Original Articles
Early Access
Detecting bacterial pustules on soybean plants by hyperspectral imaging
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.
Published: 25 February 2026
133
Views
66
Downloads
Authors
Bacterial pustules are a major threat to soybean cultivation but are difficult to detect early because they manifest on the bottoms of leaves. In this study, hyperspectral imaging was applied to detect bacterial pustules on soybean plants. Images were preprocessed, and representative central wavelengths (i.e., bands) were identified through a two-sample t-test to calculate vegetation indices (VIs) for non-inoculated (i.e., control) and inoculated (i.e., treated) groups. Three machine-learning models were applied to classify infected soybean plants based on the VIs: partial least-squares discriminant analysis, support vector machine, and random forest (RF). The best classification performance was achieved by the RF model using five VIs with an overall accuracy (OA) of 0.89 and kappa coefficient (KC) of 0.77. The RF model also achieved an OA of 0.77 and KC of 0.55 when tested on a dataset before the expression of symptoms. The results of this study can potentially be applied to developing a multispectral image sensor that can be mounted on various platforms for the early detection of bacterial pustules on soybean crops.
Downloads
Download data is not yet available.
Citations
Bajwa SG, Rupe JC, Mason J, 2017. Soybean disease monitoring with leaf reflectance. Remote Sens 9:127. DOI: https://doi.org/10.3390/rs9020127
Bannari A, Morin D, Bonn F, Huete AR, 1995. A review of vegetation indices. Remote Sens Rev 13:95-120. DOI: https://doi.org/10.1080/02757259509532298
Bertoglio C, Moura Duin I, Netzel de Matos J, Rodrigues Ribeiro N, Pereira Leite R, Balbi-Peña MI, 2023. Comparative study of inoculation methods to determine aggressiveness of Xanthomonas citri pv. glycines isolates. Agronomy 13:1515. DOI: https://doi.org/10.3390/agronomy13061515
Breiman L, 2001. Random forests. Mach Learn 45:5-32. DOI: https://doi.org/10.1023/A:1010933404324
Congalton RG, 1991. A review of assessing the accuracy of classifications. Remote Sens Environ 37:35-46. DOI: https://doi.org/10.1016/0034-4257(91)90048-B
Constantin EC, Cleenwerck I, Maes M, Baeyen S, Van Malderghem C, De Vos P, Cottyn B, 2016. Genetic characterization of Xanthomonas strains. Plant Pathol 65:792–806. DOI: https://doi.org/10.1111/ppa.12461
Hancock DW, Dougherty CT, 2007. Relationships between blue- and red-based vegetation indices and leaf area and yield of alfalfa. Crop Sci 47:2497–2502. DOI: https://doi.org/10.2135/cropsci2007.01.0031
Hansen PM, Schjoerring JK, 2003. Reflectance measurement of canopy biomass and nitrogen status in wheat. Remote Sens Environ 86:542–553. DOI: https://doi.org/10.1016/S0034-4257(03)00131-7
Hartman GL, Rupe JC, Sikora EJ, Domier LL, Davis JA, Steffey KL, 2015. Compendium of soybean diseases and pests. St. Paul, American Phytopathological Society; pp. 56–59. DOI: https://doi.org/10.1094/9780890544754
Henning AA, Almeida ÁMR, Godoy CV, Seixas CDS, Yorinori JT, Costamilan LM, Dias WP, 2014. [Manual de identificação de doenças de soja].[Book in Portuguese]. Brasilia, EMBRAPA Editora.
Hong SJ, Kim YK, Jee HJ, Lee BC, Yoon YN, Park ST, 2010. Selection of bactericides for controlling bacterial pustule. Res Plant Dis 16:266-273. DOI: https://doi.org/10.5423/RPD.2010.16.3.266
Jinendra B, Tamaki K, Kuroki S, Vassileva M, Yoshida S, Tsenkova R, 2010. Near infrared spectroscopy for rapid diagnosis of virus-infected soybean. Biochem Biophys Res Commun 397:685–690. DOI: https://doi.org/10.1016/j.bbrc.2010.06.007
Jones SB, 1987. Bacterial pustule disease microscopy. Phytopathology 77:266–274. DOI: https://doi.org/10.1094/Phyto-77-266
Kang IJ, Kim KS, Beattie GA, Chung H, Heu S, Hwang I, 2021. Population genetics of X. citri pv. glycines in Korea. Plant Pathol J 37:652-661. DOI: https://doi.org/10.5423/PPJ.FT.11.2021.0164
Kang YS, Park JW, Jang SH, Song HY, Kang KS, Ryu CS, Kim GH, 2021. Spectral band selection for detecting fire blight. Korean J Agric Forest Meteorol 23:15-33.
Kauth RJ, Thomas GS, 1976. The tasselled cap transformation. Proc LARS Symposium; p. 159.
Lee SD, 1999. Occurrence and characterization of major plant bacterial diseases in Korea. PhD Thesis, Seoul National University.
Li S, Zheng Z, Wang Y, Chang C, Yu Y, 2016. Hyperspectral band selection using multiple classifiers. Pattern Recognit Lett 83:152-159. DOI: https://doi.org/10.1016/j.patrec.2016.05.013
Marcílio WE, Eler DM, 2020. Assessing SHAP values for feature selection. Proc 33rd SIBGRAPI Conf IEEE; pp. 340–347. DOI: https://doi.org/10.1109/SIBGRAPI51738.2020.00053
Mahlein AK, Steiner U, Hillnhütter C, Dehne HW, Oerke EC, 2012. Hyperspectral imaging for sugar beet disease symptoms. Plant Methods 8:3. DOI: https://doi.org/10.1186/1746-4811-8-3
Marston ZPD, Cira TM, Knight JF, Mulla D, Alves TM, Hodgson EW, et al., 2022. SVM classification of plant stress. J Econ Entomol 115:1557-1563. DOI: https://doi.org/10.1093/jee/toac077
Pearson RL, Miller LD, 1972. Remote spectral measurements as a method for determining plant cover. In: Proc 8th Int Symp Remote Sens Environ. Ann Arbor; pp. 1357–1379.
Peñuelas J, Filella I, Biel C, Serrano L, Savé R, 1993. Reflectance at 950–970 nm as water status indicator. Int J Remote Sens 14:1887–1905. DOI: https://doi.org/10.1080/01431169308954010
Purcell LC, Salmeron M, Ashlock L, 2014. Soybean growth and development. In: Arkansas soybean production handbook. Fayetteville, University of Arkansas Press.
Ray SS, Jain N, Arora RK, Chavan S, Panigrahy S, 2011. Hyperspectral potato disease detection. J Indian Soc Remote Sens 39:161-169. DOI: https://doi.org/10.1007/s12524-011-0094-2
Richardson AJ, Wiegand CL, 1977. Distinguishing vegetation from soil background. Photogramm Eng Remote Sens 43:1541-1552.
Sarhrouni E, Hammouch A, Aboutajdine D, 2012. Mutual-information-based hyperspectral feature selection. arXiv:1210.0052v1.
Schaad NW, 2008. Emerging plant pathogenic bacteria. In: Fatmi M. (Ed.), Pseudomonas syringae pathovars and related pathogens. Berlin, Springer. DOI: https://doi.org/10.1007/978-1-4020-6901-7_38
Shea Z, Singer M, Zhang B, 2020. Soybean production and improvement. London, IntechOpen Publ. DOI: https://doi.org/10.5772/intechopen.91778
Skoneczny H, Kubiak K, Spiralski M, Kotlarz J, Mikiciński A, Puławska J, 2020. Fire blight detection in apple. Remote Sens 12:2101. DOI: https://doi.org/10.3390/rs12132101
Vincini M, Frazzi E, D’Alessio P, 2008. Canopy-scale chlorophyll vegetation index. Precis Agric 9:303–319. DOI: https://doi.org/10.1007/s11119-008-9075-z
Wang X, Zhang X, Zhou G, 2017. Rice disease detection using NIR spectra. J Indian Soc Remote Sens 45:785–794. DOI: https://doi.org/10.1007/s12524-016-0638-6
Wei X, Johnson MA, Langston DB, Mehl HL, Li S, 2021. Identifying optimal wavelengths for disease signatures. Remote Sens 13:2833. DOI: https://doi.org/10.3390/rs13142833
Widodo A, Yang BS, 2007. SVM for machine fault diagnosis. Mech Syst Signal Process 21:2560-2574. DOI: https://doi.org/10.1016/j.ymssp.2006.12.007
Wrather A, Koenning S, 2009. Effects of diseases on soybean yields. Plant Health Prog 10:24. DOI: https://doi.org/10.1094/PHP-2009-0401-01-RS
Yang C, Everitt JH, Bradford JM, Murden D, 2004. Airborne hyperspectral imagery for cotton yield variability. Precis Agric 5:445–461. DOI: https://doi.org/10.1007/s11119-004-5319-8
Zhang N, Yang G, Pan Y, Yang X, Chen L, Zha, C, 2020. Hyperspectral-based plant disease detection review. Remote Sens 12:3188. DOI: https://doi.org/10.3390/rs12193188
CRediT authorship contribution
All authors made a substantive intellectual contribution, read and approved the final version of the manuscript and agreed to be accountable for all aspects of the work.
Supporting Agencies
Rural Development Administration, Republic of KoreaData Availability Statement
The datasets used and analyzed during the current study are available upon reasonable request from the corresponding author.
How to Cite
“Detecting bacterial pustules on soybean plants by hyperspectral imaging” (2026) Journal of Agricultural Engineering [Preprint]. doi:10.4081/jae.2026.1941.
Copyright (c) 2026 the Author(s)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
