Character Segmentation in the Development of Palmyrene Aramaic OCR

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60460709%3A41110%2F23%3A97610" target="_blank" >RIV/60460709:41110/23:97610 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.researchgate.net/publication/374469082_Character_Segmentation_in_the_Development_of_Palmyrene_Aramaic_OCR" target="_blank" >https://www.researchgate.net/publication/374469082_Character_Segmentation_in_the_Development_of_Palmyrene_Aramaic_OCR</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-45010-5_7" target="_blank" >10.1007/978-3-031-45010-5_7</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Character Segmentation in the Development of Palmyrene Aramaic OCR

Popis výsledku v původním jazyce

In this study, we present the research plan and the segmentation solution in progress for our Palmyrene OCR web and mobile application from sandstone tablet photographs, which will be publicly available on the ml-research.pef.czu.cz web portal in the next steps of the research. In this paper, we compare mathematical segmentation methods with artificial intelligence methods, highlighting the advantages and disadvantages of each solution, and propose a fully automated OCR procedure from photographs using convolutional neural networks exclusively and present a development model of our solution. We also present a partially completed segmentation dataset of the Palmyrene letters to demonstrate the functionality of the proposed procedure. We hope to complete the Palmyrene OCR soon, thus making the writings of ancient Palmyra accessible to the scientific community and the public, signifying progress in the area of Digital Humanities. Since the algorithm is not completely ready yet, we also present its development model here.

Název v anglickém jazyce

Character Segmentation in the Development of Palmyrene Aramaic OCR

Popis výsledku anglicky

In this study, we present the research plan and the segmentation solution in progress for our Palmyrene OCR web and mobile application from sandstone tablet photographs, which will be publicly available on the ml-research.pef.czu.cz web portal in the next steps of the research. In this paper, we compare mathematical segmentation methods with artificial intelligence methods, highlighting the advantages and disadvantages of each solution, and propose a fully automated OCR procedure from photographs using convolutional neural networks exclusively and present a development model of our solution. We also present a partially completed segmentation dataset of the Palmyrene letters to demonstrate the functionality of the proposed procedure. We hope to complete the Palmyrene OCR soon, thus making the writings of ancient Palmyra accessible to the scientific community and the public, signifying progress in the area of Digital Humanities. Since the algorithm is not completely ready yet, we also present its development model here.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Model-Driven Organizational and Business Agility

ISBN

978-3-031-45010-5

ISSN

1865-1356

e-ISSN

—

Počet stran výsledku

16

Strana od-do

80-95

Název nakladatele

Springer

Místo vydání

Cham, Switzerland

Místo konání akce

Zaragoza, Spain

Datum konání akce

1. 1. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—