MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641097735 A) filed by G. Pullaiah College Of Engineering And Technology on August 12, 2026, for Blind-Friendly Medicine Identification Using Cnn And Ocr.
Inventors include S. Fowzia Sultana; Boda Eramma; T. Ranjitha Devi; and Dr. N. Geetha Rani.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: The development of AI in healthcare offers new capabilities for a variety of tasks, such as drug identification and extraction of important information from prescriptions. The project aims to develop a system that identifies drugs using images by applying CNN and OCR. The process begins with data collection from prescription images, including colour and grayscale transformations to ensure consistency. OCR is used to extract information from images, such as drug names, dosage instructions, and expiration dates. Specialized extraction techniques, such as grey-level co-occurrence matrix (GLCM) and mean difference model, are used to represent important images to improve classification accuracy. The extracted features were fed into a CNN-based model that was trained to classify drug images into specific categories. The performance of the model was measured with different performance metrics like accuracy, loss, and completion time. The output of the system is presented to users through a web application developed with Flask and Streamlit frameworks. The web interface allows users to upload images of prescriptions and receive text and audio-video descriptions of prescription information. The system is intended for both doctors, the visually impaired. This work not only helps detect narcotics but also provides effective solutions for providing important information about narcotics. The system combines machine learning, OCR, and web development to prove a powerful tool for improving healthcare processes.
Disclaimer: Curated by HT Syndication.
