Object identification and feedback for visually impaired impared people

For the visually impaired population of the world, this project works as an assistant which detects and identifies objects in real-time, detects obstacle and provides feedback. The main purpose is to assist the visually impaired user just by receiving an audio query and giving an audio feedback. The

Project Title

Object identification and feedback for visually impaired impared people

Project Area of Specialization

Artificial Intelligence

Project Summary

For the visually impaired population of the world, this project works as an assistant which detects and identifies objects in real-time, detects obstacle and provides feedback. The main purpose is to assist the visually impaired user just by receiving an audio query and giving an audio feedback. The system consists of a microcomputer attached with a depth sensing camera and an audio I/O device.

First, the visually impaired user sends audio query, like "find me a Chair". The system then processes the query via speech-to-text and compares the keyword i.e. Chair. After this, if the requested object is identified in the surrounding then the system will provide audio feedback i.e. "Chair is 2 meter away. Turn right (according to obstacle-avoidance)". If the object is not identified, then system replies "Object not found. Change your direction please".

Project Objectives

Following objectives are listed below to fulfill the completion of this project:

1. Detect and Identify Objects in real-time
2. Detect Obstacle
3. Measure Distance & Direction of Object from visually impaired user
4. Give Voice Feedback i.e. Distance & Direction of object from visually impaired user and also the name of object
5. Provide Obstacle-free path via audio

Project Implementation Method

Whole system is embedded in Nvidia Jetson Nano (microcomputer), attached with Depth sensing camera and an audio IO device (headset). To power the microcomputer, a 20,000 Mah power bank is fixed with it. Keeping the camera at sight level, the objectives will be fulfilled as follows:

• Objects will be detected and identified in real-time
• Objects' distance and direction will be calculated
• Once the user sends audio query about an object, the system will give its distance and direction
• The system will detect obstacle and then provide obstacle-free path

For the purpose of project completion, the end deliverable will be in the form of project only. For transformig this project into a viable product, additional assembling, manufacturing and framing costs will be considered.

Benefits of the Project

Following benefits from this project are mentioned below:

1. Assistance for Visually Impaired Population via AI edge device (no need of internet)
2. Easily implentable for outdoor assistance
3. Can be integrated by Google Maps
4. Scalable to deploy in industries, health sectors, security and military

Technical Details of Final Deliverable

Following are the technical details of final deliverable:

• All the system is run in Nvidia Jetson Nano
• Object detection and identification is done in 45 FPS
• Using Intel RealSense Depth Sensing Camera, distance and direction is measured (upto 10 meters)
• Also, obstacles are detected and system provides obstacle-free path
• Via headset, audio input is given and feedback is received

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Health

Other Industries

Medical

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com