How to Compute a Shape: Optical Character Recognition for Hieratic

Item Type Book Section
Abstract This paper is written in the framework of the project Altägyptische Kursivschriften and presents an experiment for applying OCR to Hieratic, examining the final step of the OCR pipeline. A convolutional neural network is used to classify individual hieratic characters. The results prove that the classification of hieratic characters is in principle possible, but they also show, where improvements and combinations with other methods promise a better performance of recognition models.
Authors Bermeitinger, Bernhard; Gülden, Svenja A. & Konrad, Tobias
Editors Gracia Zamacona, C. & Ortiz-García, J.
Language English
Subjects computer science
HSG Classification contribution to scientific community
HSG Profile Area None
Date 1 February 2022
Publisher Editorial Universidad de Alcalá
Place of Publication Alcalá de Henares, Spanien
Series Name Monografias de Oriente Antiguo
Page Range 121-138
Number of Pages 188
Title of Book Handbook of Digital Egyptology: Texts
ISBN 978-84-18979-09-5
Official URL https://publicaciones.uah.es/export/sites/publicac...
Contact Email Address bernhard.bermeitinger@unisg.ch
Depositing User Bernhard Bermeitinger
Date Deposited 09 Feb 2022 14:35
Last Modified 20 Jul 2022 17:47
URI: https://www.alexandria.unisg.ch/publications/265815

Download

[img] Text
07 MOA1 Bermeitinger_Gulden_Konradt.pdf - Published Version

Download (2MB)

Citation

Bermeitinger, Bernhard; Gülden, Svenja A. & Konrad, Tobias: How to Compute a Shape: Optical Character Recognition for Hieratic. In Gracia Zamacona, C. & Ortiz-García, J. (ed.): Handbook of Digital Egyptology: Texts. Alcalá de Henares, Spanien : Editorial Universidad de Alcalá, 2022, S. 121-138.

Statistics

https://www.alexandria.unisg.ch/id/eprint/265815
Edit item Edit item
Feedback?