Mango Leaf Disease Detection using VGG16 Convolutional Neural Network Model

• Published in: 2024 3rd International Conference for Innovation in Technology (INOCON) pp. 1 - 6
• Main Authors: Kaur, Gurjot, Sharma, Neha, Malhotra, Sonal, Devliyal, Swati, Gupta, Rupesh
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.03.2024
• Subjects: Agriculture; Convolutional Neural Network; Deep learning; Image Classification; Maintenance; Mango leaf Disease; Plantations; Productivity; Technological innovation; Training; VGG16
• DOI: 10.1109/INOCON60754.2024.10511415

This study proposes the VGG16 Convolutional Neural Network model. The paper suggests employing the VGG16 architecture to classify photos of mango leaf diseases into eight unique categories, inspired by recent achievements in automated image recognition. The model is subjected to intensive training using a dataset consisting of 4000 images. The training process involves five epoch, with each epoch utilizing a batch size of 64. The VGG16 model described in this research achieves a remarkable accuracy rate of 94%. The findings of this study highlight the considerable potential of utilizing advanced deep learning methods within the agricultural industry. Specifically, these techniques offer a reliable and effective framework for accurately and efficiently identifying diseases affecting mango leaves. The research findings have broader significance outside the academic realm, as they provide practical answers for farmers and other stakeholders in the field of agriculture. This empowers them with a crucial tool for detecting diseases at an early stage, enabling rapid intervention and ultimately ensuring the maintenance of mango tree health and productivity.