Detection and Classification of Corn Leaf Diseases Using Resnet-18

Abstract

Crop disease is a major problem for farmers, especially in Nepal, where many people depend on agriculture for their livelihood. Traditional methods of disease detection are often time-consuming, labor-intensive, and prone to human error. This study proposes an automated system for detecting and classifying corn leaf diseases using deep learning techniques. The ResNet-18 model is employed to detect and validate corn leaves and to classify diseases such as Common Rust, Blight, and grey Leaf Spot. The system achieved an accuracy of 91%, with a precision of 91.91% and a recall of 91.44%. By leveraging image processing and deep learning, this research provides a scalable and cost-effective solution for early disease detection in corn leaves, aiding farmers in timely intervention and crop management.