Summary
Tomato crops face big losses due to diseases, which affects food supply worldwide. The authors created a computer model that mixes two powerful ways to process images, called convolution and attention, to recognize tomato diseases quickly and accurately. They made the model use fewer parameters by using a special method called DeepSeek Mixture of Experts, which means it can run faster without losing how well it works. Their tests on popular tomato image datasets showed the model works very well with high accuracy and is much smaller than other models. This work helps make smart tools for farmers to detect diseases early with less computing power.
What this means in practice
- •For agricultural technology developers: Create lightweight apps that diagnose tomato diseases on low-power devices using efficient deep learning models.$Commercial implications: Enables development of portable agricultural diagnosis tools sold to farmers and agribusinesses for early disease detection.
- •For mobile app developers: Incorporate accurate, small-footprint plant disease recognition models into smartphone apps for everyday users.
Authors
Md Nadim Mahamood, Md Arif Shahriar, Md Shafi Ud Doula, Kamrul Hasan
Abstract
The world population is growing rapidly, and technology is improving in parallel. Meeting the huge demand for food for these 7 billion people not only depends on increasing food production but also on reducing food loss. Crop losses due to disease affect both the food supply and the financial and economic stability of a country. Tomatoes are among the top food-producing crops globally, and a significant portion of this production is lost due to disease. People have used Machine Learning techniques for feature extraction and early diagnosis of tomato diseases, and nowadays, Deep Learning-based models are widely used for disease recognition. However, most existing models are highly parameter-intensive, which increases the time required for training and inference. As a result, while lightweight models are more suitable for user-friendly applications, they often show a reduction in performance. To balance performance and model size, we propose CoAtNet-DeepMoE, a Convolution-Attention hybrid architecture for rich feature extraction, further enhanced with a DeepSeek Mixture of Experts to substantially reduce the number of parameters without sacrificing accuracy. We evaluate our model on both balanced and imbalanced datasets from Kaggle and PlantVillage, demonstrating robustness and achieving 99.80% accuracy, 99.80% precision, 99.80% recall, and 99.80% F1-score on Kaggle, and 99.83% accuracy, 99.85% precision, 99.76% recall, and 99.80% F1-score on PlantVillage, representing state-of-the-art performance with only 2.47M parameters. The source code will be available at https://github.com/nadimbrur/CoAt-MoE.