Saturday, July 21, 2012

CDLI News: Berlin cuneiform, and a few words concerning CDLI images

From Bob Englund 
Berlin cuneiform, and a few words concerning CDLI images 
I am pleased to pass on to AWOL subscribers notice of the insertion to CDLI of new VAM-Berlin images. Scanned by Ludek Vacin of the Free University, Berlin, with the assistance of Joachim Marzahn, Curator of the VAM cuneiform collection, the raw files were processed to fatcrosses in Los Angeles and are viewable at <> (including Köcher, BAM [medical texts], Falkenstein, LKU, and van Dijk VS 24 [literary texts], Jakob-Rost, VS 28 [Hittite], and Matouš, LTBA 1 [lexical tests]; Ulrike Steinert kindly assisted me in piecing together the difficult <> = BAM 3, 237(+?) 238). 
I take this opportunity to offer a few remarks concerning CDLI's policy governing the web dissemination, and the anticipated use of these and other images of physical artifacts in our pages (CDLI use of published line art copies of cuneiform texts was described at <>). In collaboration with collection managers or owners, CDLI staff generally flatbed scan all available artifacts at a resolution of 600ppi and save image files in 24bit RGB format (<>; other standards apply to the scanning of glossy photos or photos in paper publications). The resulting raw scan image files are merged into CDLI "fatcrosses" and further processed to enhance the immediate visual usefulness of the image files. In this processing stage, we generally restrict ourselves to the use of the relatively non-invasive Photoshop tools of levels adjustment (channel: RGB), desaturation (compensating for pixel enhancement occurring in levels adjustments) and "magic wand" and "spot healing brush" to remove, from the black background, remnants of dust and specks that fall from the tablets to the scanner surface, as well as to erase eventual scanner glass scratches visible in background and on uninscribed tablet surfaces. These 600-ppi RGB tif fatcross images are then batched to 300 and 75ppi jpegs, based on an artifact representation of approx. 1:1 at 75ppi on a computer monitor (and thus, at 300ppi, a 4x enlargement of the original artifact). To care for a certain presentation consistency, 75ppi thumbnails of larger artifacts are set at maximally 300x800pixels. Thus CDLI's 600ppi archival tif fatcrosses are not currently distributed through our web server, and permission to receive and use these files must be directed to the respective collections; commercial use of any images of cuneiform artifacts kept in public collections
and made available on CDLI pages is prohibited without the explicit permission from officials of the collections themselves--no such permission can be given by CDLI.

It is obvious to users of
CDLI pages that our images are of uneven quality. <> are three recent VAM examples of what a simple flatbed achieves with willing tablets and an expert scanner. Large accounts with one or both sides rounded, indeed any tablet whose inscribed surfaces are not flat, not to mention irregularly shaped tablets, barrels, and so on, however, may eventually require higher-end imaging (cones appear sufficiently clear directly off the flatbed), as will artifacts with the subtle impressions of cylinder seals. Our policy is to flatbed everything in a collection, and to let images that result from this benign process serve to determine where additional imaging is necessary to achieve a satisfactory digital facsimile. There appears, in the end, to be no real sense in expending substantially more resources to create high-resolution digital facsimiles of cuneiform artifacts than is necessary to satisfy research needs. We are using conventional digital cameras where requested by collection collaborators, and are partnering with imaging specialists at USC in the US (<>), at Oxford and
Southhampton Universities in the UK (<>), and at the University of Leuven in Belgium, in the implementation of Reflectance Transformation Imaging ("dome capture"), to be reserved by
CDLI associates for specific artifact targets, in particular for tablets with cylinder seal impressions; and with the MPIWG-Berlin in the use of 3D scanners (Breukmann) in the Jena Hilprecht collection (cp. <>). But we note that the size of these images currently make their full-resolution viewing online unfeasible, and that, ultimately given the numbers of artifacts in need of digital capture worldwide, imaging should be seen as a tool to achieve dependable and searchable transliterations in the first instance, as the basis for photographically documented paleographies in the second, and, as much as we may be baffled by efforts undertaken to hinder digitization initiatives even in areas of violent conflict, as true and lasting 3D or 3D-like digital facsimiles only in the third.

No comments:

Post a Comment