Image sizes

[Marcus Smith]

Hi!

I have a practical question about describing digital image sizes. Of course, we can do this with the dimension property, in pixels height/width, bytes, etc. However, in practice when building an application, it's useful to be able to easily address images that are "thumbnail size", or "full size" (or some medium "preview" size in between).

Short of setting a range of pixel dimensions that might be appropriate for e.g. thumbnails, and querying the data based on images that fall within that range, are there any other solutions for such size classifications in Linked Art? How do existing applications, such as Lux, handle this question?

This arises partly from the practicalities of building a web interface, and partly because we have an institutional partner mapping their data to Linked Art. Technical limitations that need not concern us here mean that they are unable to include any technical information about their images – no sizes in bytes or pixels, not even whether they're JPEG format or PNG! – but they *can* (somehow) say whether an image is small (thumbnail) medium, or original size. But I can't see a way to include that information in the Linked Art mapping.

with thanks in advance,

Marcus

---

Marcus Smith

Verksamhetsutvecklare / Development Officer

Riksantikvarieämbetet / Swedish National Heritage Board

[l.f...@aboriginal-art.de]

 

Dear Marcus Smith and Linked-Art team,

 

In answer to Marcus’ question, I recall some years ago using various (windows-pc based) apps or pipelines to extract

from datablobs their correct .png or .jpg etc format and pixel sizes (and various metadata). I believe finding/using such tools
to fit Marcus’ datasystem would be a better long-term solution for handling import of images and later scaling of web-interface thumbnails.

 

NOTE: I joined this mailinglist because I seek (longterm) to encourage an institutional partner to use Linked-Art methods.

Currently I am a lowly freelance researcher, as shown here: https://ORCID.org/0009-0002-5240-0578

 

Best wishes, and just ignor this if it is not helpful 😉

/Lindsay

--
You received this message because you are subscribed to the Google Groups "Linked Art" group.
To unsubscribe from this group and stop receiving emails from it, send an email to linked-art+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/linked-art/b8e2e55b-50ad-44ee-aa4a-603c38de640bn%40googlegroups.com.

[Robert Sanderson]

Thanks Marcus and Lindsay!

The basic digital image pattern:

   https://linked.art/model/digital/#digital-images

The way I would handle "thumbnail" vs "full" is to add new `classified_as` entries to the DigitobalObject that currently just has the "Digital Image" classification.

The way that LUX deals with this is to cheat slightly. We *only* put thumbnail size as a representation into the record, and for full size we reference the IIIF Manifest for the object as per https://linked.art/model/digital/#iiif-manifests 

So for example in https://lux.collections.yale.edu/data/object/047181f4-2fdf-43fd-abbe-f16d7f9edc10 there's a `representation` property that points to the thumbnail which the results card uses (e.g. the first record here: https://lux.collections.yale.edu/view/results/objects?q=%7B%22text%22%3A%22dort%22%2C%22_lang%22%3A%22en%22%7D ), but is only classified as a digital image, and a `subject_of` that references the manifests which the object record view uses:  https://lux.collections.yale.edu/view/object/047181f4-2fdf-43fd-abbe-f16d7f9edc10 

Would a "thumbnail" vs "full" vocabulary term be useful? Almost certainly, I think. I'll make a vocabulary issue in git, but it seems like an easy one to just solve :)

Rob

--
You received this message because you are subscribed to the Google Groups "Linked Art" group.
To unsubscribe from this group and stop receiving emails from it, send an email to linked-art+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/linked-art/b8e2e55b-50ad-44ee-aa4a-603c38de640bn%40googlegroups.com.

--

Rob Sanderson

Senior Director for Digital Cultural Heritage

Yale University

[Marcus Smith]

Thanks Rob!

> Would a "thumbnail" vs "full" vocabulary term be useful? Almost certainly, I think. I'll make a vocabulary issue in git, but it seems like an easy one to just solve :)

Yes, I think that would be perfect! We at least would probably want entries for something like, "thumbnail", "high-res", and "original" (which might well be the same image as "high-res", but not always). Possibly something "medium" in between. But I daresay there are other cases that might need to be catered for as well.

best wishes,

Marcus