Automated Detection and Quantification of Stomata

Abstract

This thesis presents an approach for the automated detection and quantification of stomata using machine learning techniques. The study focuses on employing the YOLOv8 model to analyse video data of leaf epidermal imprints, significantly improving the efficiency and accuracy of stomatal detection compared to traditional manual methods. The results highlight the model's ability to handle varying focal depths within video frames, ensuring consistent stomatal counts. Future research directions include expanding the dataset and incorporating advanced image analysis techniques to further enhance detection accuracy.