Performance Evaluation and Cross-Dataset Validation of Machine Learning Algorithms for Crop Disease Identification Using Rice and Maize Leaf Images

The present study introduced a comprehensive machine learning-based framework for crop disease identification from the rice and maize leaf images. DT, RF, SVM, KNN and the proposed model were evaluated through a comparative study, using a self-prepared dataset as well as a public dataset called PlantVillage. To evaluate the classification performance, accuracy, precision, recall (sensitivity) and specificity were used to provide a comprehensive assessment of each algorithm. The experimental results showed that the proposed model aims to have a consistent classification accuracy across all the datasets in comparison to the conventional machine learning classifiers and achieved classification accuracy of 97.42% on the self-prepared dataset and 98.36% on the PlantVillage dataset. To demonstrate the robustness and generalization ability of the proposed framework, it was validated across datasets, both in a controlled environment and in real-world agricultural settings. The developed system was also validated by the agricultural experts which in turn showed the practical reliability. The proposed approach provides an effective solution for an intelligent crop disease diagnosis and precision agriculture. The next steps include the incorporation of deep learning methods, extending the work to other disease/crop species, and the creation of a mobile or edge-based disease diagnosis system for field use, which is near real-time.  

1.    Ngugi, H. N., Akinyelu, A. A., & Ezugwu, A. E. (2024). Machine Learning and Deep Learning for Crop Disease Diagnosis: Performance Analysis and Review. Agronomy, 14(12), 3001. https://doi.org/10.3390/agronomy14123001 2.    Phadikar, S.; Sil, J.; Das, A.K. Classification of Rice Leaf Diseases Based on Morphological Changes. Int. J. Inf. Electron. Eng. 2012, 2, 460–463. 3.    Wang, Y.; Wang, Y.; Zhao, J. MGA-YOLO: A lightweight one-stage network for apple leaf disease detection. Front. Plant Sci. 2022, 13, 927424 4.    Annu Singla, Ashima Nehra, Kamaldeep Joshi, Ajit Kumar, Narendra Tuteja, Rajeev K. Varshney, Sarvajeet Singh Gill, Ritu Gill, Exploration of machine learning approaches for automated crop disease detection, Current Plant Biology, Volume 40, 2024, 100382, https://doi.org/10.1016/j.cpb.2024.100382. 5.    Chihaoui, N., Waleed, A., Mufleh, A.S.S. et al. A comprehensive review on AI-based crop disease detection using leaf image classification and explainable AI. Discov Appl Sci 8, 663 (2026). https://doi.org/10.1007/s42452-026-08684-0 6.    Thakur, P.S.; Sheorey, T.; Ojha, A. VGG-ICNN: A Lightweight CNN model for crop disease identification. Multimed. Tools Appl. 2023, 82, 497–520. 7.    Shandilya, G., Gupta, S., Mohamed, H. G., Bharany, S., Rehman, A. U., & Hussen, S. (2025). Enhanced Maize Leaf Disease Detection and Classification Using an Integrated CNN-ViT Model. Food science & nutrition, 13(7), e70513. https://doi.org/10.1002/fsn3.70513 8.    Dolatabadian, A., Neik, T.X., Danilevicz, M.F., Upadhyaya, S.R., Batley, J. & Edwards, D. (2025) Image-based crop disease detection using machine learning. Plant Pathology, 74, 18–38. Available from: https://doi.org/10.1111/ppa.14006 9.    Dey, B.; Haque, M.M.U.; Khatun, R.; Ahmed, R. Comparative performance of four CNN-based deep learning variants in detecting Hispa pest, two fungal diseases, and NPK deficiency symptoms of rice (Oryza sativa). Comput. Electron. Agric. 2022, 202, 107340. 10. Suresh Timilsina, Sandhya Sharma, Satoshi Kondo, Advancements in maize leaf disease detection, segmentation and classification: A review, Biosystems Engineering, Volume 255, 2025, 104162, https://doi.org/10.1016/j.biosystemseng.2025.104162.