| Autor: |
Ansari, Nazneen, Lopes, Monalisa, Pandey, Shreya, Tuscano, Max, Lobo, Benen, Lobo, Sinon |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2024, Vol. 3188 Issue 1, p1-8, 8p |
| Abstrakt: |
Braille is an essential means of tactile reading and writing for individuals who are blind or visually impaired, where raised dots serve as representations of alphabet letters. The alphabet is represented by each letter in a different way, and each letter also has combing signs that function similarly to punctuation. Six dots are combined. Usually, to read Braille, one must trace each line from left to right using their fingertips and sense of touch. Braille is present in just a very small portion of written or printed texts. A system for automatic recognition of Braille Text to English or English Text to Braille is necessary to bridge the gap between visually sighted people and other people via which they both may communicate. There is always a need for a system that can do both i.e. can take Braille written text or text images and convert them into equivalent English text via which keen sighted people can understand braille text and easily communicate with visually impaired people and can take English written text or text images from the system and convert them into equivalent Braille code. Our system will take the dataset of Braille Character Images which will be used to identify the Braille text characters from images. Braille text Images will be converted into English text using CNN model. The system will utilize the potent Open-Source Tesseract OCR Engine to extract the English text from images utilizing and then convert the text to Grade 1 Braille which is the most basic format and Grade 2 Braille which is a little advanced. Our project provides an efficient way of converting to text between English Language and Braille Code. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
