Annotation Metadata Model

Metadata Model

Metadata model for markups and annotations is still in flux; here are some basic attributes that should be included in any zoomable image markup.

1. UID ( text String?)
2. Object Type (Square/Rect/Circle/Point/Polyline/Freehand)
3. Border Color ( #ffff00 )
4. Markup Opacity ( 0- 1.0)
5. Fill Color (Hex color)
6. Rater Name ( Text String of who generated the annotation; or algorithm name/version)
7. Mark up Label ( Text)
8. List/Ontology ID (data set/value list for #7)
9. Notes (add'l freetext)
10. Markup Image Name (e.g. name/link to image your drawing on)
11. Browser Used for Markup
12. Time stamp of markup creation ( javascript getDate?)13/ ) Time Stamp of markup creation
13. Revision and/or UID of original markup version
14. md5sum of image
15. Annotation data server (where_annotated,server used, maybe just use hash of domain name....)
16. Zoom Level

More Detailed Information.

0. some UID (probably hash of rater_id and timestamp; although unlikely 2 people are generating a markup at the same microsecond)
1. object type (square/rect/circle)
2. Vertices/points
3. Border Color
4. Opacity
5. Fill Color
6. rater_name (who did it or what algorithm), will probably use a Version if it's a computer myalgo.1, myalgo.2, etc..
7. zoom_of_markup
8. mark up label (ideally from semistructured list using autocomplete)
9. List/Ontology/whatever ID
10. Notes
11. Image Name being marked up

11. Browser Used

12. Time Stamp of markup creation
13. Revision and/or ID of original version--- debating what's easier to implement
14. md5sum of image
15. where_annotated (server used, maybe just use hash of domain name....)

Note: as these images get converted to a "web" format prior to visualization, it's possible various versions of the "same" image have been created; unlikely, but who knows...