Towards Style-Based Dating of Historical Documents

Abstract

Abstract-Estimating the date of undated medieval manuscripts by evaluating the script they contain, using document image analysis, is helpful for scholars of various disciplines studying the Middle Ages. However, there are, as yet, no systems to automatically and effectively infer the age of historical scripts using machine learning methods. To build a system to date medieval documents is a challenging problem in several aspects: 1) As yet, no suitable reference dataset of medieval handwriting exists; 2) relatively little is known about the evolution of writing styles in the Middle Ages, and especially in the later Middle Ages. Our Medieval Paleographic Scale (MPS) project aims at solving these problems. We have collected a corpus of charters from the Medieval Dutch language area, dating from the period 1300 to 1550. A global and local regression method is proposed for learning and estimating the year in which these documents were written, using several features which have been successfully used in writer identification. The proposed system can serve as a basic tool for the medievalist or paleographer. The experimental results of the proposed method demonstrate its effectiveness.

Historical Manuscript Production Date Estimation Using Deep Convolutional Neural Networks

“…The assumption was that in a small volume of the feature space, the parameters of the best SVR estimator might differ from the best SVR on the whole training set. This approach is used in [3] for estimating the production dates of a dutch charter collection, which is similar to SDHK. The authors used hand crafted features, setting their approach apart from ours.…”

“…In our earlier work [1], [2], we have showed that a Gaussian Process (GP), in combination with handcrafted ink stroke descriptors, can be used for manuscript dating. In [3], Support Vector Regression (SVR) with a two phase estimation refinement have been used in combination with handcrafted features. In this paper, we evaluate these regression techniques with novel features learned from a CNN.…”

Digital Approaches to Paleography and Book History: Some Challenges, Present and Future

“…237 and 247-8), but his challenge has met with relatively little attention. Some work has been done on dating manuscripts based on the assumption that variations in script develop over time (He et al, 2014), but the objective here is still to determine dates rather than to investigate the evolution per se (but cf. Poulle, 1977).…”

Historical Document Dating Using Unsupervised Attribute Learning

“…Automatic date estimation provides an efficient tool for scholars who work with a large digitized database. Therefore, it has attracted several researches [1], [2], [3], [4], [5]. Document representations play an important role in automatic date estimation based on the writing style.…”

“…For handwriting, the possible attributes are: (1) the wide loops and narrow loops created by the letters 'e', 'o' et al (2) long crosses on 't' and short crosses. (3) large gaps between words and crowded words.…”