Identification of Mango Leaf Diseases Using a Convolutional Neural Network

Abstract

Mango cultivation contributes substantially to agricultural production in many regions, yet mango trees are affected by several diseases that can reduce crop yield and quality. Timely identification of these diseases is therefore important for effective disease management and sustainable production. In this study, a Convolutional Neural Network (CNN) is applied to identify diseases from mango leaf images. The developed CNN achieved an accuracy of 96.88% when evaluated using 2000 high-quality images obtained from the Mendeley dataset. The image collection represents Gall Midge, Powdery Mildew, Sooty Mould, and healthy leaves. The obtained accuracy, recall, and F1-score indicate that the model can distinguish diseased leaves with relatively few false positive and false negative classifications. The findings demonstrate the potential of deep-learning-based image analysis as a tool for supporting plant disease management and, consequently, agricultural productivity and food security.