As part of a digitization project, some post-production is necessary to obtain surrogates that match the original’s detail, sharpness, and tonal range. The nature of the project governs the amount of processing required.

For example, automatic processing is appropriate for homogenous files, whereas manual adjustments may be performed on individual images.

Institutions should strive to provide faithful reproductions of the originals. Images may be framed to show beyond the originals’ edges when appropriate, rotated, deskewed, or sharpened. These adjustment processes, however, should not be used to overcome the proper equipment operation.

Color management begins with correct scanner operation and capture time to make the original scan as accurate as possible. Next, the original document should be compared to the image under controlled viewing conditions on a color-controlled monitor. Finally, using software such as Photoshop, the image may be adjusted to match the colors closely.

File-Naming Practices

Finding and classifying digital images is easier if file naming conventions are followed. Files move through servers and systems during their lifespan. Computers cannot interpret logical relationships in a collection of images. Therefore, these relationships must mirror how the files are named. File-naming conventions help sort like objects together in search results and create persistent URLs for digital surrogates. Archivists should follow conventions when creating file names to ensure consistency and simplify processing.

Files have unique names and are independent of the location within a directory structure for context. At a minimum, file names should combine information about the item’s location, a unique identifier, and a file type extension.

Derivative files should have similar file names as their master files, with a usage indicator appended to the item number. Master files require no usage appendage. For example, suppose a digital object consists of multiple files. In that case, each file name must contain the object’s identifier, appended with a unique sequence number containing enough digits to account for all items in the collection.

Guaranteeing Integrity

Quality control is an essential component in digitization to guarantee integrity. The conditions for quality require the identification of the desired result and production goals. The quality of image capture cannot be any better than the source image of a scan; the source imposes the upper limit on image quality. Therefore, archivists should define acceptable levels of digital image quality based on the attributes of the source, the capability of the digital imaging system used, and the output of the digitization process to be judged against.

A quality review process is necessary for in-house and outsourced images. In both cases, archivists must ensure that images meet specifications. Contracts with vendors for digitization services include provisions for the re-digitization of images that fail to meet specifications, and institutions contracting with vendors must have a means of identifying these images quickly. When the vendor sends derivative images, they should also be subject to quality review.

Ideally, quality control should be performed on all master and derivative images. The quality control of image files can be maintained by determining an appropriate percentage of images to check, depending on the project’s size and the scanning technician’s skill. Many projects adopt sampling to reduce the costs of this process, such as checking only 10% of the images. The people other than those performing the digitizing should conduct the quality review process.

A complete quality review process is no substitute for training and supervision of digitizing staff. It is also only effective if image specifications have been well-defined and tested at the start of the project, based on an understanding of the source material and digital imaging best practices.

Documenting Efforts

Creating and maintaining digitization documentation takes effort. Documentation conveys the processes behind the construction of digital collections, including project plans, metadata guides, and workflows. It provides benefits in the long run as digital collections are maintained and enhanced. In addition, if digitization projects standardize systems and processes, each new project will require less unique documentation.

The blog was originally published on Lucidea's blog.