• Classification and segmentation of hyperspectral images: algorithms and tools;
• Classification and segmentation of hyperspectral images: real-life use cases;
• Unsupervised, semi-supervised, supervised learning for hyperspectral classification and segmentation;
• Multitemporal and multisensor analysis;
• Event detection and tracking;
• Prediction from hyperspectral data;
• Deployment of hyperspectral classification and segmentation techniques in hardware-constrained environments;
• Robustness of automated hyperspectral classification and segmentation;
• Interpretability of hyperspectral classification and segmentation algorithms;
• Concept drift in hyperspectral data analysis;
• Change and anomaly detection in hyperspectral images.

Dr. Jakub Nalepa
Guest Editor

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. Electronics 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 2400 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.

• Hyperspectral imaging
• Classification
• Segmentation
• Regression
• Hyperspectral Unmixing
• Machine learning
• Deep learning
• Supervised, unsupervised, semi-supervised learning
• Image processing and analysis