Deep Transfer Learning for writer identification in medieval books

In digital paleography, recent technology advancements are used to support paleographers in the study and analysis of ancient documents. One main goal of paleographers is to identify the different scribes (writers) who wrote a given manuscript. Deep learning has recently received much attention from researchers as classification system and has been applied to many domains. However, this approach is based on the hypothesis that large amount of labeled data are available. To overcome this drawback, transfer learning techniques have been proposed. These techniques use parts of large deep networks, learned by using very large image datasets, as starting points for the learning of networks to solve a more specific classification problem. In this paper, we present a deep transfer learning based tool to help paleographers in identifying the parts of a manuscript that were written by the same writer. The proposed approach has been tested on a set of digital images from a Bible of the XII century. The achieved results confirmed the effectiveness of the proposed approach.