REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOS
The handwritten character recognition is still a major challenge in the field of computer vision, primarily due to the diversity of styles that people can write, which makes it difficult to generalize the problem. In addition, there is also the difficulty in defining the descriptors that best ch...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | Portuguese |
Published: |
Universidade do Oeste Paulista
2014-08-01
|
Series: | Colloquium Exactarum |
Subjects: | |
Online Access: | http://revistas.unoeste.br/revistas/ojs/index.php/ce/article/view/1095/1164 |
_version_ | 1811244589663649792 |
---|---|
author | Clariane Silva Menezes Leandro Luiz de Almeida Francisco Assis da Silva Mário Augusto Pazoti Almir Olivette Artero |
author_facet | Clariane Silva Menezes Leandro Luiz de Almeida Francisco Assis da Silva Mário Augusto Pazoti Almir Olivette Artero |
author_sort | Clariane Silva Menezes |
collection | DOAJ |
description | The handwritten character recognition is still a major
challenge in the field of computer vision, primarily due to the diversity
of styles that people can write, which makes it difficult to generalize
the problem. In addition, there is also the difficulty in defining the
descriptors that best characterize the character and build high
performance OCR systems. This paper presents a system for
recognizing handwritten characters offline, using Artificial Neural
Networks Modular Class with classic backpropagation training
algorithm, besides the methods used for feature extraction. Although
training of neural classifiers require long processing and recognition of
62 classes of characters, few studies have considered the results
obtained from the experiments are shown very promising, achieving
hit rates above 90%. |
first_indexed | 2024-04-12T14:27:50Z |
format | Article |
id | doaj.art-bad462acf1924497954fa156fe87329b |
institution | Directory Open Access Journal |
issn | 2178-8332 |
language | Portuguese |
last_indexed | 2024-04-12T14:27:50Z |
publishDate | 2014-08-01 |
publisher | Universidade do Oeste Paulista |
record_format | Article |
series | Colloquium Exactarum |
spelling | doaj.art-bad462acf1924497954fa156fe87329b2022-12-22T03:29:23ZporUniversidade do Oeste PaulistaColloquium Exactarum2178-83322014-08-01060217018310.5747/ce.2014.v06.n2.e083REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOSClariane Silva Menezes0Leandro Luiz de Almeida1Francisco Assis da Silva2Mário Augusto Pazoti3Almir Olivette Artero4UNOESTEUNOESTEUNOESTEUNOESTEUNOESTEThe handwritten character recognition is still a major challenge in the field of computer vision, primarily due to the diversity of styles that people can write, which makes it difficult to generalize the problem. In addition, there is also the difficulty in defining the descriptors that best characterize the character and build high performance OCR systems. This paper presents a system for recognizing handwritten characters offline, using Artificial Neural Networks Modular Class with classic backpropagation training algorithm, besides the methods used for feature extraction. Although training of neural classifiers require long processing and recognition of 62 classes of characters, few studies have considered the results obtained from the experiments are shown very promising, achieving hit rates above 90%.http://revistas.unoeste.br/revistas/ojs/index.php/ce/article/view/1095/1164Pattern RecognitionArtificial Neural NetworkImage ProcessingOCR SystemsComputer Vision. |
spellingShingle | Clariane Silva Menezes Leandro Luiz de Almeida Francisco Assis da Silva Mário Augusto Pazoti Almir Olivette Artero REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOS Colloquium Exactarum Pattern Recognition Artificial Neural Network Image Processing OCR Systems Computer Vision. |
title | REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOS |
title_full | REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOS |
title_fullStr | REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOS |
title_full_unstemmed | REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOS |
title_short | REDES NEURAIS CLASSE MODULAR APLICADAS NO RECONHECIMENTO DE CARACTERES MANUSCRITOS |
title_sort | redes neurais classe modular aplicadas no reconhecimento de caracteres manuscritos |
topic | Pattern Recognition Artificial Neural Network Image Processing OCR Systems Computer Vision. |
url | http://revistas.unoeste.br/revistas/ojs/index.php/ce/article/view/1095/1164 |
work_keys_str_mv | AT clarianesilvamenezes redesneuraisclassemodularaplicadasnoreconhecimentodecaracteresmanuscritos AT leandroluizdealmeida redesneuraisclassemodularaplicadasnoreconhecimentodecaracteresmanuscritos AT franciscoassisdasilva redesneuraisclassemodularaplicadasnoreconhecimentodecaracteresmanuscritos AT marioaugustopazoti redesneuraisclassemodularaplicadasnoreconhecimentodecaracteresmanuscritos AT almirolivetteartero redesneuraisclassemodularaplicadasnoreconhecimentodecaracteresmanuscritos |