Version 1
: Received: 1 October 2024 / Approved: 2 October 2024 / Online: 3 October 2024 (04:11:17 CEST)
How to cite:
Thomas, S.; Varga, L. A.; Harter, N.; Zell, A.; Voegele, R. T. Detection of Phakopsora pachyrhizi Infestation in Soybean via Hyperspectral Imaging and Data Analysis. Preprints2024, 2024100142. https://doi.org/10.20944/preprints202410.0142.v1
Thomas, S.; Varga, L. A.; Harter, N.; Zell, A.; Voegele, R. T. Detection of Phakopsora pachyrhizi Infestation in Soybean via Hyperspectral Imaging and Data Analysis. Preprints 2024, 2024100142. https://doi.org/10.20944/preprints202410.0142.v1
Thomas, S.; Varga, L. A.; Harter, N.; Zell, A.; Voegele, R. T. Detection of Phakopsora pachyrhizi Infestation in Soybean via Hyperspectral Imaging and Data Analysis. Preprints2024, 2024100142. https://doi.org/10.20944/preprints202410.0142.v1
APA Style
Thomas, S., Varga, L. A., Harter, N., Zell, A., & Voegele, R. T. (2024). Detection of <em>Phakopsora pachyrhizi</em> Infestation in Soybean via Hyperspectral Imaging and Data Analysis. Preprints. https://doi.org/10.20944/preprints202410.0142.v1
Chicago/Turabian Style
Thomas, S., Andreas Zell and Ralf Thomas Voegele. 2024 "Detection of <em>Phakopsora pachyrhizi</em> Infestation in Soybean via Hyperspectral Imaging and Data Analysis" Preprints. https://doi.org/10.20944/preprints202410.0142.v1
Abstract
Phakopsora pachyrhizi, the causative agent of the Asian Soybean Rust, is one of the most prominent causes of yield loss in soybean production worldwide. In this study, a combination of hyperspectral imaging with advanced data analysis methodology is shown to accurately detect soybean rust symptoms in early stages and differentiate them from other factors at leaf scale. Data analysis was performed with neural networks based on training data from a small subset of the respective experiments and compared with classical machine learning methods for its accuracy and early disease detection capabilities. Results show that even with a comparably small subset of training data the analysis through neural networks can outperform classical machine learning methods in a complex dataset with high variability in the designated classes. In two separate experiments it was possible to accurately detect disease in inoculated leaves before symptoms were visible to the human eye. These results do not only show the potential of this methodology for precision farming applications, but also allow accurate and objective assessment of symptom severity and development for scientific studies and resistance breeding.
Biology and Life Sciences, Agricultural Science and Agronomy
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
