|
| Pilot Project Report -- Functions |
ODIN is designed to be simple and generic so it meets the basic document storage and retrieval needs of most of the candidate departments. Additional features can be added later but they were not included in the pilot. The scanning module allows documents to be scanned into electronic images, one at a time or in batches. The image is displayed on the screen to verify readability.
The indexing module displays each image and requires the user to enter identifying information. For the pilot, the images are identified by student, document title, date, and page number. The student number, social security number or name is entered and checked against a copy of the mainframe student database. If only part of a name is given, ODIN presents a list of all matching names and numbers so the user can click on the correct one.
The retrieval module displays the table of contents for a file based on student number, social security number, or name. All three keys can be used to search for a document even though only one is needed when the document is stored. If a partial name is entered, the user gets to choose from a list. In many cases, questions can be answered from the table of contents which shows the document title and date for every page in the file. Clicking on a document in the table of contents displays the image scaled horizontally to fit in the browser. Clicking again enlarges it to a scrollable full resolution picture.
Most of the administrative filing systems on campus contain a wide
variety of documents. Some are simple printed text, some include
signatures, and many contain handwritten information in pre-printed
forms. Very few of the existing records lend themselves to Optical
Character Recognition (OCR) or handwriting recognition software. For
these documents to be available electronically they must be stored as images.