Alisson Pereira Anjos, Francisco Assis da Silva, Leandro Luiz de Almeida, Danillo Roberto Pereira, Mário Augusto Pazoti, A. O. Artero, M. A. Piteri
{"title":"RECONHECIMENTO DE CÉDULAS DO REAL A PARTIR DE IMAGENS USANDO CNN PARA AUXILIAR DEFICIENTES VISUAIS","authors":"Alisson Pereira Anjos, Francisco Assis da Silva, Leandro Luiz de Almeida, Danillo Roberto Pereira, Mário Augusto Pazoti, A. O. Artero, M. A. Piteri","doi":"10.5747/ce.2021.v13.n1.e352","DOIUrl":null,"url":null,"abstract":"Real Banknotes recognition through touch has always been a problem found by visually impaired. The advancement of technology makes it possible to solve this problem computationally. In this work, we present a method to perform Real Banknotes recognition from images using computational vision and artificial intelligence algorithms. The results show that computational cost and recognitionrate are acceptable for use in uncontrolled environments. The processing time for recognition of each Real banknote was 200 milliseconds, with an accuracy of 91.67%.","PeriodicalId":30414,"journal":{"name":"Colloquium Exactarum","volume":"24 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Colloquium Exactarum","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5747/ce.2021.v13.n1.e352","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Real Banknotes recognition through touch has always been a problem found by visually impaired. The advancement of technology makes it possible to solve this problem computationally. In this work, we present a method to perform Real Banknotes recognition from images using computational vision and artificial intelligence algorithms. The results show that computational cost and recognitionrate are acceptable for use in uncontrolled environments. The processing time for recognition of each Real banknote was 200 milliseconds, with an accuracy of 91.67%.