Design of a solution for the development of an application that will interpret the recognition of Mexican sign language through Convolutional Networks

• DOI: hhttps://doi.org/10.22533/at.ed.3174182421066

• Palavras-chave: Lenguaje de señas mexicano, inteligencia artificial, red neuronal, Dispositivo móvil, TensorFlow Lite.

• Keywords: Mexican Sign Language, artificial intelligence, neural network, Mobile device, Tensor Flow Lite.

• Abstract:

  This document addresses the problems faced by those people who have a hearing disability [1] in such a way that their main form of communication is through sign language, which for our work only considers Mexican and the interpretation of this by people who don't know him. The central objective of this document is to propose a design to generate an application for mobile devices and that, through the use of artificial intelligence using convolutional neural networks; be able to interpret personalized sign language transmitted by the hearing impaired person. Some previous works on the interpretation of sign language by other researchers are reflected, exposing its main characteristics and the differences with the one presented here. In the methodology part, the artificial intelligence techniques that can be applied to this topic are explained, justifying why convolutional neural networks are considered in this work through the Tensor Flow libraries. In conclusion, the design proposed in this article will allow the recognition of personalized Mexican Sign Language.