Machine Learning (ML)-Based Braille Lippi Characters and Numbers Detection and Announcement System for Blind Children in Learning

Introduction

L. Braille created the Braille Lippi writing or printing method for blind people, which uses combinations of palpable dots or points to form letters, characters, etc. that can be read by touch. People who are blind or partially sighted can read for the rest of their lives if they learn Braille. Learning Braille early on is especially beneficial for literacy because Braille is a considerably more effective medium for comprehending punctuation, grammar, and spelling than audio. Books and publications aren't the only things that are written and typed in Braille. Additionally, it is used to mark commonplace objects like prescriptions as well as door signs, elevator keypads, and restaurant menus in public areas. It is used to make a variety of papers, including bank statements, more easily accessible. For example, three different Braille cells are used to write the word “can” in Alphabetic Braille—one cell for every one of the three letters within the word. If your primary hobbies are making grocery lists, playing board and card games, writing down phone numbers, reading lift buttons and memorizing room numbers, The Frenchman Louis Braille created it in 1824. Each letter uses a different pattern to make up the six dots that make up a Braille “cell,” which resembles a domino. Around 250 letters (phonograms), numbers, punctuation, formatting symbols, contractions, and abbreviations make up this language (logograms). Some English Braille characters, like “ch,” correlate to several print letters. The Braille cell, also known as the basic Braille character, is composed of six dots arranged in a rectangle shape, three dots high and two dots wide. These six dots can form other symbols in some cases. Known as uncontracted and contracted Braille, Grades 1 and 2 are the two most popular types of Braille. The most fundamental type of Braille is Grade 1, often known as uncontracted Braille or alphabetic Braille.

Due to its usefulness in several daily tasks, Braille character & digits recognitions have gained importance in today's digital environment. The fact that numerous recognition systems are being developed or proposed for use in a variety of sectors at which high categorizing efficiency is necessary in recent years serves as evidence of this. People have the ability to finish more complicated tasks that might otherwise require an extended period of time and become expensive due to the aid of systems that identify Braille Lippi letters, characters, and numerals. The biological Braille networks that enable people to learn and model non-linear and complicated relationships can serve as inspiration for the Braille Lippi recognition systems. Therefore, it is created by using the Artificial Neural Network (ANN) by Liyakat (2023), Liyakat(2024). People can distinguish between various Braille Lippi items, such as numbers, letters, and characters, thanks to their Braille’s. Humans can choose to interpret Braille Lippi letters and numerals in a variety of ways because they are biased. On the other hand, impartial computer systems can complete extremely difficult jobs that could demand a lot of time and effort from people to complete them in the same way.

The human visual system is mostly used when reading Braille Lippi characters, letters, words, or numerals. It never seems difficult to read handwriting, but it's not as easy as people believe. Even though everything is done unconsciously, people can still interpret what they see by using the information that their Braille has been taught. Only when trying to build a computer system that can read Braille Lippi does the difficulty of visual pattern recognition become obvious. The best strategy for creating systems that can recognize Braille characters is thought to be the use of ANNs. When reading Braille Lippi in condensed manner, NNs assist in simulation, how human Braille functions. It enables technology to read as well as, if not better than, humans. The ability of a neural network to extract meanings from complex data and find trends from data that are difficult to spot by either other manual techniques or humans makes it the most suitable type of algorithm for the suggested system by Mulani (2019). The major goal of the study is to create a model using the Convolution Neural Network (CNN) concept that will be used to read Braille Lippi numerals, letters, and sentences from an image and proclaim them.