Embedded Systems and TensorFlow Frameworks as Assistive Technology Solutions

Area of Science:

• Computer Science
• Artificial Intelligence
• Assistive Technology

Background:

• Visually impaired individuals face challenges in navigating and understanding their surroundings.
• Existing assistive technologies may lack real-time environmental perception capabilities.

Purpose of the Study:

• To design and develop a wearable computer vision system for the visually impaired.
• To leverage deep learning for real-time object recognition in assistive technology.

Main Methods:

• Utilized a single board computer and smart glasses with a camera.
• Employed the Google TensorFlow machine learning framework for image classification.
• Integrated audio feedback for user interaction.

Main Results:

• The system enables users to explore and identify objects in their environment.
• Real-time classification of acquired images was achieved.
• Enhanced user awareness of the surrounding environment was demonstrated.

Conclusions:

• The proposed wearable computer vision system effectively aids visually impaired users.
• Deep learning frameworks like TensorFlow are crucial for developing advanced assistive technologies.
• The system offers a promising solution for improving environmental interaction for the visually impaired.