Application to assist visually impaired people in the recognition of Real money bills with Artificial Neural Networks technique

Authors

DOI:

https://doi.org/10.5335/rbca.v14i3.13058

Keywords:

Banknotes, visual impairment, image recognition, artificial neural networks

Abstract

Visual impairment affects about 3.4% of the Brazilian population, that is, about 443 thousand people. Among the various problems faced by them, the use of money stands out, for which they depend on the help of people they know to carry out the identification of the money bills. In this sense, this work aimed to develop an application for smartphones that helps them to recognize Brazilian’s money bills. Artificial Neural Networks were used for the recognition process, which have been gaining visibility when it comes to image pattern recognition. As results, it is highlighted that the Brazilian’s money bill recognition model had an accuracy of 95% on the test dataset. An evaluation process of the developed application was carried out with five visually impaired users, who used it and answered questions in order to measure their level of satisfaction regarding the use of the mobile application.

Downloads

Download data is not yet available.

Published

2022-07-19

Issue

Section

Original Paper

How to Cite

[1]
2022. Application to assist visually impaired people in the recognition of Real money bills with Artificial Neural Networks technique. Brazilian Journal of Applied Computing. 14, 3 (Jul. 2022), 1–16. DOI:https://doi.org/10.5335/rbca.v14i3.13058.