How can I study my Written Artefact if I only have digital images?

Kyle Ann Huskin, Hussein Adnan Mohammed

The study of written artefacts results in an ever-increasing amount of digital data. Colour images from institutions or one’s own camera are almost always saved as a single image comprised of red, green, and blue channels – that is, as an ‘RGB image’.  It is possible to use image-editing software like Adobe Photoshop or its free counterpart, GIMP, to manipulate such images in order to enhance specific characteristics. Useful features in these programs include the curves function, brightness-contrast enhancements, and hue-saturation adjustments. One can also treat an RGB image as a three-band dataset and manipulate it in more methodical, scientific ways using free image-processing software like Hoku and ImageJ. Because an RGB image is technically a multi-band image, it is possible to perform statistical image processing functions like PCA, ICA, and MNF on it in order to separate different components.

Digital data can also be used to extract information: for example to detect visual-patterns, count text/non-text lines, and analyse handwriting styles. The manual analysis of this data is typically time consuming and can be subject to human error and bias. Therefore, we developed a set of state-of-the-art Pattern Analysis Software Tools (PAST), which can be used for the automatic analysis of visual-patterns in digitised written-artefacts. These software tools have been developed to facilitate a more efficient study of written artefacts and to help the scholars benefit from the rapid advancements in the fields of pattern analysis and artificial intelligence.  More information can be found here.