Sustainable Plant Disease Detection with CNN: Achieving High Accuracy in Multi-Crop Classification

Crop production has a major impact in contributing to the agricultural sector and the overall economy of every nation. Diagnosing plant diseases is a vital factor in sustaining a country with a strong and sustainable the economy of agriculture. To ensure a productive and sustainable agricultural industry, moreover in order to avoid unnecessary waste of resources and funds, it is essential to detect plant diseases promptly and efficiently. Many plant diseases cause crop growers to lose a substantial amount of money each year. Deep learning can greatly assist farmers in promoting sustainability by enabling early detection of diseases in plant leaves, which helps prevent crop loss. In this study dataset is collected from Plant Village which includes images of plant leaves from 15 different categories across three plant types: potato, tomato, and bell pepper. The proposed model accuracy was 98.47% for potato and 99.3% for bell pepper, 97.72% for tomato, and 98.57% for detecting diseases across all three plants combined. The results indicate that proposed model performs effectively in detecting and classifying plant leaf diseases with a high degree of accuracy. Comparative findings show that the proposed approach outperforms other models in accuracy, contributing to a more sustainable approach to agriculture.