EDANet: A Novel Architecture Combining Depthwise Separable Convolutions and Hybrid Attention for Efficient Tomato Disease Recognition

Ibrahim ; Yusuf ; O. Momoh ; Muyideen ; O. Shobowale ; Kafayat ; Mukhtar Abubakar ; Zainab ; Yahaya ; Basira

Journal of Computing Theories and Applications 📅 2025 Vol 3, No 2 🔗 DOI: 10.62411/jcta.14620

Abstract

Tomato crop yields face significant threats from plant diseases, with existing deep learning solutions often computationally prohibitive for resource-constrained agricultural settings; to address this gap, we propose Efficient Disease Attention Network (EDANet), a novel lightweight architecture combining depthwise separable convolutions with hybrid attention mechanisms for efficient Tomato disease recognition. Our approach integrates channel and spatial attention within hierarchical blocks to prioritize symptomatic regions while utilizing depthwise decomposition to reduce parameters to only 104,043 (multiple times smaller than MobileNet and EfficientNet). Evaluated on ten tomato disease classes from PlantVillage, EDANet achieves 97.32% accuracy and exceptional (~1.00) micro-AUC, with perfect recognition of Mosaic virus (100% F1-score) and robust performance on challenging cases like Early blight (93.2% F1) and Target Spot (93.6% F1). The architecture processes 128×128 RGB images in ~23ms on standard CPUs, enabling real-time field diagnostics without GPU dependencies. This work bridges laboratory AI and practical farm deployment by optimizing the accuracy-efficiency tradeoff, providing farmers with an accessible tool for early disease intervention in resource-limited environments.

Keywords

Attention mechanisms; Computer vision; Deep learning; Depthwise separable convolutions; Lightweight models; Plant disease classification; Precision agriculture; Tomato disease detection

How to Cite

Ibrahim, et al. (2025). EDANet: A Novel Architecture Combining Depthwise Separable Convolutions and Hybrid Attention for Efficient Tomato Disease Recognition. Journal of Computing Theories and Applications, 3(2). https://doi.org/10.62411/jcta.14620

Ibrahim, Yusuf; O. Momoh, Muyideen; O. Shobowale, Kafayat; Mukhtar Abubakar, Zainab; Yahaya, Basira, "EDANet: A Novel Architecture Combining Depthwise Separable Convolutions and Hybrid Attention for Efficient Tomato Disease Recognition," Journal of Computing Theories and Applications, vol. 3, no. 2, 2025.

Ibrahim, Yusuf; O. Momoh, Muyideen; O. Shobowale, Kafayat; Mukhtar Abubakar, Zainab; Yahaya, Basira. "EDANet: A Novel Architecture Combining Depthwise Separable Convolutions and Hybrid Attention for Efficient Tomato Disease Recognition." Journal of Computing Theories and Applications, vol. 3, no. 2, 2025.

Ibrahim, et al. (2025) 'EDANet: A Novel Architecture Combining Depthwise Separable Convolutions and Hybrid Attention for Efficient Tomato Disease Recognition', Journal of Computing Theories and Applications, 3(2). doi: 10.62411/jcta.14620.

Ibrahim, Yusuf; O. Momoh, Muyideen; O. Shobowale, Kafayat; Mukhtar Abubakar, Zainab; Yahaya, Basira. EDANet: A Novel Architecture Combining Depthwise Separable Convolutions and Hybrid Attention for Efficient Tomato Disease Recognition. Journal of Computing Theories and Applications. 2025;3(2).

An Attention-Enhanced CNN–RBF Framework for Network Intrusion Detection in Imbalanced Traffic

Kabura, Fabrice; Nsabimana, Thierry

A Lightweight Maize Leaf Disease Recognition Using PCA-Compressed MobileNetV2 Features and RBF-SVM

Abubakar, Mustapha; Ibrahim, Yusuf; Ajayi, Ore-Ofe; Saminu, Sani Saleh

Multimodal Deep Learning for Pneumonia Detection Using Wearable Sensors: Toward an Edge-Cloud Framework

Ibam, Emmanuel Onwako; Oluwagbemi, Johnson Bisi

Android Malware Detection Using Machine Learning with SMOTE-Tomek Data Balancing

Masari, Maryam Sufiyanu; Danladi, Maiauduga Abdullahi; Onyinye, Ilori Loretta; Tohomdet, Loreta Katok

Hybrid Real-time Framework for Detecting Adaptive Prompt Injection Attacks in Large Language Models

Prakash, Chandra; Lind, Mary; De La Cruz, Elyson

Integrating Fully Homomorphic Encryption and Zero-Knowledge Proofs for Efficient Verifiable Computation

Qureshi, UmmeAmmara; Doshi, Bhumika; More, Aditya; Joshi, Kashyap; Kumar, Kapil

Tren Sitasi Jurnal