fairfolio
Saved Projects

AI Eye for the Visually Impaired Using Deep Learning Neural Network

AJAS · 2025

Muhammad Fawwaz · North Carolina Academy of Science

Overview

Visual impairment poses significant challenges globally, with millions facing blindness or severe visual impairments. While solutions like echolocation-based canes exist, affordability and accessibility remain challenging, especially in developing countries. This research addresses these challenges by focusing on supporting visually impaired individuals in their daily activities through object identification and recognition. The affordable, novel artificial Intelligence (AI) and machine learning (ML) image classification model integrates into a wearable assistive device, object detection conveyed through text-to-speech, and provides comprehensive details about the user's surroundings. Technical integration involves open-source hardware and software for cost-effective development, prioritizing portability to enhance independence and safety. The project comprises hardware and software design phases. Hardware setup involves configuring an edge computing device with essential peripherals and a camera module, enabling real-time object identification triggered by a push button. Software development utilizes TensorFlow for building the deep learning neural network model, trained on annotated images collected from various sources. Results indicate an 87% classification accuracy, with potential for further improvement. Real-time testing demonstrates promising accuracy, albeit with room for enhancement. Conclusions highlight the project's contribution to reducing device costs and providing affordable assistive technology. The rapid evolution of AI and edge computing holds promise for future enhancements. The "AI Eye" emerges as a beacon of hope for visually impaired individuals worldwide. Open-source design fosters future expansion opportunities, leveraging community support. This research aims to significantly improve the lives of visually impaired individuals through innovative technology.

Competition history

  • AJAS 2025 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

Source: AAAS Annual Meeting (Confex) / American Junior Academy of Science

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google