Computer Science Journal of Moldova (Aug 2017)
On Digitization of Romanian Cyrillic Printings of the 17th-18th Centuries
Abstract
The paper describes in details recognition of Romanian texts of the \nth{17}--\nth{18} centuries printed in the Cyrillic script, and their conversion to the modern Latin script. The challenges are discussed, and solutions of problems are proposed. The elaborated technology and a tool pack include historical alphabets, sets of recognition patterns, and spelling dictionaries in the corresponding orthographies for ABBYY Finereader. In addition, virtual keyboards, fonts, a transliteration utility, and the user manual were developed. This permits successful recognition of old Romanian texts in the Cyrillic script. Transliteration to the Latin script grants no-barrier access to historical documents.