A Weighted Ensemble of EfficientNetV2 Variants with Optimizer Tuning for Tomato Leaf Disease Classification

Abstract

Tomatoes are one of the most widely grown horticultural crops but are highly susceptible to leaf diseases that could significantly affect yield and quality. Early and accurate disease detection is necessary to enhance crop productivity and food security. The present study presents a deep learning system for automatic tomato leaf disease classification using a weighted ensemble of EfficientNetV2 models. Seven variants of EfficientNetV2 (B0-B3, S, M, and L) were fine-tuned on an augmented version of the filtered PlantVillage dataset. To improve feature visibility under varying lighting conditions, images were preprocessed using contrast-limited adaptive histogram equalization. Five optimization algorithms, which include Adam, Adamax, AdamW, Nadam, and RMSProp, were tested to assess their impact on model convergence and generalization. Among individual models, Adam-optimized EfficientNetV2L performed the best in accuracy with a measure of 99.50%. For classifiability resilience improvement, 16 ensemble configurations (eight unweighted and eight weighted ensembles using different combinations of EfficientNetV2-S, M, and L variants and five optimizers) were explored, with the best weights discovered using brute-force grid search. The best-performing weighted ensemble achieved 99.89% in accuracy, precision, recall, and F1-score, which was evaluated on a held-out test set not used during training or validation, demonstrating its strong generalizability. The proposed framework offers a scalable and reliable solution for early tomato disease detection with significant potential for real-time plant health monitoring in precision agriculture.

Received: 10 September 2025 | Revised: 22 December 2025 | Accepted: 6 February 2026



Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.



Data Availability Statement

Data are available from the corresponding author upon reasonable request.



Author Contribution Statement

Aliyu Tetengi Ibrahim: Conceptualization, Methodology, Formal analysis, Writing – original draft. Ibrahim Hayatu Hassan: Conceptualization, Methodology, Formal analysis, Resources, Data curation, Writing – original draft. Mohammed Abdullahi: Validation, Formal analysis, Resources, Data curation, Writing – review & editing, Visualization. Abeer Rashad Mirdad: Conceptualization, Software, Validation, Investigation, Writing – original draft, Writing – review & editing, Supervision, Project administration. Muhammad I. Khan: Methodology, Validation, Investigation, Resources, Data curation, Writing – review & editing, Visualization, Project administration. Saeed Ali Bahaj: Conceptualization, Software, Writing – original draft. Fatima Khan Nayer: Investigation, Visualization, Supervision, Project administration.