Image-text linking

with No Comments

During the early beginnings of my PhD research one of my supervisors, Melissa Terras, and I were doing a bit of research into Image Annotation Tools for reading ancient documents. Melissa put out a request on Stoa for more information about image markup tools to which Tom Elliott replied that he and Seam Gillies “(mostly Sean) took a quick whack today at the basic tracing task (and producing SVG from it) using the OpenLayers toolkit.” (Terras 2008). Most of the links from the post are dead and gone now, but Sean’s presentation of the tool and the example is still available.
Encouraged by Tom I decided to try it out with the now reconsidered ‘Frisian Ox Sale’ tablet. Below you can see the SVG result of the pivotal word ‘quem’ (There are more interesting images of the tablet on the eSAD website).
However, we didn’t really get much further with the concept of text-image linking and image annotation as a part of the eSAD project. But now I am working on a project where we would like to look into the possibilities of publishing texts next to images with links between the two on a line, word or character level. So now I am trying to find out how far the world of Digital Humanities has come on this subject since it last had my great interest.

Now you might ask – why do you need to link ancient document transcription to images of the same document?

“Handwritten documents can, and are, regularly digitized, but human beings must produce transcriptions for them (if this is done at all) and the transcriptions are typically linked to the images only on the level of the page.” (Cayless 2008)

It would be very beneficial to be able to produce links between much smaller regions of a page and the corresponding transcription or commentary on the word or character. This could for example be a link between a what is though to be a word in a handwritten document and a discussion in the corresponding edition about what the word might mean. It would also make navigating between the image and the transcription of some of our larger texts much easier for the reader.

So far I have looked at the following possibilities developed specially for my field:

Another option is Image mapping, which uses the <map> tag in HTML to draw areas on an image that can then be linked.
GIMP was a plugin enabling image mapping. I gave it a try using part of the Vindolanda Tablet 607. It is just about possible to see that I have  drawn a polygon around the first character. GIMP now requires me to input a URL to link this area of the image to. I can use a relative link but in this case I won’t.

 

I saved the image map in the same folder as my original image as 607.map. I made a little screen recording of  opening the map file. It basically shows what the exported .map file looks like. I added some basic HTML to make the file viewable in my browser where I loaded it and taa daa if I click on the region where I added the link, sure enough, it opens the linked webpage.

 
If anyone can think of other linking tools that would work for our project please throw me a link. It would be much appreciated.

 

Cayless, Hugh A. (2008) Linking Page Images to Transcriptions with SVG. Presented at Balisage: The Markup Conference 2008, Montréal, Canada, August 12 – 15, 2008. In Proceedings of Balisage: The Markup Conference 2008. Balisage Series on Markup Technologies, vol. 1. doi:10.4242/BalisageVol1.Cayless01.

Terras, Melissa (2008) Palaeographic Image Markup Tools (blog post) http://www.stoa.org/?p=776, February 19, 2008.