Classification of Tomato Plant Diseases by Detecting Leaf Using YOLOv5 Models

Loh, Kai Yang (2024) Classification of Tomato Plant Diseases by Detecting Leaf Using YOLOv5 Models. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text Full Text-Loh Kai Yang.pdf Restricted to Registered users only Download (13MB)

Abstract

The purpose of this research is to assess the effectiveness of YOLOv5 models in classifying tomato plant diseases by detecting them on the leaves. The ultimate goal is to develop a system that can automatically recognize and identify diseases in tomato plants. The YOLOv5 model showed exceptional performance by finishing its training in only 0.314 hours. The precision achieved was remarkable, as both "Early Blight" and "Late Blight" had a precision score of 100%. The high recall rates observed in all classes demonstrate its effectiveness in identifying important examples. The accuracy of the model is highlighted by its mAP50 score of 98.4% for the "all" class. In summary, this study proposes potential real-world uses for classifying diseases and managing crops in the field of agriculture. The model showed consistent improvement throughout the training process, which makes it a dependable option for detecting tomato leaf diseases. This, in turn, enhances agricultural efficiency by utilising computer vision and AI analysis. The detection process resulted in a match of 92.86% in only 71.7 milliseconds.