Detecting Mango Leaf Diseases Using Google Teachable Machine for Sustainable Agriculture

Mango cultivation is a critical component of agricultural economies, but it faces substantial challenges due to various leaf diseases that can drastically reduce crop quality and yield. Traditional methods of disease detection, which rely on manual visual inspection, are often slow, labor-intensive, and prone to inaccuracies, especially in early disease stages. This study presents an automated mango leaf disease recognition system developed using Google Teachable Machine, an accessible custom AI development platform. The system was trained on a dataset of 4000 images and is capable of detecting seven common mango leaf diseases: Anthracnose, Bacterial Canker, Cutting Weevil, Die Back, Gall Midge, Powdery Mildew, Sooty Mould, as well as identifying healthy leaves. The model achieved an impressive overall accuracy of 99.6%, with high precision, recall, and F1 scores across all disease categories. Compared to similar studies, this system demonstrated superior performance, making it a reliable tool for early and accurate disease detection. The adoption of this system can empower farmers with a cost-effective and efficient solution for disease management, reducing the reliance on excessive pesticide use and supporting more sustainable agricultural practices.