Text reader for the visually impaired

Description

The estimated number of people visually impaired in the world is 285 million, yet we see belittle of innovation focused on improving their lives, essentially because the market is not as economically powerful as of general public. Harsh reality is, as I looked online for what innovative things already exist, I saw braille readers costing around few hundred dollars. 71% of the world's population is living off $10 or less a day, Imagine how could those people with no income could afford those things. I used my skills to make a device which will help the visually impaired or blind in reading the regular text. Camera and a small braille cell will be on the finger tip of the person and just by scrolling(camera) over the text, he/she can read through the text, this project relies heavily on OCR (Optical Character Recognition) and Image processing.

Details

Video of six LEDs(will be replaced by solenoids) representing a single Braille character or cell--

The Github repository to python 3 and 2.7 script with Arduino code

Dependencies-

OpenCV

Numpy

Pytesseract (You also need to install the main Tesseract OCR engine, whether in windows or linux. More information - https://github.com/10ishq/Brailled-Text-to-Braille-Reader-with-Optical-Character-Recognition- )

Pillow

Project Logs

Collapse

Regarding Braille actuators
Tanishq • 04/14/2018 at 15:15 • 0 comments

In the final project, Braille actuators will be really really tiny stepper motor I saw the other day on ebay, meanwhile it arrives from china, I made my own actuators with the most simplest solenoid lying around me: 5V relays.

By simple modification I managed to put six of them on a Prefboard inside an Acrylic case.

It is definitely not a final prototype or stage of the hardware, just did because I wanted to know, as soon as possible, what those people who are actually visually impaired think about the project, I'll go to the nearby school for the children with disabilities and film their views about the project and post the video here.

Stay tuned.
Running in Raspberry Pi
Tanishq • 04/14/2018 at 14:12 • 0 comments

The program successfully ran in Raspberry Pi after installing the required dependencies.
The python script in Raspberry is running on python 2.7 interpreter, so those who prefer using python 2.7 can also accomplish the project.
All the files are on the same GitHub link.