Medicinal plants, cultivated and harvested across India for centuries, are plentiful in Indian forests, which are a amusing source of valued medicinal herbs. Due to their importance in healthcare, these plants have been the focus of wide research. However, the procedure of precisely categorizing medicinal plants remains time-consuming and entails a high level of proficiency. To shorten this task, deep learning techniques are proposed in this paper. The present study discovers the use of three models artificial Neural Networks, 2D Convolutional Neural Networks, and 3D Convolutional Neural Networks were applied to the combination of Medicinal Leaf Dataset and the Indian Medicinal Plant Dataset. The 3D CNN model outperformed the remaining by efficiently apprehending spatial and depth features of plant images. A thorough analysis of both successful and failed cases was accompanied, signifying the leverage of deep learning techniques to expressively improve Ayurvedic plant identification. Both qualitative and quantitative results and analysis shows the importance of the method.