<![CDATA[Herbal plants are a source of natural ingredients widely used in traditional medicine and alternative therapies. However, the process of identifying herbal plant types is still often done manually, potentially leading to errors, especially for the general public. The aim of this research is to implement an intelligent system capable of classifying herbal plant types based on leaf images, thereby helping to identify herbal plants independently while providing educational value. The research utilizes an image dataset of several types of herbal plant leaves that were processed and trained using Google Teachable Machine. This platform facilitates training machine learning models through an image-based approach without requiring complex programming knowledge. The dataset is divided into several classes according to the type of herbal plant, and the training process is carried out by varying parameters such as epoch, batch size, and learning rate to obtain the most optimal configuration. The results of this research create a web-based intelligent system application that can effectively classify herbal plant images, provide educational information to users, and can be further developed into a practical tool for herbal plant identification. The research also highlights several challenges, such as the need for a sufficiently large and diverse dataset, variations in lighting conditions and image backgrounds, and the model's ability to perform consistently under real-world conditions.]]>
Copyrights © 2025
