Online First

 Preview Page

Research Article

Standardized Image-Based Phenotyping of Tomato Bacterial Wilt Symptom Severity using Deep Learning to Support Resistance Screening
Junesung Lee1
,
You-Jin Lee1
,
Ji-Su Kwon1
,
Seon-In Yeom1*
1Department of Horticulture, Division of Applied Life Science (BK21 Four Program), Institute of Agriculture & Life Science, Gyeongsang National University, Jinju 52828, Korea
*Corresponding Author
PDF XML Citation
Abstract

Bacterial wilt, caused by the soil-borne bacterium Ralstonia pseudosolanacearum, severely limits tomato production, making objective and reproducible disease phenotyping essential for resistance breeding. In this study, a convolutional neural network (CNN) image classification model based on transfer learning with EfficientNetB0 was developed to classify tomato bacterial wilt severity under controlled inoculation and imaging conditions to support standardized resistance screening. Four susceptible tomato cultivars were inoculated with R. pseudosolanacearum strain SL341 using drench and petiole methods to generate diverse wilt symptoms. A total of 1,500 standardized multi-view images from 776 individual plants were categorized into low-severity, intermediate-severity, and severe-wilt classes. To prevent data leakage, images were partitioned at the plant level into training, validation, and independent test sets at an approximate ratio of 64:16:20. On-the-fly image augmentation was applied during training to improve generalization. The final model correctly classified 288 of 300 test images, achieving 96.0% overall accuracy. Performance was highest for severe symptoms and lowest for the intermediate class, indicating greater difficulty in distinguishing transitional disease stages. A Grad-CAM analysis showed that predictions were based on biologically relevant plant regions rather than background features. Training and validation curves also showed stable convergence without substantial overfitting. These results demonstrate that deep-learning-assisted image classification can provide an objective, reproducible auxiliary tool for standardized symptom scoring, thereby relieving the phenotyping bottleneck in tomato bacterial wilt resistance screening.

Keywords
convolutional neural network
disease classification
machine learning
Ralstonia pseudosolanacearum
Solanum lycopersicum
References
1

Abade A, Ferreira PA, de Barros Vidal F (2021) Plant diseases recognition on images using convolutional neural networks: A systematic review. Comput Electron Agric 185:106125. https://doi.org/10.1016/j.compag.2021.106125

10.1016/j.compag.2021.106125
2

Allen C, Prior P, Hayward A (2005) Bacterial wilt disease and the Ralstonia solanacearum species complex. APS Press, St. Paul, MN, USA

3

Baek Y, Sul S, Cho YY (2023) Estimation of Days after Transplanting using an Artificial Intelligence CNN (Convolutional Neural Network) Model in a Closed-type Plant Factory. Hortic Sci Technol 41:81–90. https://doi.org/10.7235/HORT.20230008

10.7235/HORT.20230008
4

Bock CH, Chiang KS, Del Ponte EM (2022) Plant disease severity estimated visually: a century of research, best practices, and opportunities for improving methods and practices to maximize accuracy. Trop Plant Pathol 47:25–42. https://doi.org/10.1007/s40858-021-00439-z

10.1007/s40858-021-00439-z
5

Furbank RT, Tester M (2011) Phenomics–technologies to relieve the phenotyping bottleneck. Trends Plant Sci 16:635–644. https://doi.org/10.1016/j.tplants.2011.09.005

10.1016/j.tplants.2011.09.005
6

Gao Y, Wang L, Liu R, Tian J, Cai K (2023) Physiological response and proteomic profiling of biochar-induced tomato resistance to bacterial wilt. Sci Hortic 317:112055. https://doi.org/10.1016/j.scienta.2023.112055

10.1016/j.scienta.2023.112055
7

Hayward A (1991) Biology and epidemiology of bacterial wilt caused by Pseudomonas solanacearum. Annu Rev Phytopathol 29:65–87. https://doi.org/10.1146/annurev.py.29.090191.000433

10.1146/annurev.py.29.090191.000433
8

Jafar A, Bibi N, Naqvi RA, Sadeghi Niaraki A, Jeong D (2024) Revolutionizing agriculture with artificial intelligence: plant disease detection methods, applications, and their limitations. Front Plant Sci 15:1356260. https://doi.org/10.3389/fpls.2024.1356260

10.3389/fpls.2024.135626038545388PMC10965613
9

Jelali M (2024) Deep learning networks-based tomato disease and pest detection: a first review of research studies using real field datasets. Front Plant Sci 15:1493322. https://doi.org/10.3389/fpls.2024.1493322

10.3389/fpls.2024.149332239524561PMC11543466
10

Kang WH, Lee J, Koo N, Kwon JS, Park B, Kim YM, Yeom SI (2022) Universal gene co-expression network reveals receptor-like protein genes involved in broad-spectrum resistance in pepper (Capsicum annuum L.). Hortic Res 9:uhab003. https://doi.org/10.1093/hr/uhab003

10.1093/hr/uhab00335043174PMC8968494
11

Kim N, Kang WH, Kwon JS, Kim HJ, Park B, Yeom SI (2023) Optimization of the TRV-based Gene Silencing Efficiency and Growth Condition on Solanum lycopersicum var.‘Hawaii7996’Resistant to Bacterial Wilt Disease. Hortic Sci Technol 41:429–436. https://doi.org/10.7235/HORT.20230039

10.7235/HORT.20230039
12

Kim N, Lee J, Yeom SI, Kang NJ, Kang WH (2024) The landscape of abiotic and biotic stress-responsive splice variants with deep RNA-seq datasets in hot pepper. Sci Data 11:381. https://doi.org/10.1038/s41597-024-03239-7

10.1038/s41597-024-03239-738615136PMC11016105
13

Kwon JS, Lee J, Shilpha J, Jang H, Kang WH (2024a) The landscape of sequence variations between resistant and susceptible hot peppers to predict functional candidate genes against bacterial wilt disease. BMC Plant Biol 24:1036. https://doi.org/10.1186/s12870-024-05742-w

10.1186/s12870-024-05742-w39482582PMC11529287
14

Kwon JS, Nam JY, Yeom SI, Kang WH (2021) Leaf-to-Whole Plant Spread Bioassay for Pepper and Ralstonia solanacearum Interaction Determines Inheritance of Resistance to Bacterial Wilt for Further Breeding. Int J Mol Sci 22. https://doi.org/10.1101/2021.01.27.428365

10.1101/2021.01.27.428365
15

Kwon JS, Shilpha J, Lee J, Yeom SI (2024b) Beyond NGS data sharing for plant ecological resilience and improvement of agronomic traits. Sci Data 11:466. https://doi.org/10.1038/s41597-024-03305-0

10.1038/s41597-024-03305-038719829PMC11079010
16

LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521:436–444. https://doi.org/10.1038/nature14539

10.1038/nature14539
17

Lee J, Kim Y, Kwon JS, Yeom SI, Kang WH (2026) Convolutional Neural Network-Based Image Classification of Bacterial Wilt Symptoms in Pepper (Capsicum annuum L.). Hortic Sci Technol 44. https://doi.org/10.7235/HORT.20260022

10.7235/HORT.20260022
18

Lee SJ, Shin MH, Jayasooriya LSH, Wijethunga WMUD, Lee SK, Cho JG, Jang SH, Cho BK, Kim JG (2024) Prediction of Fruit Maturity of ‘Mihwang’ Peach using Deep Learning. Hortic Sci Technol 42:80–93. https://doi.org/10.7235/HORT.20240007

10.7235/HORT.20240007
19

Leite D, Brito A, Faccioli G (2024) Advancements and outlooks in utilizing Convolutional Neural Networks for plant disease severity assessment: A comprehensive review. Smart Agric Technol 9:100573. https://doi.org/10.1016/j.atech.2024.100573

10.1016/j.atech.2024.100573
20

Mahmood ur Rehman M, Liu J, Nijabat A, Faheem M, Wang W, Zhao S (2024) Leveraging convolutional neural networks for disease detection in vegetables: A comprehensive review. Agronomy 14:2231. https://doi.org/10.3390/agronomy14102231

10.3390/agronomy14102231
21

Ramanjot, Mittal U, Wadhawan A, Singla J, Jhanjhi N, Ghoniem RM, Ray SK, Abdelmaboud A (2023) Plant disease detection and classification: A systematic literature review. Sensors 23:4769. https://doi.org/10.3390/s23104769

10.3390/s2310476937430683PMC10223612
22

Rivera-Zuluaga K, Hiles R, Barua P, Caldwell D, Iyer-Pascuzzi AS (2023) Getting to the root of Ralstonia invasion. In Semin Cell Dev Biol, Vol 148. Elsevier, pp 3–12. https://doi.org/10.1016/j.semcdb.2022.12.002

10.1016/j.semcdb.2022.12.002
23

Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2017) Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proc. IEEE Int. Conf. Comput. Vis., pp 618–626. https://doi.org/10.1109/ICCV.2017.74

10.1109/ICCV.2017.74
24

Shilpha J, Kang WH (2025) Molecular and genomic insights into viral resistance in Capsicum spp.: pathogenesis, defense mechanisms, and breeding innovations. Front Plant Sci 16:1716114. https://doi.org/10.3389/fpls.2025.1716114

10.3389/fpls.2025.171611441404141PMC12703788
25

Shilpha J, Lee J, Kwon JS, Lee HA, Nam JY, Jang H, Kang WH (2024) An improved bacterial mRNA enrichment strategy in dual RNA sequencing to unveil the dynamics of plant-bacterial interactions. Plant Methods 20:99. https://doi.org/10.1186/s13007-024-01227-x

10.1186/s13007-024-01227-x38951818PMC11218159
26

Tan M, Le Q (2019) Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning, Vol 97. PmLR, pp 6105–6114

27

Ueyama T, Tsuzuki M, Tateda S, Terazawa Y, Ikeuchi A, Kiba A, Ohnishi K, Hikichi Y (2026) Contribution of a Novel TetR/AcrR Family Transcriptional Regulator RalT of Ralstonia pseudosolanacearum Strain OE1‐1 to the Fine‐Tuning of Its Virulence. MicrobiologyOpen 15:e70229. https://doi.org/10.1002/mbo3.70229

10.1002/mbo3.7022941724708PMC12928067
28

Vásconez JP, Vásconez IN, Moya V, Calderón-Díaz M, Valenzuela M, Besoain X, Seeger M, Cheein FA (2024) Deep learning-based classification of visual symptoms of bacterial wilt disease caused by Ralstonia solanacearum in tomato plants. Comput Electron Agric 227:109617. https://doi.org/10.1016/j.compag.2024.109617

10.1016/j.compag.2024.109617
29

Wang G, Sun Y, Wang J (2017) Automatic image‐based plant disease severity estimation using deep learning. Comput Intell Neurosci 2017:2917536. https://doi.org/10.1155/2017/2917536

10.1155/2017/291753628757863PMC5516765
30

Wu S, Su H, Gao F, Yao H, Fan X, Zhao X, Li Y (2023) An insight into the prevention and control methods for bacterial wilt disease in tomato plants. Agronomy 13:3025. https://doi.org/10.3390/agronomy13123025

10.3390/agronomy13123025
Information
  • Publisher :KOREAN SOCIETY FOR HORTICULTURAL SCIENCE
  • Publisher(Ko) :한국원예학회
  • Journal Title :Horticultural Science and Technology
  • Journal Title(Ko) :원예과학기술지
  • Received Date : 2026-09-15
  • Revised Date : 2026-09-22
  • Accepted Date : 2026-10-08
  • DOI :https://doi.org/10.7235/HORT.20260049
Journal Informaiton Horticultural Science and Technology Horticultural Science and Technology
Online Submission
  • scopus
  • NRF
  • KOFST
  • crossref crossmark
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close