Artificial Eye obstacles detection and recognition System

Visually impaired and completely blind people face constant challenges and issues in their daily lives. The lack of proper vision impacts their mental health due to which have low self-esteem and confidence in themselves. This project focuses on developing a system that will help completely blind pe

Project Title

Artificial Eye obstacles detection and recognition System

Project Area of Specialization

Artificial Intelligence

Project Summary

Visually impaired and completely blind people face constant challenges and issues in their daily lives. The lack of proper vision impacts their mental health due to which have low self-esteem and confidence in themselves. This project focuses on developing a system that will help completely blind people in their routine activities so that they can live an easy and independent life. Some of the common challenges which are faced by the visually impaired and blind people include traveling from one place to another, lack of ability to see any objects or hurdles present in front of them, and being uncomfortable walking with smart sticks, smart shoes these are previous research projects related to improving blind person life. For catering to these challenges of blind people, this project proposes to develop a system i.e., “Artificial Eye: Obstacle detection and recognition system”. This project wants to contribute to resolving the challenges of blind people in society in their daily lives by using a system that is linked with hardware and software technology. This project also describes how a person can wear this system that is comfortable to him/her. This also helps in improving blind persons’ mental health so he/she will not feel inferior because of their disability. This system trains the model using TensorFlow lite and detects and recognizes the object using OpenCV, that implemented on raspberry pi 3B and pi camera module utilizing raspberry pi 3 and pi camera it also estimates the distance between the detected object and blind person. The presence of detected objects and the distance estimated by the system can be heard by the blind person via headphones connected to Raspberry pi.

Project Objectives

The aim is to develop a system that helps a blind person’s daily routine, and he feels free while working. Same as a normal person.

The main objectives of this system are:

• To train a model for detecting objects in front of a blind person using TensorFlow Lite.
• To detect any obstacle in front of a blind person using OpenCV.
•  To convert the name of the detected object to the voice format and then send it to the headset using pyttsx3.

Project Implementation Method

The methodology is divided into two modules:

Module 1: Obstacle detection and distance estimation:

Module 1 is divided into two phases:

• Phase 1: Initially, we train the model by giving images of objects used in our daily routine using TensorFlow, that trained model is converted into TensorFlow lite, then in the next stage by using OpenCV to detect the obstacle and recognize that detected object.
• Phase 2: Next, it estimates the distance between a blind person and an object.

Module 2: Text to voice conversion and sending a voice message:

Module 2 is divided into two phases:

• Phase1: detected and recognize object text is converted into voice
• Phase 2: Sending voice messages to a blind person via headphones

These methods are implemented in raspberry pi 3B, then we keep this system on a band that band is attached to a blind person’s arm.

Benefits of the Project

1. The main benefit of this project is to keep relaxed and comfortable a blind person while using this system and he/she can live a normal life without any dependencies.
2. It also benefits in the medical field while working on a blind person’s self-esteem and mental health because he never feels inferior.

Technical Details of Final Deliverable

The tools that we used in our project:

1. Hardware:

• Raspberry pi
• Raspberry pi camera
• Power bank 
• Headphones
• SD card
• Card reader
• HDMI Cable
• VGA Cable
• Arm Band

1. Software:

• Python
• TensorFlow Lite
• OpenCV
• Pyttsx3

The final deliverable is a system that helps a blind person while his daily routine in detecting and recognizing objects and estimating the distance between the object and a blind person, this system also sends the message via headphones to a blind person.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Health

Other Industries

Medical

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT)

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

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