Ayurvedic leaf detection using Raspberry Pi and deep learning

Electronics and Communication Engineering

The accurate identification of medicinal plants is crucial for the efficacy and safety of Ayurvedic medicine. Traditional methods, reliant on manual expertise, are time-consuming, subjective, and not scalable. This paper presents an automated system for classifying Ayurvedic medicinal plants based on leaf images using deep learning techniques. The proposed system employs a pipeline comprising image acquisition, preprocessing (resizing, denoising, background segmentation), and classification via a Convolutional Neural Network (CNN) model. The model is trained on a curated dataset of leaf images from various Ayurvedic species. A key contribution is the development of a responsive web application that provides real-time classification, delivering results with associated confidence scores and detailed medicinal information in under 200 milliseconds. The system achieved a classification accuracy of over 95% on test data, demonstrating its potential as a reliable tool for botanists, Ayurvedic practitioners, and students. This work bridges traditional botanical knowledge with modern artificial intelligence, promoting accessibility and accuracy in medicinal plant identification.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.