Desenvolvimento de uma inteligência artificial para a classificação de sinais sonoros de trânsito

Abstract

In Brazil, there is a low number of people who are hearing impaired and are drivers of vehicles in traffic (IBGE, 2010), one of the causes of this are the difficulties faced by them when issuing or renewing a CNH (National Driver's License). The expansion of Assistive Technology development has enabled the PcD community to have a better quality of life and social inclusion. The purpose of this project is to develop an Artificial Intelligence, an Artificial Neural Network model which is efficient in classifying whistle, horn and siren sounds, which in a future perspective, can be implemented in an alert system to assist hearing impaired drivers. Furthermore, the presented convolutional-type Artificial Neural Network (CNN) model and the database created, give scope for the use of these components in other applications.