Integrating Spectroscopy and Machine Learning for Crop Phenotyping

Dear Colleagues,

The escalating demands being placed on global agriculture, driven by climate change and population growth, necessitate accelerated crop improvement and sustainable management practices. Central to these challenges is crop phenotyping, the quantitative assessment of plant traits. Historically, phenotyping has relied on laborious manual measurements that created bottlenecks in research and breeding.

The development of spectroscopic technologies over the past two decades has, however, revolutionized this field. Techniques such as near-infrared (NIR) spectroscopy and hyperspectral and multispectral imaging provide rich, noninvasive datasets capturing the biochemical and physiological status of plants. Importantly, the increasing availability of compact and portable instruments, coupled with decreasing costs of hardware and sensors, has made these technologies more accessible for research, applied agricultural settings, and commercial applications. At the same time, advances in computational power have been essential, enabling the handling of this high-dimensional information. The progress in artificial intelligence (AI), particularly in machine learning (ML) and deep learning, has been substantial, providing sophisticated tools capable of interpreting these complex spectral signatures and moving beyond traditional chemometric methods to reveal intricate patterns related to plant performance.

This Special Issue of Agriculture aims to capture the latest innovations emerging from this synergy. We seek to compile cutting-edge research demonstrating how this integration enhances the accuracy, efficiency, and scalability of phenotyping across diverse contexts.

We invite the submission of high-quality original research articles, comprehensive reviews, and perspectives, and we strongly encourage submissions detailing novel applications of spectroscopy and imaging for trait estimation, stress detection, yield prediction, and the analysis of fruit quality, composition, and ripening dynamics.

Furthermore, we are interested in the implementation of advanced computational methods, encompassing sophisticated algorithms designed to analyze high-dimensional data, such as convolutional neural networks (CNNs) for spatial/spectral analysis, transformers for modeling spectral sequences, and advanced computer and machine vision techniques for high-throughput morphological analysis. We also welcome research utilizing established ML methods (e.g., SVMs, random forests, and PLS) in novel comparative studies.

A major area of focus is the integration and practical application of these technologies. We welcome research presenting data fusion strategies that combine spectral data with other modalities, such as LiDAR or thermal imaging. Equally important is the development of robust ML models designed for real-world field conditions, including studies utilizing transfer learning and domain adaptation. Finally, we solicit papers addressing the need for model interpretability, such as the application of explainable AI (XAI) or other feature selection methods for interpreting spectral models in agricultural science applications.

We look forward to receiving your contributions to this exciting and vital area of research.

Dr. Jan Skvaril
Prof. Dr. Stefka Atanassova
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. Agriculture 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 2600 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.

• crop phenotyping
• machine learning
• deep learning
• hyperspectral imaging
• nir spectroscopy
• computer vision
• sensor fusion
• explainable AI (XAI)
• high-throughput phenotyping
• precision agriculture