2022 Fiscal Year Final Research Report
Character Recognition of English Early Printed Books with AI-OCR
| Project/Area Number |
20K20705
|
|---|
| Research Category |
Grant-in-Aid for Challenging Research (Exploratory)
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Medium-sized Section 2:Literature, linguistics, and related fields
|
|---|
| Research Institution | Senshu University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2020-07-30 – 2023-03-31
|
| Keywords | 文字認識 / 初期印刷本 / 手稿写本 / 写本言語の比較 |
|---|
| Outline of Final Research Achievements |
Research on AI-OCR (Optical Character Recognition) will promote study of Medieval English literature. The present study concerns Early English printed books such as Chaucer's The Canterbury Tales: The Knight's Tale and William Langland's Piers Plowman. Progress in AI-OCR programs has gained a fruitful result. However, there are still problems for recognition: slightly printed letters and ligatures (connected letters such as w = ea, w = he, etc.)
AI-OCR Project, University of Innsbruck, Austria has kindly given me a chance to investigate recognition of hand-written manuscripts of Canterbury Tales and Piers Plowman. The Created Model for our MSS is quite suitable for recognition.
|
| Free Research Field |
人文情報学
|
| Academic Significance and Societal Importance of the Research Achievements |
AI-OCRは現代社会においてDX, IoT分野と同様に社会的貢献を果たすことが可能である。従来、ほとんど研究対象にならなかった初期印刷本、手稿写本の特徴を捉え、比較検討することが可能になる。さらに、Digital Humanitiesという分野において、資料全体を文字イン指揮することにより、重要な研究資料を作成することになる。
さらに、それらを集積して、E-Librayを設立すれば、世界のどの場所にいても、世界で1つしかない資料を調査・研究が可能である。
|
