Published in: Journal of Computer Science and Engineering in Innovations and Research (ISSN No: 3049-1762 online)
Publication Date: May 15, 2025
Mango cultivation is one of the most important agricultural activities in West Bengal, India, and the economy of the region depends on it to a great extent. However, Mango trees are prone to various diseases, which can severely reduce yield and quality. Therefore, early, and accurate detection of these diseases is essential for taking corrective measures on time and reducing losses.
ResNet9 architecture is proposed for the detection and classification of three major Mango leaf diseases found in West Bengal. An extensive dataset of images of Mango leaves, including healthy and diseased leaves, was created by collecting samples from the field and other publicly available sources. The ResNet9 model was modified and trained on this dataset, showing high accuracy in distinguishing between healthy and diseased leaves and classifying the specific diseases.
Result shows that the ResNet9-based approach is effective for automated disease identification and thus holds great promise for real-time, on-field diagnostics. This methodology has the potential to empower local Mango farmers with an affordable and efficient tool for early disease detection, upgrading their crop management practices for better productivity in the West Bengal Mango industry.
Leaf disease detection, Convolutional Neural Network (CNN), Deep Learning (DL), Residual Network (ResNet)