Predicting Diseases of Mango Leaves Using ResNet50 and CNN

Mango leaf disease presents a significant challenge to mango trees worldwide, with untreated infections leading to severe leaf damage and potential tree mortality. Early detection is crucial to mitigating the effects of these diseases and preventing their spread. Our dataset includes 4,000 images categorized into eight classifications of mango leaf diseases, with 3,200 images for training and 800 for validation. This study employs a Convolutional Neural Network (CNN) Transfer Learning Model, specifically ResNet50, trained with carefully tuned hyperparameters to achieve an accuracy of 97%. This innovative approach significantly reduces the time and effort required for traditional inspections, offering a precise and automated solution for disease detection. By improving diagnosis and supporting sustainable farming practices, this research contributes to the development of scalable solutions for managing mango crops effectively.