Machine Annotation for Digital Imagery of Historical Materials Using the ALIP System

Manual annotating digital imagery of historical materials is a labor-intensive task for many historians. In this paper, we introduce the application of the ALIP (Automatic Linguistic Indexing of Pictures) system, developed at The Pennsylvania State University, to the problem of machine assisted annotation of these images. The ALIP system learns the expertise of a human annotator from a small collection of representative images. The learned knowledge about the domain-specific concepts is stored as a dictionary of statistical models in a computer-based knowledge base. When a new image is presented to ALIP, the system computes the statistical likelihood of the image resembling each of the learned statistical models and the best few are further studied for the purpose of annotating the image with keywords. The Penn State research team applied their ALIP system to the EMPEROR images base and metadata created by C.-c. Chen. Promising results have been obtained.