Identification of Medicinal Plant Using Deep Learning Techniques

Abstract

The identification of medicinal plants plays a vital role in traditional medicine and botanical research, yet it often demands significant expertise and time. This study aims to simplify and improve the classification process by applying deep learning techniques to recognize 40 different species of medicinal plants. We investigate the effectiveness of Convolutional Neural Networks (CNN), MobileNet, and a hybrid model combining MobileNet with Recurrent Neural Networks (RNN). By training these models on a wide variety of plant images, we assess their performance based on metrics such as accuracy, precision, and recall. While the CNN serves as a strong baseline for plant image classification, MobileNet offers a more efficient solution suited for environments with limited resources. Additionally, the hybrid MobileNet+RNN model is explored to assess its potential in extracting sequential and contextual features. This work aims to enhance the accessibility and reliability of automated plant identification systems, supporting researchers, herbalists, and practitioners in their work and improving the overall process of medicinal plant classification.