Citation

Rafay, Muhammad and Lim, Sin Liang (2025) Deep Learning-Based Tomato Crop Health Monitoring Using ResNet101V2. In: 2025 Multimedia University Engineering Conference, MECON 2025, 21 July 2025 - 23 July 2025, Cyberjaya, Malaysia.

Text 48.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Plant diseases pose a critical threat to global food security, contributing to an estimated 20–40% reduction in agricultural output. Traditional detection methods are labor-intensive, slow, and error-prone, often delaying effective intervention. This study proposes an automated tomato crop health monitoring system using deep learning, specifically the ResNet101V2 architecture with transfer learning, to detect both foliar and fruit diseases. The model was trained and evaluated on three datasets: the PlantVillage dataset with approximately 16,000 annotated tomato leaf images across ten classes, a 7226-image fruit dataset with four categories (Unripe, Ripe, Old, Damaged), and a custom three-class fruit dataset (Reject, Ripe, Unripe) containing 2400 images. Preprocessing techniques such as contrast enhancement, sharpening, and normalization were applied to improve feature extraction, while augmentations like rotation, zoom, and brightness adjustment enhanced model robustness. Performance was measured using Accuracy, Precision, Recall, and F1-score. The ResNet101V2 model achieved 93.29% accuracy on the leaf dataset, 97.65% on the four-class fruit dataset within 60 epochs, and 92.50% on the three-class dataset using only preprocessing—surpassing augmentation-based results. These findings demonstrate the model’s robustness, efficiency, and potential for practical deployment in intelligent crop health monitoring systems.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Deep learning, Plant disease detection
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics S Agriculture > S Agriculture (General)
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	18 Mar 2026 04:32
Last Modified:	18 Mar 2026 05:26
URII:	http://shdl.mmu.edu.my/id/eprint/15538

Downloads

Edit (login required)