Image/slide database

[Ben Richardson]

Mary

In Western Australia, we have a system that links the name of the subject of
the image to the name identifier in the Herbarium, along with some
attributes to record the type of photograph.

To achieve this, images are classified into four general categories:
environment (e), flower (f), habit (h), and diagnostic (d).

"Environment" (e) refers to images of a typical environment in which the
target species is found. Ideally the species should be identifiable in the
image although, for some small or cryptic species, this may not always be
possible.

"Flower" (f) refers to images in which the main emphasis is on a close-up of
a flower or flowers.

"Habit" (h) refers to images in which the normal growth habit of a mature
plant can be seen.

"Diagnostic" (d) refers to images in which one or more diagnostic features
are highlighted.  For example this may refer to buds, seed pods or other
fruits, bark, leaf-shape and so on.

A replicate number is also used when there is more than one image of a
particular category.

Our images are archived in TIFF format with compression onto CD-ROM discs.

An image of the first recorded flower image of Eucalyptus camaldulensis in
our system, thus has the filename 5580if1.tif.

It is a simple but effective method to store each image, and if we need the
extra organisation, as we do with our taxon maps, we also use the in-house
family and genus identifiers to create a file system hierarchy.  Thus
Eucalyptus camaldulensis might be stored in a filesystem using a path:

273/1113/5580if1.tif

We use this filename approach in FloraBase,
http://florabase.calm.wa.gov.au/, to enable dynamic building of pages, since
all name-related content is indexed by its name identifier.

We use Max, http://www.naturebase.net/science/max/max.htm, a plant name
management tool for Windows-based PCs to manage our image database.

The database contains fields that link the filename (our primary key) to
data such as the photographer's name, where the photo was taken, scan date,
scan operator, file path, file size in pixel height x width, and so on.

The next stage in the process is to have this meta-data online so that
people can see what we have for donation of new images, and to be able to
link into other projects across Australia.

Hope this helps,
Ben

> -----Original Message-----
> From: Taxacom Discussion List [mailto:TAXACOM at USOBI.ORG]On Behalf Of
> Mary Barkworth
> Sent: Thursday, 21 March 2002 4:47 AM
> To: TAXACOM at USOBI.ORG
> Subject: Image/slide database
>
>
> We have started thinking seriously about how to store scans
> and digital
> pictures in such a way that we know where to look for them.
> Basically, I
> am thinking of being able to locate them by taxon name and by
> structure(s) shown.  I am sure that others have already
> figured out how
> to do this. If there are suggestions as to canned programs that others
> have found useful, or if people would be willing to share a database
> structure that they use, it would be great.  I just do not
> want to waste
> energy trying to design something that already exists.
>
> Oh - some slides/scans are linked to specimens in the herbarium - but
> not all of them.
>
> Mary