Sugarcane Disease Detection and Classification Using Deep Learning Techniques

This Agriculture remains a primary occupation for most people in India, with sugarcane being the predominant crop cultivated in the Sangli district. The cropping season and duration of sugarcane are influenced by the variety planted and the timing of seeding. However, sugarcane is highly sensitive to environmental factors and other conditions, leading to an increase in plant diseases. To report these tasks, this paper emphases on emerging an application powered by an efficient algorithm to deliver highly accurate results. While existing systems primarily classify sugarcane as either healthy or unhealthy, our approach takes it a step further by diagnosing specific diseases affecting the crop. This paper proposes the learning models: Convolutional Neural Network and RESNET50 with a multiclass classification and got a performance of 95.29% with RESNET50 model. The model categorizes sugarcane leaves into one of three classes: Healthy, Red Rot, Bacterial Blight. This system aims to significantly simplify disease detection and enhance decision-making for farmers, ultimately improving agricultural outcomes.