I work at a University, and I've just developed a workflow to digitize our cuneiform collection. I don't know anyone on campus that can read cuneiform. Is there anyone here that can tell me whether or not this scan is viable?

by dsquard

Here are some pictures: In my excitement to post this, I realized I didn't edit the photos properly, will post fixed link Monday!

I am more than happy to give a link so you can download the OBJ, PLY, STL, or whatever you'd like. I believe the STL may be best, simply because the artifact seems to be more legible without the texture.

Thank you in advance, I hope this post conforms to the rules here, I know this isn't the typical kind of question that is posed.

rcxheth

I am able to read cuneiform, but the scan is pretty unclear. Most of my experience comes from inscriptions that have been rendered onto a page. The signs on the right side of the scan could be a list of numbers, but it's difficult to tell.

Dokugumo

Something you might try is to inverse the digital image. That is, render the digital image as if it were putty placed onto the cuneiform cylinder. This has been applied to cordmarked pottery from the Mississsippian period U.S. by archaeologist at UW- Madison, and it give a much better idea of cord impressions than simply the 3D rendering of the artifact itself. Perhaps this would help in deciphering the text.

dahud

Might shape-preserving smoothing algorithms be used to make the cuneiform symbols show up better? They appear to be the only large-scale deformations on the object, but all the small variations make it look busy.

datherine

Is there an OCR (optical character recognition - creates typed text from an image of text)program for Cuneiform yet? I've made a simple OCR once before and after reading how many cuneiform tablets have yet to be translated, I would be interested in working on a project that could facilitate this.