International Journal on Science and Technology

E-ISSN: 2229-7677   •   Impact Factor: 9.88

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Call for Paper Volume 17 Issue 4 October-December 2026 Submit your research before last 3 days of December to publish your research paper in the issue of October-December.

AI- Powered Visual Assistant for Text Reading and Indian Currency Recognition to Visually Impaired

Author(s) Ms. Sneha Kumari, Prof. Sujitha S, Ms. Puja Sharma
Country India
Abstract Visually impaired individuals often face challenges in reading printed text and identifying currency. This project introducesan Android-based AI vision assistant that provides audio and haptic feedback through two modes: Read Mode and Currency Mode.Read Mode uses Google ML Kit offline OCR to extract printed text and convert it into speech using text-to-speech technology.Currency Mode identifies six Indian denominations: ₹10, ₹20, ₹50, ₹100, ₹200 and ₹500.The currency classification model uses MobileNetV2 transfer learning with 1,505 images, divided into training, validation and testingsets using stratified sampling. Data augmentation was applied to improve robustness against real-world image variations. Training wasperformed in two stages: initially, the MobileNetV2 base was frozen while the classification layers were trained. Subsequently, thefinal 30 layers were fine-tuned using a lower learning rate, with Batch Normalization layers kept frozen.
Keywords Assistive Technology, Visual Impairment, Optical Character Recognition (OCR), Indian Currency Recognition, MobileNetV2, Transfer Learning, Text-to-Speech, Android Application, Data Augmentation
Field Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In Volume 17, Issue 4, October-December 2026
Published On 2026-11-21

Share this