Advanced Audio Aid for Blind people using Raspberry Pi

Fig.1: images/Advanced Audio Aid for Blind people using Raspberry Pi _1639949666.png The implementation of purposed system using Raspberry Pi 4 model B. It takes i

Project Title

Advanced Audio Aid for Blind people using Raspberry Pi

Project Area of Specialization

Artificial Intelligence

Project Summary

• One of the most important sense in human life is vision, without it one’s life is totally filled with darkness.
• According to WHO globally the number of people of all ages visually impaired or blind is 2.2 billion. Unfortunately, there are around 2.4 million of people are blind in our beloved country Pakistan.
• Humanity has always high importance and the blind community is an essential part of the humanity. All the innovations and technologies are developed to make the human life easier and comfortable with each aspect. However, the blind community visionary disability would reduce their chance of using such innovated products. Therefore, blind community considered themselves as a burden for the society, and they do not engage in basic routine activities which results in isolating the blind people from the society. Visual impairment can result at any stage of life. For example, blindness can result by birth borne blind naturally.
• Generally, this group of community is less enthusiastic to participate in additional normal life activities. The visual impair people mainly faces severe problem to navigate the surrounding environment easily. Such an individual is not familiar with the architectural structure of paths, roads, buildings, and different kinds of obstacles. It is hard for these people to recognize the intensity of loss caused by any accident with obstacles. They are only aware of obstacles within their building by their counted steps.
• The aim of the proposed work is to turn the real time world into an audio world by telling blind person about the objects in their way and tell the path. This will enable blind persons to navigate independently without any external help just by using the object detection system in real time.
• It determines real-time artifacts through the camera and warn visually impaired person about the object and its position through the audio output. Another benefit of this project is that it will help them in studies we used OCR system that recognize the characters of printed text and read it which is helpful in studies for Visually impaired people. These programs use image processing and deep learning techniques.
• We also added a Google assistant so that user can search things over internet.
• The main goal of the proposed work is to provide visually impaired people with high accuracy, best performance results and a realistic choice to make the world a happy place for them. The goal of our thesis project is to make a device, or we can say it a helping hand for visually impaired persons that help them enough to face the hardness of time and life.
• This project is helpful in the following features:

1.  Object Detection Recognition
2.  Read printed text.
3.  Act as a guide while walking (obstacle avoider)
4. Google Assistant

Project Objectives

• Implementation of an Advanced Audio Aid for blind people using Raspberry Pi.
• To design and develop the object detection recognition application, read printed text application, based on deep learning techniques that has more accuracy.
• To guide visually impaired person via audio output.
• To make a google assistant so that visually impaired person can search things over internet.
• To make it able to recognize the object with the high processing speed.
• To make capable of reading text quickly to speed up study process.
• To makes navigation easier for blind person as it will warn about obstacles in the way, which may save from many accidents.
• To make it Affordable.
• The sole purpose of this project is to make this world a better place for blind people.
• Process large amount of data in real time environment mainly used for safety and study purpose.

Project Implementation Method

Fig.1: images/Advanced Audio Aid for Blind people using Raspberry Pi _1639949666.png

• The implementation of purposed system using Raspberry Pi 4 model B. It takes inputs via webcam, ultrasonic sensor and USB mini microphone and process it according to selected algorithm.
• Raspberry Pi 4 model B is used because it is more power as compared to Raspberry Pi 3 model B.
• Firstly, we install and setup Raspbian OS and enable pi camera and connect ultrasonic sensor and Mini USB microphone.
• Secondly, we program the system for each algorithm (OCR, Object detection and Obstacle avoider) using OpenCV Python and implement it on Raspberry Pi via command line in Raspbian OS.
• Thirdly, Implement Google Assistant API.
• Lastly, mounted the setup on a pair of eyeglasses and can provide real-time auditory feedback to the user through a headphone. Camera and sensor are used for distance measurement between the obstacle and the user, Mini USB microphone is used to take voice command for Google assistant.

• 

             Fig.1: Hardware configuration of proposed system         

Benefits of the Project

• Deep learning techniques are used to get high accuracy.
• Help blind people in their daily life.
• Makes navigation easier for visually impaired person and saves their lives from many deadly accidents.
• Also work as reading assistant, read printed text which speeds up study process.
• A helping hand for blind people to recognize things of surrounding.
• Visual impaired person can search things over internet which in turns help in education as well.

Technical Details of Final Deliverable

images/Advanced Audio Aid for Blind people using Raspberry Pi _1639949667.png

The project technical flow will be under as: 
 

• The system is fully based on 4th Industrial revolution-based technologies like loT, Big-Data ,Cloud infrastructure,Wearable and implantables, NeuroTech and Artificial Intelligence.
• The prototype system will be delivered along with its technical detailed reports. The system mainly consists of two parts:
• A Raspberry Pi 4 model B (hardware) takes real-time data via pi camera and ultrasonic to detect obstacle and Mini USB Microphone to take voice commands.
• The Algorithms for each application implemented on Raspberry Pi 4 model B.
• For this finaly project Python language is used. Python reads like our everyday English language; thus, it makes use of Deep learning techniques easier and less complex. Python's concise syntax implies that it requires less coding time that most other programming languages and permits the developer to rapidly test algorithms without executing them.
• OpenCV-Python is used for each algorithm.

• Compared to traditional computing techniques, Deep learning techniques have high accuracy.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Medical

Other Industries

Education

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT), Cloud Infrastructure, NeuroTech, Wearables and Implantables, Big Data

Sustainable Development Goals

Good Health and Well-Being for People, Industry, Innovation and Infrastructure

Required Resources

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