Message-Id: <mailto:199507082317.SAA03622@library.wustl.edu> Date: Sat, 8 Jul 1995 14:42:56 UNDEFINED From: "Kevin C. Marsh" <mailto:IAI@NEOSOFT.COM> Subject: Re: Image database for Queensland University of Technology To: Multiple recipients of list IMAGELIB
In article <mailto:Pine.3.89.9507061501.D11437-0100000@redash.qut.edu.au> Graham Dawson mailto:<mailto:g.dawson@QUT.EDU.AU> writes:>At QUT we are considering building an image database to be
>used for a variety of university functions, both teaching and
>administrative.
>We have in mind something along the lines of images stored in
>JPEG format and indexed by a relational database. This
>database would likely be accessed via a search engine that
>underlies a Web Home Page.
I have used freeWAIS-sf to build such a system at NASA and I'm currently helping the Florida State Archives Photographic Collection build a similar system. Unfortunately, neither system is available for public Web access yet.
My suggestion is to create an HTML document (filename.htm) and a small GIF file (filename.gif) for each image (filename.jpg). The HTML documents can be generatred as a report from a database like Paradox, or from a MARC compatible library system. You then use freeWAIS-sf to index the HTML documents, and use the SFGate CGI script to search that index from a Web form. Each HTML document will include the GIF file as an in-line image, and that image will also act as a link to the JPEG file.
See http://neosoft.com/~iai/florida/florida.htm for an example of how this will work.
By using a WAIS search engine and standard field names you will open up the possibility of simultaneously searching multiple image collections. For example, I have configured a Web page to perform a lattitude/longitude search against both the USGS maps collection and the NASA earth observation photos collection.
Please feel free to send e-mail if you would like more information about this approach.
Kevin C. Marsh, Executive Director, Information Access Institute mailto:IAI@neosoft.com