Sistema de reconhecimento de padrões do alfabeto da língua brasileira de sinais utilizando microcontrolador

Abstract

The Brazilian Sign Language is the language used by most of the deaf people in Brazil. Currently, there is a challenge to make the society more inclusive for people who have some kind of disability, and a proposal for inclusion is encouraging the learning of Sign Languages for communication between deaf and hearing people may be possible without the use of interpreters. This work aims to develop a system using sensor glove, for the recognition of patterns of the Brazilian Sign Language signs constituting the letter of the alphabet. A simplified technique for project sensor glove is employed using handmade inductive sensors, and the development of the electronic circuit for sensor response signal conditioning. The data is processed using a microcontroller systems development board using an pattern recognition algorithm based on artificial neural network. The letters recognized by the system are displayed on a digital display. Despite the limitations of information available from the sensor glove, the developed sensors showed significant potential in determining hand shapes. The results obtained with the pattern recognition system showed the viability for use the system in applications for development skills in dactylology for beginners in Brazilian Sign Language.