Tomato Plant Disease Classification with Pesticide Suggestion Using Yolov8 Algorithm

Tomato plant diseases can adversely affect agricultural output, leading to significant financial losses and reducing global food security. If neglected, these illnesses can also reduce yields and raise farmer costs, disrupting the supply chain and raising consumer prices. Traditional disease detection methods rely on agronomists visually assessing plants, physical inspections, and specialist knowledge. Due to overlapped symptoms, the traditional method is time-consuming, prone to human error, and usually results in a misdiagnosis. Machine learning-based techniques are now practical choices for automating the detection of plant leaf diseases. The proposed deep learning-based modified YOLOv8 model accurately analyzes visual patterns and symptoms from a Plant Village data set to achieve training precision of 94% and validation accuracy of 91%. The proposed deep learning classification model evaluates the accuracy, loss and F1 score parameters. A web-based interface incorporating the model predictions enables farmers to input plant photos and get immediate evaluation and pesticide recommendations.