How Good Can It Get? Analysing and Improving OCR Accuracy in Large Scale Historic Newspaper Digitisation Programs

• Published in: D-Lib magazine Vol. 15; no. 3/4
• Main Author: Holley, Rose
• Format: Journal Article
• Language: English
• Published: 01.03.2009
• Subjects: Antiquarian materials; Australia; Digital preservation; Digitization; Electronic newspapers; Library materials; Library resources; National libraries; Newspapers in microform; Optical character recognition; Optical character recognition devices; Quality control

Details the work undertaken by the National Library of Australia Newspaper Digitisation Program on identifying and testing solutions to improve OCR accuracy in large scale newspaper digitisation programs. Gives a state of the art overview of how OCR software works on newspapers, factors that effect...