ANN and CNN Based Ensemble Learning for Recognizing Renowned Medicinal Plants

This paper focuses on the automated identification and classification of medicinal plants using artificial neural networks and convolutional neural networks. This paper focuses on the ten most renowned leaves of Bangladesh. These are Bohera, Devilbackbone, Haritoki, Lemongrass, Nayontarat, Neem, Pathorkuchi, Thankuni, Tulsi, and Zenora. The paper emphasizes how these technologies could help scientific research, advance environmental preservation, and integrate traditional medicinal products with modern medicine. Since the beginning, medicinal plants have been used to heal illnesses because they contain the active compounds in medicine, which are derived from various plant sections. Natural and organic items are being used by customers more frequently, and therapeutic plants have a large economic impact. Bangladesh is a tropical nation with a wide variety of medicinal plants, and it is thought that roughly 1,500 of the estimated 5,000 plant species there have medicinal properties. The system uses three pre-trained models (VGG16, ResNet50, and ResNet152) as base models and adds custom layers to adapt them to the specific task. The system also uses a majority voting mechanism to combine the predictions from the different models. The system achieves an impressive accuracy of 99% on a dataset of 5000 images of 10 medicinal plant leaves. This work offers a potential method for the automated identification and classification of medicinal plants, which may have significant implications for medical care, scientific study, and environmental preservation.