Image to interpretation : an intelligent system to aid historians in reading the Vindolanda texts /
The ink and stylus tablets discovered at the Roman fort of Vindolanda are a unique resource for scholars of ancient history. However, the stylus tablets in particular are extremely difficult to read. This book details the development of what appears to be the first system constructed to aid experts...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | eBook |
| Language: | English |
| Language Notes: | English. |
| Published: |
Oxford ; New York :
Oxford University Press,
2006.
|
| Series: | Oxford studies in ancient documents.
|
| Subjects: | |
| Online Access: | Connect to the full text of this electronic book |
| Summary: | The ink and stylus tablets discovered at the Roman fort of Vindolanda are a unique resource for scholars of ancient history. However, the stylus tablets in particular are extremely difficult to read. This book details the development of what appears to be the first system constructed to aid experts in the process of reading an ancient document, exploring the extent to which techniques from artificial intelligence can be used to develop a system that could aid historians in reading the stylus texts. Using knowledge elicitation techniques (borrowed from artificial intelligence and engineering science), a model is proposed for how experts construct a reading of a text. A prototype system is presented that can read in image data and produce realistic and plausible textual interpretations of the writing that appears on the documents. Incorporating knowledge elicited from experts working on the texts, and utilizing image processing techniques developed in engineering science to analyze the stylus tablets, the book includes a corpora of letter forms generated from the Vindolanda text corpus, and a detailed description of the architecture of the system. This research presents the first stages towards developing a cognitive visual system that can propagate realistic interpretations from image data. |
|---|---|
| Physical Description: | 1 online resource (xi, 252 pages) : illustrations |
| Format: | Master and use copy. Digital master created according to Benchmark for Faithful Digital Reproductions of Monographs and Serials, Version 1. Digital Library Federation, December 2002. |
| Bibliography: | Includes bibliographical references (pages 231-247) and index. |
| ISBN: | 9781435621169 1435621166 9780191708121 0191708127 1280845694 9781280845697 9786610845699 6610845697 0191525448 9780191525445 |