Project Methodology
Image Selection and Manipulation for the Kelsey Museum of Archaelogy and ILS603 Image Database Course
What follows is a brief description of how our group decided on which images to include and how to process them.
1. Selection: The selection of images was done on a random basis in order to show what could be done with this image collection. We chose those that had viewable quality and were not one-of-a-kind originals which could not be replaced.
2. Processing: After we gathered the images, we scanned them on an HP ScanJet II flatbed scanner at a resolution of 600 dots per inch (dpi). The reasoning behind such a relatively high-resolution scan is that these files will be stored by the Kelsey Museum and may be used or re-processed in the future.
We compressed the fairly large files (500K and more) using a JPEG compression scheme so as to make them readily available online via a Web browser.
One method we used to manipulate and further process the images was adjusting their Brightness/Contrast levels. Some of the prints were of low quality and had faded with time. Adjusting the brightness assisted us in bringing them closer to their original state.
A second manipulation decision was based on the yellowing of the photographs. To be true to the original black and white photographs, we scanned them in black and white. One reason was to try and get as close as possible to the original photos, and another, to save disk space, as color information dramatically increases amount of memory required.
The third, and most common, image manipulation, was spot removal. The photographs, transported during the expedition as glass negatives, initially produced remarkably good prints. However, as the years went by, the negatives began showing signs of deterioration, resulting in scratches, spots, and "holes" visible on subsequent prints. We tried to remove these blemishes using tools provided by Adobe Photoshop.
Database Considerations
1. The database was created using Filemaker Pro
2. The fields:
- Country (Text)
- City (Text)
- Date Taken (Text)
- Year Taken (Number)
- Photographer (Text)
- Caption (Text) : (whatever remained after inputting the above fields)
All the info for the above fields was transcribed directly from the
typed captions for the photos. The fields below contain new information:
- Objects (Text): list (up to 4) significant objects on a photo
- Date Scanned (Date)
- Scanner Type (Text): name of the flatbed scanner used
- Manipulations (Text): restorative work on digitized images (e.g.
spot removal, sharpening).
The last 3 fields are not being displayed to the user - they are
provided only as part of the documentation for future expansion of the database.
- Accession number (Text)
Thesaurus
We have created a mini-thesaurus of terms used in the Object field so as
to assist the user in his choice of words in searching. The terms were generated by viewing the images and selecting descriptors which represented the images. The thesaurus has a syndetic structure as its base.
Exhibition History
The information for the exhibition's history was taken from a previous exhibit catalog, provided by the Kelsey Museum. Additional information was gleaned from the Francis W. Kelsey Collection's correspondence files at the Bentley Historical Library, University of Michigan.
Acknowledgement
Many thanks, from the database authors, to the Kelsey Museum of Archaelogy staff, especially to Mrs. Robin Meador-Woodruff, for their help and cooperation during our work on this project.
