Classification and Identification of Saccharum Officinarum Diseases using Deep Transfer Learning Computer Vision

This research paper explores the classification and identification of sugarcane leaf diseases using deep learning algorithms, specifically focusing on YOLOv11. Sugarcane is a critical crop for sugar and ethanol production, yet it is vulnerable to various diseases that can significantly affect yield and quality. This study targets five specific disease classes: healthy leaves, rust, red rot, yellow leaf diseases, and mosaic disease. By utilizing a comprehensive dataset of high-resolution images, the YOLOv11 model is trained to accurately detect and classify these diseases. The results indicate a substantial improvement in classification accuracy compared to traditional methods, demonstrating the potential of deep learning techniques in agricultural disease management. This research aims to provide farmers and agricultural stakeholders with an effective tool for early disease detection, ultimately contributing to better crop health and productivity. A deep transfer learning object identification model YOLOv11 is used to form sugarcane leaf diseases. As per the study results, we achieved mAP range of 80%-99.29%.