Towards Automatic Transcription of Estrangelo Script

Towards Automatic Transcription of Estrangelo ScriptHugoye: Journal of Syriac StudiesSyriaca.org: The Syriac Reference PortalThe National Endowment for the HumanitiesThe International Balzan Prize FoundationDavid A. Michelsoncbsc_adminRecord added to Zotero bycbsc_adminSyriaca.org: The Syriac Reference Portalhttp://syriaca.org/cbss/M7XH5GUY/tei

Distributed under a Creative Commons Attribution 3.0 Unported License.

2024-08-15-04:00

Born digital.

CREATED: This bibl record was autogenerated from a Zotero record.William F.ClocksinPrem P. J.FernandoTowards Automatic Transcription of Estrangelo Scripthttp://syriaca.org/cbss/M7XH5GUYhttps://www.zotero.org/groups/a_comprehensive_bibliography_on_syriac_studies/items/M7XH5GUYhttps://www.zotero.org/groups/4861694/items/M7XH5GUYHugoye: Journal of Syriac Studies6:22003This paper surveys several computer-based techniques we have developed for the automatic transcription of Estrangelo handwriting from historical manuscripts. The Syriac language has been a neglected area for research into automatic handwriting transcription, yet is interesting because the preponderance of scribe-written manuscripts offers a challenging yet tractable medium between the extremes of type-written text and free handwriting. The methods described here do not need to find strokes or contours of the characters, but exploit characteristic measures of shape that are calculated by geometric moment functions. Both whole words and character shapes are used in recognition experiments. After segmentation using a novel probabilistic method, features of character-like shapes are found that tolerate variation in formation and image quality. Each shape is recognised individually using a discriminative support vector machine with 10-fold cross-validation. We describe experiments using a variety of segmentation methods and combinations of features. Images from scribe-written historical manuscripts are used, and the recognition results are compared with those for images taken from clearer 19th century typeset documents. Recognition rates vary from 61-100% depending on the algorithms used and the size and source of the data set. Clocksin, William F., and Prem P. J. Fernando. “Towards Automatic Transcription of Estrangelo Script.” Hugoye: Journal of Syriac Studies 6:2 (2003). William F. Clocksin and Prem P. J. Fernando, Towards Automatic Transcription of Estrangelo Script, Hugoye: Journal of Syriac Studies 6:2 (2003).Syriac studiesXX CEdigital