Database Support for Image-Based Electronic Editions

Image-based electronic editions are an emerging type of a multimedia information system. They allow scholars and students to study editions of ancient texts, including a transcript, a comprehensive glossary, and full critical apparatus, in the context of their manuscript source or sources. Building image-based electronic editions poses a two-fold challenge. The humanities editor must be able to annotate manuscript images to represent a wide range of features. Because these annotations can be quite dense, with complex interrelations, the computer scientists must develop mechanisms for representing them in a way that associates the images and encoded text, enable efficient querying of both. This paper addresses the latter problem, by proposing data structures to collect, store and query image-based annotations and to preserve image-to-text and image-to-markup mappings.