Ayurvedic, herbal, and other conventional medicines all make the extensive use of medicinal plants. These medicinal plants are considered as one of the richest bio source of traditional plant medicine. It is difficult to identify medicinal plants from leaf images obtained from various sources. An automated system for medicinal plant classification based on its leaves using image processing and deep neural networks will play a significant role in the field of plant medicine. This study presents a novel method for classifying and identifying medicinal plants using deep neural network and image recognition. The models used in our study includes ResNet 50, VGG 16, and Inception V3 and performed classification task on medicinal plants. The accuracy rate of the ResNet model was highest among the model used and shown 85%. A comprehensive experiment on various models for medicinal plant recognition demonstrate the effectiveness for correctly classifying medicinal plants.