We have talked about why digitisation of the Chester Beatty collections is especially important and we have talked about when a digital file becomes a digital asset. However, we are yet to provide an overview of our digital workflow within the Digitisation Department at the Chester Beatty from a digital photographer’s role.
Since our Digital Department was established, we have been continuously striving to build upon our digital asset management and digital workflow to be as comprehensive, sustainable, quality assured and efficient as possible. For this matter our digital workflow system involves a strict protocol and the use of multiple file formats which are used for different purposes within our digitisation output. This blog will provide an overview of our digital workflow system at the Chester Beatty and discuss the different purposes of each file format we use and how we preserve our digital assets.
To begin with, during the image capture phase of our digitisation process we only use raw file format, and all of our digital processing and adjustments are then done directly from the raw file. Adjustments consist of cropping each image and altering metadata when needed. We work with raw file format to avoid any unnecessary data loss and to maintain the highest quality and flexibility of the final image. A raw file is uncompressed and contains the absolute maximum amount of information and original data generated by a camera sensor. Therefore, a raw file allows for optimal processing and adjustment control over the final image, this is also why a raw file is often referred to as the digital photographer’s equivalent to a film negative. However, one of the downsides to working with raw file format is that they do have large file sizes, meaning extra storage is often needed when shooting high quantities of raw images as we do at the Chester Beatty. A raw file size typically roughly matches the size of the camera sensor, for example if you are using a camera equipped with an 80-megapixel sensor your file sizes will be roughly around 80 to 100 megabytes. The other downside is that not many software platforms can handle or render such a large file size, therefore specific software needs to be installed to be able to work with raw format. In our studio we use Capture One software for all of our image capture conducted in the studio as well as for all of our image processing of object images.
After each image capture session is completed at the end of the day two copies are made of each image onto separated servers as a security precaution to avoid accidental loss or corruption. Before exporting beyond the raw image, the most important step is to ensure all appropriate metadata is attached to each image, including the object number, page or print number and copyright details. The metadata is double checked by a co-worker as a quality assurance protocol to confirm all metadata is correct, whilst the exposures and cropping of each image is also double checked before beginning the exports.
Overall, five different files are created for each image, which consist of a high-resolution preservation tiff, access jpeg, another access jpeg exported at lower resolution and a re-exported raw file. The Raws are re-exported to include the correct metadata and file name. While access jpegs retain the same numbering as the preservation Tiff file, lower resolution jpegs are exported in order to create a chronologically ordered image sequence. This sequence allows the manuscript to be viewed in the same order on our collections online to how the manuscript itself is read.
The preservation tiff is used as an archive image, this file is exported to the highest resolution practical without overbearing our storage, which is 16bit, 300dpi and is exported with an Adobe 1998 colour profile. Tiffs are the preferred archival format as they are uncompressed files, meaning image quality and metadata is maintained when exporting a raw image to a high-resolution tiff.
The access jpeg is the primary file format used for the Chester Beatty online collections website, exported to be 300dpi at 80% scale to the original file, all jpegs are exported with the smallest colour profile sRGB Colour Space Profile. Jpegs are lossy compressed files, meaning if the file was uncompressed the decompressed image would have slightly less information than the original image. Jpegs are typically the universal file format for web use due to their small file size, which allows them to be easily uploaded or downloaded. However, jpegs are not necessarily practical for archival use due to quality loss. A single low-resolution jpeg is also created of each digitised object and exported at 20% scale to the original image, this file is used as a thumbnail viewer for our Chester Beatty collections database.
After all of the files have been exported and added to our collection’s online website,all backup files and jpegs are deleted whilst the archival tiffs and re-exported Raws are kept. The archival files are stored on one of our backed-up servers to protect against accidental loss or corruption.
Considering we have recently upgraded from our Phase One IQ3 80mp digital back to an IQ4 150mp digital back, exhaustion of storage space is an issue we need to consider and is a matter we are using the current lockdown to plan for. Our strategy thus far is to increase our storage space and consider digitising certain parts of collections that contain minimal detail at a lower resolution that is similar to our old digital back of 80mp. We anticipate that our new potential need for greater storage space will be a matter of trial and error, however certainly will not be a challenge we won’t be able to overcome.
Caroline Harding, Digital Photographer at the Chester Beatty