Braille is a writing system used by the visually impaired. It has seen many changes over time and now been simplified to a simple 6 dot-cell mapping to the English Alphabet. The way it works is each alphabet is mapped to a cell where one or more than one dots are raised into a little bump. The reader then feels these bumps and figures out the alphabet. We worked on making a machine where an image of a printed text (say, from a textbook) is taken via a camera, processed by the Raspberry Pi using Optical Character Recognition API into text and the said text processed out as bumps using 6 servo motors in place of 6 dots of a Braille cell. The servo motors are placed inside a black box with a simple mechanical CAM system through six holes on the top through which the actuators poke out the symbol. This, we believe, will have some meaningful impact on society, and in particular to the visually impaired in accessing the books and texts which have never been published in Braille.
The first step of the flow of the project is to capture an image of a printed text and using OCR to decipher the image to text file. Following this we map the characters of the text to the corresponding Braille cells. This mapping table is between the English characters and a six bit string where each bit corresponds to the YES/NO to the presence of a bump in each hole of the Braille cell. These bits, therefore, command the servo of that hole to turn on and off. The servos will be set up with a simple mechanical CAM system where the rotational motion of the servo can be converted into linear motion of a shaft placed in each hole. There are four push buttons implemented to the system with individual functionalities. First button will start and quit the process. Second button is to toggle between the modes “slow” and “fast” where slow corresponds to outputting one alphabet at a time at the push of the third and fourth to go forward and backward. Whereas the fast mode outputs the characters at a pre determined speed.