Leveraging Inception V3 for Precise Early and Late Blight Disease Classification in Potato Crops

Abstract

The global significance of potato (Solanum tuberosum) as a staple food necessitates innovative approaches to safeguard its production against detrimental fungal pathogens, notably early and late blight diseases.These afflictions not only jeopardize crop health but also exacerbate economic pressures on agricultural stakeholders by diminishing yields and necessitating increased use of chemical treatments, which in turn degrades soil quality and elevates disease susceptibility.In response to these challenges, the deployment of Artificial Intelligence (AI) offers a transformative solution for the automatic detection of plant diseases, minimizing human intervention and maximizing diagnostic precision.This study employs the Inception V3 model, a deep learning algorithm, on a dataset comprising 2,152 potato leaf images, segmented into training and validation subsets at an 80:20 ratio.Through the application of image augmentation techniques such as scaling, rotating, and flipping, the diversity of the dataset was substantially enhanced, facilitating the model's capacity to distinguish between early blight, late blight, and healthy leaves with an unprecedented accuracy of 98.60%.This research not only underscores the efficacy of AI in revolutionizing agricultural disease management but also contributes to the broader discourse on sustainable farming practices by reducing reliance on chemical interventions.The findings advocate for the integration of AI-driven diagnostic systems in agricultural management, promising significant advancements in crop preservation and economic sustainability.

Topics & Concepts