Empowering Farmers: AI-Driven Potato Leaf Disease Identification and Actionable Solutions

Potato farming is vital for food security and livelihood worldwide, yet diseases like Early Blight, Late Blight, and Black Scurf pose significant threats to crop yield. Traditional disease detection methods are often slow, inaccurate, and inaccessible for many smallholder farmers. This paper presents an AI-driven system that leverages deep learning and computer vision to enable fast, accurate, and automated identification of potato leaf diseases. Utilizing a lightweight MobileNetV2 model with transfer learning, our approach ensures reliable classification even on low-resource devices with offline capability. Beyond detection, the system delivers clear, actionable treatment recommendations tailored for practical application by farmers. Designed with user-friendly interfaces and multilingual support, this solution significantly reduces dependency on expert intervention, fostering timely disease management and promoting sustainable farming practices. Field testing and user feedback demonstrate that the system effectively empowers farmers with accessible technology, helping to increase productivity and reduce crop loss. Keywords: Potato Leaf Disease Detection, Early Blight, Late Blight, Deep Learning, Convolutional Neural Networks (CNN), MobileNetV2, Transfer Learning, Computer Vision, Precision Agriculture, Smart Farming, Image Classification, Artificial Intelligence in Agriculture, Plant Disease Diagnosis, Streamlit Application