Medicine Detection.pdf

The RECOGNIZATION HOLIC application uses deep learning techniques to identify medicines. It works by scanning the image of a medication label, extracting the text using optical character recognition (OCR) and TensorFlow, and then matching the text with a dataset of drug names and descriptions. The application provides reminders to take medications on time, prescription information, and aids chronic patients in taking multiple medications correctly.

Key features include:

Medicine recognition using deep learning
Optical character recognition (OCR) to extract text from scanned images
TensorFlow and Keras for character recognition training
Firebase for storing and retrieving data
Google Cloud Text-to-Speech API for multilingual voice descriptions
Reminder option with medicine photo and time to take medication

The application helps patients take their medications properly and at the right time, reducing the risk of taking the wrong medication. It is particularly useful for elderly people and those with chronic illnesses who may have difficulty managing their medications.

Related works include applications that use computer vision and OCR to track and trace wine bottles, recognize characters on metal workpieces, and extract information from invoices and receipts. Other studies have used deep learning models to recognize handwritten characters, offline Latin numerals, and alphabetic characters.

The proposed methodology involves:

Scanning the image of a medication label using a mobile camera
Extracting the text from the image using OCR and TensorFlow
Matching the text with a dataset of drug names and descriptions
Providing reminders and prescription information to the user
Using Firebase to store and retrieve data
Integrating Google Cloud Text-to-Speech API for multilingual voice descriptions.