Imaging

Today I tried my hand at using some of the fancy equipment in the imaging studio at USC. Matt Gainer and Giao (pronounced Yiao) Luong are digitizing USGS topo maps of California from the early 1900's. They have a Sinar P 4x5 camera on a huge 9' camera stand with a BetterLite Super 6K2 Scanback device in it. The camera faces straight down to a 32"x46" black platform that is large enough to handle the maps, posters and other large formats. BetterLite has its own scanning software (ViewFinder) that controls the camera and Scanback, captures the image, renders a tiff file, and sends it to a Mac with Adobe Photoshop for post-processing and quality control (cropping, rotating and converting etc.). KinoFlo daylight-balanced fluorescents (2 sets of 4Bank) shine on the black platform. A QP color target card is scanned with the image with a white/grey/black color scale to help assesss color and tonal values.

Once the image is created, it is examined for focus around the edges, saved, and reviewed to make sure the color is OK. (Careful color evaluation is not necessary for the topos. Images created for the USC Fisher Gallery collection on the other hand will require critical color evaluation.) The images are stored on portable hard drives (Firewire) until they are tfp'd to the staging server. (Portable hard drives make it easy to move work from one person and/or station to another.)

Renditions are later created on the staging server according to the specifications of the collection. Typical derivative renditions of the images include thumbnails, 256x256 jpegs for a quick view, 1024x1024 jpegs, and MrSID compressed images where zooming is required, such as with these topo maps. Tiffs are stored on tape as the archival digital images but not used in the public interface. The imaging process to create content runs parallel to the creation of metadata describing the images. Metadata is sometimes pre-existing, sometimes created from scratch, or sometimes migrated from Excel spreadsheets, as with the Seaver collection. Eventually the metadata and content are brought together. (Part of my work for the last two weeks has been using the ingest system to create metadata for the LA Examiner photos and adding links to the photo images.)