Medicinal Plants Recognition Using Deep Learning
Yousef Sharrab, Dimah Al-Fraihat, Monther Tarawneh, Ahmad Sharieh
Abstract
The use of medicinal plants has been a longstanding practice in traditional medicine worldwide. Accurately identifying medicinal plants is crucial for determining their medicinal properties and potential applications. However, it can be a challenging task due to the complexity of their appearance. Variations in growth stage, lighting, and imaging conditions can make classification challenging, which limits the application of traditional methods for plant identification. This paper proposes a deep learning-based approach that uses a convolutional neural network (CNN) based on the VGG-16 model. With a dataset of 25,686 images, the CNN is capable of learning and representing complex features in images, enabling it to recognize and classify medicinal plants with high accuracy. The proposed approach can efficiently classify plants with different growth stages, lighting conditions, and imaging settings, providing a reliable tool for plant identification. We achieved an impressive recognition rate of 98%, demonstrating the feasibility of using deep learning techniques for accurate plant classification. The proposed approach has enormous potential for providing healthcare professionals and herbal medicine researchers with a reliable tool for identifying herbal plants. the study represents an essential advancement in the use of deep learning techniques for medicinal plant recognition, overcoming the challenges posed by their complex appearance. The proposed approach has farreaching implications and can significantly impact the field of herbal medicine research, enabling researchers and healthcare professionals to identify and classify medicinal plants more accurately