Handwriting Analysis Tools

The field of computational palaeography has been developing for decades but it only recently starting to show some real promise. Some very good online articles are:

Here are some other tools you can use to play with manuscripts and their scripts:

Archetype Ink

As it describes itself on its Twitter bio, Archetype (previously DigiPal) is "an integrated suite of web-based, open source tools for the study of medieval handwriting, art and iconography. The tool remains in development and though only an early version exists for Windows, a stable version exists for Mac and Ubuntu/ Linux. You install the program via Docker. Here are instructions about how to do so and how to get started; more documentation on their Github site.

The tool is designed to identify letter forms, compare them, and allow you to pinpoint key features you can use to identify scripts.

HAT-2

One promising project is being worked on at the Centre for the study of Manuscript Cultures at the University of Hamburg (Germany). As it describes itself in the manual, HAT-2 is "a software tool that can be used to analyse handwriting styles. Several different handwriting styles (scribal hands) can be analysed concurrently and sorted according to their similarity to a questioned or unknown style (query). A similarity score will be calculated for each predefined style (scribal hand) to create a relative comparison between them with respect to an unknown style."

You can download the installation file here. This tool is designed for use on Windows computers, but can be used on a MacOS via a virtual machine or Wine. The manual with instructions for set up and how to use the tool is located in the zipfile.

You can find a very dense and complex explanation of the work underlying the Handwriting Analysis Tool (HAT) here:

H. Mohammed, V. Märgner, T. Konidaris, and H. S. Stiehl, “Normalised local näive bayes nearest-neighbour classifier for offline writer identification”, in 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR). IEEE, 2017, pp. 1013–1018.

Last updated