ASSISTIVE DEVICE TO AID VISUALLY IMPAIRED PERSON

According to world health organization(WHO) millions of people are going through visual impairness. People with visual impairment experience numerous difficulties in their everyday life. With the advanements in technologies , many developers have engineered various devices for visually challenged pe

Project Title

ASSISTIVE DEVICE TO AID VISUALLY IMPAIRED PERSON

Project Area of Specialization

Artificial Intelligence

Project Summary

According to world health organization(WHO) millions of people are going through visual impairness. People with visual impairment experience numerous difficulties in their everyday life. With the advanements in technologies , many developers have engineered various devices for visually challenged people. This project will assist visually challenged people in detecting and reeognizing it’s family members (known persons) , text detection and recognition from printed text using OCR and to detect and avoid the surrounding objects, all of them with voice command output

The hardware architecture comprises of jetson nano as the processing device , pi camera , ultrasonic sensor, earphone to get audio output and power bank

Project Objectives

The objectives of this project are:

1) To design face recognition system of family members or any specific persons

2) Design a system which assist blind person in detecting and recognizing any printed text .

3) To design a system to assist blind person in detecting surrounding objects and avoid it.

4) To develop a portable device

Project Implementation Method

To implement our first objective which is face recognition system we are using transfer learning on pre trained keras model like resnet or mobilenet and use that model to train our custom face recognition system. We have 3 dataset , each dataset contains 500 images , out of 500 , 400 is for training purpose  and remaining 100 is for testing purpose.

To implement our second objective we use a pre trained OCR library called tesseract . we are using image processing  techniques on images for pre processing and after preprocessing we then send the image to OCR engine to get more accurate result.

To implement our third objective we have use an ultrasonic sensor to detect any surrounding object by measuring the distance of the object from the person and then alert the blind person so that it can avoid it.

To implement our fourth objective we are using 5v 3a power supply.

Benefits of the Project

With the help of this device the visually challenged person can have more freedom. Blind person don’t know about its surroundings and they usually get injured because of their carelessness and they also don’t know which person is in front of him. So by face recognition system ,the blind person can detect its family members of friends without the help of others and they will also know if there is any objects near him or not so that it an avoid it. It also assist blind person in reading text from any printed document.

Technical Details of Final Deliverable

Projet contains the following deliverable components.

Jetson nano

Pi camera

Power bank 

Earphone

Ultrasonic sensor

We are getting a portable device which blind person can use anywhere. we are using a camera to get real time inference.The camera will extract the features using computer vision opencv library. After capturing frames we will be able to detect and recognize the faces of specific persons by using face recognition model that we have trained. if the camera is on the printed text it willuse OCR to Extract the text from the image and we will get audio output of the text. we are also using ultrasonic sensor to detect any surrounding objects by getting the distance between the person and object, so that blind person can easily avoid it. 

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Health

Other Industries

Security

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

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