AI-Powered Medicinal Herb Identification System with Integrated Knowledge Retrieval

Abstract

The interest in the use of herbal medicines and Ayurveda is growing. There is a great need to develop techniques that can aid in the identification of medicinal plants because many plants resemble each other and are known by different names in different regions. This work presents a web-based medicinal herb identification system developed using Python, the Flask framework, and the PlantNet API. The system enables users to identify medicinal herbs either by uploading plant images or by searching using herb names. An integrated knowledge base provides scientific names, medicinal uses, and safety precautions for identified herbs. The proposed system was evaluated using multiple herb species under different image acquisition conditions and achieved an overall identification accuracy of approximately 90.6%. The platform offers a simple and user-friendly interface for students, researchers, healthcare practitioners, and the general public, making medicinal plant information more accessible and promoting awareness of traditional herbal knowledge.