Implementation of Hybrid ResNet50 and XGBoost Model for Wheat Plant Disease Classification
DOI:
https://doi.org/10.30871/jaic.v10i3.13004Keywords:
Image Classification, Wheat Leaf Disease, ResNet50, Wheat Plant, XGBoostAbstract
A hybrid artificial intelligence (AI) system was successfully developed in this study, combining the ResNet50 architecture as an image feature identifier and the XGBoost algorithm for final classification. This model was used to detect six disease variations using 5,505 wheat leaf photographs. To ensure model stability, rigorous testing was conducted using two methods: Stratified 5-Fold Cross-Validation on the entire data set and independent testing using 300 images (equally divided into 50 samples per class).The test results demonstrated very solid performance. The model recorded an average global accuracy of 94.66% (±0.21%) using the K-Fold method, and an accuracy of 94.67% and a Macro F1-Score of 0.9445 in the independent testing. Through confusion matrix mapping, the model successfully classified the Healthy, Black Rust, and Septoria categories perfectly (a score of 1.00). However, there was still a minor error in the case of eight samples being confused between Brown Rust and Yellow Rust due to the visual similarity of the orange-yellowish coloration early in the infection period. Furthermore, the Feature Importance assessment demonstrated that the XGBoost decision base is transparent (Explainable AI). This AI accurately focuses on clinical signs of plants such as chlorosis symptoms and spot texture, while ignoring background objects such as weeds and soil. This combination of methods creates a stable, efficient system with a response time of only 18.5 milliseconds per photo, and a biologically valid decision base.
Downloads
References
[1] A. Aryanti, N. Iryani, and K. Khairunnisa, “A Smart Recommendation System for Crop Seed Selection Using Gradient Boosting Based on Environmental and Geospatial Data,” J. Appl. Informatics Comput., vol. 9, no. 6, pp. 2965–2973, 2025, doi: 10.30871/jaic.v9i6.10249.
[2] G. D. Auliani and A. Akbar, “Analisis Komparasi Impor Gandum Di Indonesia Setelah Dan Sebelum Konflik Rusia-Ukraina,” J. Educ. Dev., vol. 12, no. 3, pp. 195–208, 2024.
[3] N. A. Cipta, “Analisis Faktor-Faktor Yang Mempengaruhi Impor Gandum Indonesia Volume impor gandum Indonesia ( dalam juta ton ),” J. Ekon. Manajemen, dan Akunt. Prodi Ekon. Pembanguan Fak. Ekon. dan Bisnis UPN Veteran Jawa Timur, vol. 9, no. 18, pp. 2321–2331, 2023.
[4] M. Ibrohim, F. Ardiyansyah, and R. Rudyarta, “Arus Jurnal Sosial dan Humaniora (AJSH) Strategi Indonesia dalam Menekan Ketergantungan Impor Gandum dan Kedelai (2010-2023): Perspektif Ketahanan Ekonomi Nasional,” J. Ardenjaya, vol. 5, no. 2, pp. 2411–2420, 2025, [Online]. Available: http://jurnal.ardenjaya.com/index.php/ajshhttps://jurnal.ardenjaya.com/index.php/ajsh
[5] A. Hura, “Penerapan Machine Learning Untuk Prediksi Produktivitas Pertanian Berbasis Data Cuaca Di Indonesia,” JIARTIN J. Ilmu Artif. Intell. Nusant., vol. 1, no. 1, pp. 1–8, 2025.
[6] M. Purba, “Klasifikasi Penyakit dan Hama Daun Padi Menggunakan Model ResNet50 pada Dataset AgroGuard AI,” vol. 09, no. 1, pp. 179–185, 2026.
[7] B. Paula and M. R. Pribadi, “Rice Leaf Disease Classification Using ResNet - 50 : A Comparative Study of Adam ,” vol. 6, no. 1, pp. 11–20, 2026.
[8] P. U. Poonam Gupta, Surendra Waghmare, “An Advanced CNN based System for Rice Plant Diseases Detection and classification in Precision Agriculture,” J. Electr. Syst., vol. 20, no. 4s, pp. 71–80, 2024, doi: 10.52783/jes.1823.
[9] D. Margarita, H. Maulana, and E. P. Mandyartha, “Klasifikasi Penyakit Daun Padi Menggunakan Support Vector Machine Berdasarkan Fitur Mendalam (Deep Feature),” JIPI (Jurnal Ilm. Penelit. dan Pembelajaran Inform., vol. 9, no. 4, pp. 2256–2270, 2024, doi: 10.29100/jipi.v9i4.5634.
[10] M. H. Abidi, N. Khare, P. D, H. Alkhalefah, and U. Umer, “EFS-XGBoost: A robust framework for precision classification of COVID-19 cases,” Intell. Data Anal., vol. 29, no. 1, pp. 171–185, 2025, doi: 10.3233/IDA-230854.
[11] K. Kusnaeni, I. R. Adriani, M. S. Hafid, A. B. Andy B, and M. E. Rizal, “Deepfake Image Classification Using ResNet50 Feature Extraction and XGBoost Learning Model,” J. Math. Comput. Stat., vol. 8, no. 2, pp. 258–268, 2025, doi: 10.35580/jmathcos.v8i2.8387.
[12] Syafrizal, M. Afdal, and R. Novita, “Sentiment Analysis of PLN Mobile Application Review Using Naïve Bayes Classifier and K-Nearest Neighbor Algorithm,” MALCOM Indones. J. Mach. Learn. Comput. Sci., vol. 4, no. January, pp. 10–19, 2024.
[13] Y. Chen, A. Aryanti, and Q. Hong, “A Multi-Model CNN Approach Using Pre-Trained Network for Improved Hand Gesture Recognition,” pp. 1–22, 2026.
[14] M. Khalid and M. A. Talukder, "A Hybrid Deep Multistacking Integrated Model for Plant Disease Detection," in IEEE Access, vol. 13, pp. 116037-116053, 2025, doi: 10.1109/ACCESS.2025.3583793.
[15] D. Mohnish Kumaar and S. Palani, "ResNet50 Integrated Vision Transformer for Enhanced Plant Disease Classification," 2024 3rd International Conference on Artificial Intelligence For Internet of Things (AIIoT), Vellore, India, 2024, pp. 1-6, doi: 10.1109/AIIoT58432.2024.10574771.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Aryanti Aryanti, Muhammad Aulia Dzikri, Ahmad Rifqi Nugraha, Khumairah Amira Sari, Dea Oktavia
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
