Diseases in tomato plants are pathological conditions that affect tomato plant leaves. Common symptoms of this disease include changes in leaf color to yellowing and wilting, as well as the appearance of brown spots on the leaf surface. This can hinder the plant's photosynthesis process and ultimately affect the declining tomato harvest. Diseases in tomato leaves can be triggered by various factors, including fungal and viral infections. Traditionally, classifying diseases in tomato leaves can be done manually, where someone observes the characteristics of each form of leaf condition directly. However, due to human limitations and limited knowledge, manually checking or classifying tomato leaves becomes less efficient, especially when done in large quantities.
In previous research entitled Identification of Tomato Leaf Mold using the Densenet121 Model Based on Transfer Learning, measuring on the available dataset by dividing into 3 disease categories namely Leaf Mold, Healthy, and Other Diseases. It shows that the model gets 92.6% accuracy value, 93.3% precision value, and 93% recall value.
This Final Project research uses YOLOv8 as a method of analyzing and classifying diseases on tomato leaves. The YOLO model has achieved great success in the field of image processing, compared to the previous flagship models of the YOLO series (such as YOLOv5 and YOLOv7), YOLOv8 is an advanced and cutting-edge model that provides higher detection accuracy and speed.
This research will use Google Collab with Python programming language to run the simulation. System performance using YOLOv8 can be seen from the parameters of accuracy, precision, recall, F1-Score, and mAP.