Program and algorithm for segmentation and recognition of hand-printed characters using convolutional neural networks

E. S. Popova, V. G. Spitsyn, Yu A. Bolotova

Результат исследований: Материалы для типов конференцийДокументрецензирование

Аннотация

The article is devoted to the development of the algorithm for segmentation and recognition of hand-printed symbols on images. A generalized algorithm for the operation of the text recognition system is presented. The paper describes the methods of segmentation of text documents. The application of a neural network approach based on the architecture of convolutional neural networks is proposed to solve the problem of recognizing hand-printed characters. Results of numerical experiments on the recognition of USE forms on the basis of the proposed approach showed a recognition accuracy of 94.1%, which exceeds the recognition accuracy of 82% obtained with the help of ABBYY FineReader.

Язык оригиналаАнглийский
Страницы230-233
Число страниц4
СостояниеОпубликовано - 2018
Событие28th International Conference on Computer Graphics and Vision, GraphiCon 2018 - Tomsk, Российская Федерация
Продолжительность: 24 сен 201827 сен 2018

Конференция

Конференция28th International Conference on Computer Graphics and Vision, GraphiCon 2018
СтранаРоссийская Федерация
ГородTomsk
Период24.9.1827.9.18

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Fingerprint Подробные сведения о темах исследования «Program and algorithm for segmentation and recognition of hand-printed characters using convolutional neural networks». Вместе они формируют уникальный семантический отпечаток (fingerprint).

Цитировать