I’m often asked about my work in Cultural Heritage (CH) imaging, and while sometimes I answer that I take pictures of pictures, I usually talk about the fact that the aims of our particular niche are the inverse of those of just about every other type of photography. That is to say that while nearly all photography aims to create an idealized or clearly mediated view of its subjects, what we do in museums and libraries and archives is to attempt to produce as “objectively” accurate a version of an object as possible. This inversion seems particularly ironic when remembering that photography in its early period was considered a craft or tool for more truthfully depicting “reality” than was possible with drawing. Now, of course, photography is a canonical artistic medium and no one believes any longer that “the camera cannot lie”. And yet, in the imaging studios of museums, we try to tell the literal visual truth about our subjects.
What allows us this conceit is the advent of digital technology. The fact that the performance of camera sensors can be characterized to deliver consistent and objectively defined colorimetric output gives us incredible possibilities for digitization and preservation. And standards are what allow us the possibility to obtain uniformity if not “objectivity” in the field.
The primary reference for establishing CH imaging standards in the US is the Federal Agencies Digitization Guidelines Initiative (FADGI), which began in 2007 and which defines sets of guidelines for best practices “for digitized and born-digital historical, archival and cultural content”. Together with the work of the International Organization for Standardization (ISO), standards and guidelines not just for color accuracy, but for numerous other imaging parameters are currently available for measuring the quality of imaging production.
One of the barriers to achieving standards broadly in CH imaging is the proprietary nature of many of the tools we use. Camera performance can be characterized, but the need for such characterization stems from the fact that what happens under the hood–the interpolation of light data into a coherent picture—in different camera manufacturers—is opaque to the user. This brand-differentiated opacity is the reason that Canons render skin tones differently than Nikons, for example, is analogous to the idea of the ultimately untrustworthy photographic product. So an ongoing project within our community is to encourage manufacturers and vendors to design tools with such standards in mind. One possibility is for cameras to include a mode in which a particular brand’s “flavor” could be bypassed or switched off in favor of straightforward and transparent use of a sensor’s raw data for particular types of standards-based imaging.
Another barrier to achieving imaging standards is, of course, resource availability. Top-notch digital imaging programs are costly and require considerable technical expertise. But there are continual technical developments that bring the promise of reduced cost and greater user-friendliness. A recent example that is being discussed with great interest in our community is the introduction by Apple of their Object Capture API that will allow for very deep access to the Apple OS and be available for developers and manufacturers to develop applications that may significantly lower the barriers to entry for 3D imaging. Our hope in the CH imaging community is that this tool will be used to help create accurate pictures as well as “good” ones.
Interested in joining a SIG?
Through Special Interest Groups, all MCN members have the opportunity to connect with colleagues doing similar work, explore opportunities to collaborate among them, share resources and effective practices, or discuss ways to advance their area of digital practice.MCN members may join any SIG and there is no limit on the number of SIGs one may join. When you fill out the membership registration form select which SIGs you’d like to join, you can join or leave a SIG at any time by updating your profile in the New Member Portal. Join or Login today!