International Journal of Science, Engineering and TechnologyAuthors: Akash Sn, Adithiya S, Muni Praharsha M
Abstract: Visual impairment restricts independent recognition of everyday objects such as medicines and currency, leading to medication errors and financial dependency. This paper presents an AI-based companion system designed to assist visually impaired users in real-time medicine and currency recognition. The system utilizes computer vision and Convolutional Neural Networks (CNNs) to classify medicines based on packaging features and identify currency denominations using distinctive visual patterns. Optical Character Recognition (OCR) is integrated to extract drug names and dosage information from printed text. The system provides audio feedback through text-to-speech and supports voice commands for hands-free interaction. Designed for deployment on low-cost smartphone or embedded platforms, the proposed solution ensures affordability and portability. Experimental results demonstrate high recognition accuracy under varying environmental conditions. The system enhances user independence, reduces reliance on caregivers, and improves safety in medication management and financial transactions, highlighting the potential of AI-driven assistive technologies.
