Recognition of Image Pattern to Identification of Braille Characters to Be Audio Signals for Blind Communication Tools

Abstract

The five senses are a source of information in humans. The sense that is the main source of information is the sense of sight. Some humans are created with limited sense of sight. The blind performs reading and writing activities using Braille characters. Braille characters are writing or printed systems for the visually impaired in the form of codes consisting of six dots in various combinations that are highlighted on the paper so that they can be touched. Those with significant visual impairment need special education or learning services. Information is very important for everyone, including blind people. Submission of information is done through various media. One media that is often used is print media such as books. But books available on the market do not adapt the way for blind people to capture information. This research offers alternative solutions to overcome the above problems, namely communication aids for reading the blind. The implementation is in the form of a scanner and webcame that is equipped with a braille character text to speech system as an alternative to the lack of blind reading media, especially braille print books. The method, the reading of Braille character scripts by studying braille characters from a to z. First, a webcame or scanner captures braille characters. Second, the system will convert Braille characters and translate Braille characters into alphabetical form through Optical Character Recognition (OCR) image processing. Recognition of Braille character patterns in written text using Artificial Neural Networks (ANN). The results of research on braille character testing are in the form of alphabetical texts a to z, and audio signals of alphabet pronunciation. The results of testing the introduction of braille character patterns using a scanner for training data 100% and for testing data 90.38 % and 96.15 %.