Desarrollo de un sistema para reconocimiento de texto y conversión a audio, utilizando Raspberry Pi para personas no videntes

Abstract

The main objective of this project is to help people with visual disabilities. A prototype is presented whose main objective will be to detect a printed document and transform it into an audio format, which will reproduce the text detected from the previously made capture. The concept features a single module that allows an audio file to be played back using real-time image detection technology. The project uses the Raspberri Pi card for the processing part and a camera compatible with the aforementioned card for image capture. Finally, the results obtained show the design of the prototype of capture and conversion to audio, several experiments are carried out to show the precision of capture and reading of the printed text. This project has the advantage of using portable devices that apart from this prototype can be used for more tasks. According to the tests carried out, it can be determined that the optimal distance for good focus is 30 cm from the camera to the document since at that distance the entire sheet focuses, lighting is an important part when processing text printed since with good lighting the prototype can better recognize the text and deliver better information to the person requesting it, the Raspberry Pi 4 card offers the necessary resources for the correct functioning of the prototype since it has 4 GB of RAM for its Additional processing is supported by IR-CUT camera.