Image Database Project/Paper Topics

The following is a list of topics that might make good student papers or projects for Howard Besser's Fall 1995 Image Database class. This is only a partial list; students are welcome to suggest other topics and have these approved by the instructor. Some of these are papers, others are projects. Most of these are the kind of practical activities that (if successfully developed) will be made widely available and become useful to the developing field:

The Special Collections Department in Hatcher Library has recently had external requests to mount photographic images it owns onto the WWW. The need to develop a policy for this, and are interested in examining methods that will prevent these images from being re-used either without proper credit, or in an unauthorized way. Papers: Examine and evaluate the ways that various institutions (particularly archives) are planning to use to protect their images (ie. only mounting low-resolution images, labeling the institution name right on the digital photo, embedding watermarks or digital signatures, encryption, etc.). Or examine how each of the technical schemes works. Or examine policies for both digital and photographic reproductions of archival works. Projects: Develop a sample policy for the Special Collections library here. Or produce samples of each of the possible labeling and protection devices for presentation to a Special Collections meeting to develop policy.
There is not nearly enough educational materials related to Image Databases on the WWW or in print. A number of interesting papers and projects could organize and bring more of this material to more people.
- Paper: Critically review WWW sites and/or print resources for Image Database information.
- Project: Compile an online annotated index to Image DB resources.
- Project: Create an online index to collections of images on the Web.
- Project: Create an online index to the material generated by student papers in this class, organized in a way that makes it useful to future Image DB students.
- Project: Reorganize the course material and resources for this class to make them useful for the general public who want to learn about Image DBs
- Project: Develop a set of visual materials to support courses and lectures on Image DBs. These might be made into slides or overheads, posted on the Web, be incorporated into books and articles, etc. Possible materials include: sample images [that are not copyrighted] in various resolutions and bit-depths, simple line drawings of different scanner types, charts and tables that sumarize some of the information from class lectures, flow-charts illustrating various steps in the creation of an Image DB, etc.
The National Endowment for the Arts (NEA) has a possible project titled the NEA RadioArts Database Project. This will include the creation of a database model.
There are a number of interesting projects on the Berkeley campus that you can work on in conjunction with the work with Anne Frantilla, Asst Reference Archivist (annef@umich.edu), or Nancy Bartlett, Reference Archivist (nbart@umich.edu)]
Paper: Examine the role and issues involved with the use of controlled vocabulary lists (such as Union List of Artist Names [ULAN], Art & Architecture Thesaurus [AAT], Thesaurus of Geographic Names [TGN] as forms of authority control. Project: Design or implement a project using one of these forms of authority control. (For example, the Art Museum uses ULAN on a separate DOS machine; a student could either integrate this into their Embark system, or implement it into a WWW service. Or a student could do the specs for a system that would improve the workflow of using such an authority list.
Design and implement a front-end WWW searching tool that puts search requests through a hierarchical controlled vocabulary that we have available (such as the Art & Architecture Thesaurus or the Thesaurus of Geographic Names). This should give the user the option of finding terms that aren't explicitly requested (such as narrower or related terms).
Paper: Examine schemes for encapsulation or encryption of images (either to ensure the integrity of the "document", or to limit access to authorized users). Projects: Develop a prototype system to limit access to authorized users, or to guarantee authenticity of an image. or Develop a grant proposal to implement encapuslation or encryption as part of the Museum Education SIte License Project.
Paper: Compare different approaches to representation of 3-dimensional space ("virtual reality"-type systems). Examine both Quicktime VR and the Virtual Reality Modeling Language (VRML).
Project: Create a Quicktime VR application that allows users to navigate through a collection of visual materials. This must have "hot-spots" that allow the user to click on an object and get fielded information about that object.
Project: Develop a Quicktime VR project that pans large Asian scrolls (to improve legibility) or pieces of sculpture in the Art Museum collection. Possibly experiment with incorporating this into their collection management system (Embark). Work with Toni Kramer (tonik@umich.edu) on this project.
The Labadie Collection holds 500-600 color and black and white posters, of varying sizes (with few larger than 36" x 36") depicting many different social protest movements, including gay liberation, anarchism, communism, feminism, labor, black history, anti-racism, youth and student protest, and international liberation struggles. None of these are cataloged yet. A map case was recently purchased (though not in place) for storage of the posters. The project would include scanning the posters in order to make them widely available to researchers. At this time it is not easily accessible and there are no records or finding aids.
A second possible project for the Labordie Collection is to break the collection down into subject areas and chose one subject area to work with. This would include scanning the posters for the subject area of your choice and making them widely available.
A third possibility for the Labordie Collection would be to seperate the collection into their subject areas and create a list or finding aid for them. Additionally placing them into the map case for proper preservation. Work with jherrada@umich.edu on this project.
The Pacific Film Archive Library (PFA) at the UC Berkeley's University Art Museum is building a document image database of its collection of film festival program notes, newspaper and magazine articles, publicity materials, and other film documentation. Access to all images they have permission to disseminate, and all document indexing will be available over the WWW. Possible project would include developing a cgi interface to access PFA's Film Document Image Database over the WWW. [Contact Nancy Goldman at nlg@uclink.berkeley.edu if you would like to work on any of the PFA projects.]
A second possible project for the Pacific Film Archive at UC Berkeley would include researching image viewers such as xv and evaluate their effectiveness for viewing images of binary text and continuous tone images over the WWW. Considerations for this include image clarity, and ability to scroll, zoom or out, or rotate the image.
A third possibility for Pacific Film Archive at UC Berkeley is to analyze database performance and develop ways to improve query retrieval speed.
Convert the Tshirt Image Database from Filemaker into an SGML-based database (2-student projetct). Create a Tshirt DTD, write and execute a script to automatically mark up the Filemaker data, and work with John Price-Wilkin (jpwilkin@umich.edu) to use the PAT software for WWW retrieval of the SGML database. Create forms-based query of the database. Individual papers might explore various uses of SGML (such as as a database).
Paper: Compare and contrast two museum collections management systems. Project: Evaluate two museum collection management systems (ARGUS and Embark) in relation to the collection and needs of the Art Museum. Work with Toni Kramer (tonik@umich.edu) on this project.
Paper: Examine the issues in linking text and various image file formats. Project: Prepare and link JPEG and GIF images to Embark text records for the Art Museum. Work with Toni Kramer (tonik@umich.edu) on this project.
Work on developing a set of tools to evaluate the effectiveness of the Museum Education Site License Project (MESL). There are a number of valuable contributions that student projects can make in preparing a major evaluation grant to NEH that will be submitted in May. Potential projects include any or a mixture of the following: literature search on assessment tools in similar environments, examining attitudes of museum and/or university personnel towards digital images, conducting focus group studies to assess the relative importance of things that might be evaluated in the course of the project, organizinging items to examine into groups based on how they might be evaluated (electronic tracking, questionnaires, interviews, focus groups), examining feasibility and cost of various assessment methods. Evaluation could be either for a single site, or for the comparative evaluation of different sites. Possible evaluation topics for technical evaluation include: frequency of access (by image, image group, or demographic group), user's access approach (known item, subject, thumbnail browse), use purpose (class instruction, individual research, desktop publishing, etc.), images used (thumbnail, small, medium, large) vs. purpose, which images are used most frequently, use of images vs. accompanying text, etc.
Work with the staff of Cranbrook to create a WWW "tour" of Cranbrook, historical photographs, objects, buildings, etc.
Explore the ways in which a WWW image database could be tied back to a collection-level MARC record for a photograph collection/series in an OPAC. This might take the form of a research and speculative paper, and/or could involve a project experimenting with Cranbrook's implementation of Ameritech's Horizon client-server system.
Organize an online collection of images and formatted text by William Blake in a way that foregrounds Blake's (image-based) representation of the text. The argument the Blakists make is that Blake, more than most authors, tied the sense of his words to the visual representation of those words. He did his own engraving, and John Wright (English Dept. here) argues that there is significant hidden information in the strokes and design. Is metainformation (e.g., titles, first lines) valuable? How about character-based representations of the text? Should the visual be foregrounded so much that the lexical is inaccessible (e.g., that no word or phrase searches would be possible? It might be useful to bring together writing about this problem with the "evidence" of the images themselves. Work with John Price-Wilkin (jpwilkin@umich.edu) on this. [For a recent review of Blake's illuminated books, see SF Chronicle, 8/13/95]
Develop a plan for imaging a coin collection at the Kelsey, and do a small prototype of this. Work with John Price-Wilkin (jpwilkin@umich.edu) and History Professor Rudi Lindner (rpl@umich.edu).
Revise the Impact Museum WWW Site. This is an index to all WWW museum-related information throughout the world. The project has been worked on by students during previous semesters. There is a real need to develop a system to keep site up-to-date with very little maintenance.
Examine an aspect of Image DB Standards that need to be developed. This could be folded into an effort of either the Computerized Interchange of Museum Information (CIMI) and/or the efforts of the Coalition for Networked Information (CNI) in this area. This might also be helpful to the CPA/RLG task force examining long-term archival formatting/access questions for digital materials, and the OCLC/NCSA group looking at metadata for identification of networked digital information. (relevant areas that Howard is working on)
Look at technologies for intellectual property protection and/or authentication. You could synthesize and expand upon the discussion that appears in the Proceedings of the CNI/Interactive Multimedia Association conference last year on Technological Strategies for Protecting Intellectual Property in the Networked Multimedia Environment. One of the many uses of these results would be for a museum association that is designing model licensing agreements for multimedia programs.
Explore some aspect of what needs to be done to ensure longevity of electronic information long after the hardware it was designed to work with has become obsolete. Work on this might feed into the Task Force on Archiving Digital Information.
Extend Howard's Image Database bibliography. Add a comprehensive list of articles published within the past 2 years. Assign subject terms to these articles and add to the Pro-Cite database. Review previous assignment of terms and maintain consistency. Review/annotate a handful of key articles.
Develop a method for mounting Howard's Image Database bibliography onto the WorldWide Web. Figure out how to maintain updating to this Pro-Cite database, as well as how to allow searching by subject term from WWW browsers.
The University of Michigan's Museum of Art has a number of possible projects. The first possibility is to transfer a particular set of records from ARGUS to Embark with proper cleanup. Toni Kramer at tonik@umich.edu is the contact point for the UofM Museum of Art projects.
The second possible project for the UofM Museum of Art is to prepare a visitor kiosk subset of fields and introductory screens.
A third project for the UofM Museum of Art would be batch programming to transfer ARGUS to Embark, or ARGUS to FileMaker to cleanup and them to Embark.
The fourth possibility for the UofM Museum of Art is to develop and outline a process to use for conversion from one system to another.
A fifth project for the UofM Museum of Art would be development of portfolio sets in Embark. (Thematic groupings of images and text in visitor module. There can be up to 9.)
A final possible project for the UofM Museum of Art would be a comparision of image needs, formats, qualities for WWW images and for Embark use.

Return to main Image DB class page