An Efficient Way to Represent Braille using YOLO Algorithm

Ahnaf Atef Choudhury, Ramkrishna Saha, Syed Zamil Hasan Shoumo, Sadid Rafsun Tulon, Jia Uddin, Md. Khalilur Rahman

2018 · 11 citations · 5 references

Concepts

Abstract

Blindness is a curse that has affected more than 30 million people of the world hampering their day to day lives to an unbearable extent. A braille in this case is a ray of hope for the visually impaired which allows them to see and understand which their eyes do not allow them to. This paper addresses this issue and proposes to create a dynamic braille which will identify an object using a camera in real time, convert the identified image into text and finally convert the text into braille language. Around 9000 objects can be successfully detected using this model. This is an innovative way of combining the functioning of braille along with the object detection methods of image processing and thus hopefully using this the problems that arise due to blindness can be solved to some extent.

References

5