2005.08.15 23:15 "Re: [Tiff] TIFF Thumbnail", by Joris Van Damme

There is also the case where there are multiple reduced resolution images -- i.e. multiple "thumbnails" -- although I would probably only call the smallest the thumbnail.

Yes, indeed. It's not widely used, and I didn't want to confuse the issue any further, that's why I avoided mentioning it. Plus, the 'spirit' of tree-structured TIFF remains unchanged by that fact, as does the 'spirit' of the single-list pre-SubIFD TIFF. But you are of course correct.

- Some software writes thumbnails the 'old-style' pre-SubIFD way. This

  usually results in a two-page TIFF, the second page being a thumbnail
  of the first bigger image. You can detect this situation by checking

• number of pages (must be 2)
• NewSubfileType or SubfileType tags (must be present and reflect this relationship)
• sizes (2nd must be smallest)

The "number of pages" confuses me -- perhaps you meant "number of IFDs" instead? If not, are you suggesting that the pagenumber tag be used?

I should have used the word 'IFD' instead, there, that would have been not so confusing. I did mean 'IFD', and wasn't refering to the pagenumber tag or sorts.

I think, in general, you can use the word 'page' to mean 'an IFD in the primary list, and all it's child IFDs'. Thus, TIFF is multi-page, where each individual page can consist of multiple reduced copies, multiple layers, etc, by using the SubIFD tag and/or private IFD tags. But that is, of course, the post-SubIFD notion of TIFF. Thus, using that same word 'page' to have that same meaning of 'IFD in the primary list', when describing the pre-SubIFD situation, like I did, is unfortunate.

If multiple subfiles are written, the first one must be the full-resolution image.

Subsequent images, such as reduced-resolution images, may be in any order in the

TIFF file. If a reader wants to use such images, it must scan the corresponding

IFD's before deciding how to proceed."

Yes, that is the basis of the old pre-SubIFD thumbnail (or multiple reduced version) scheme. It has a number of problems, though, the need to forward-scan being just one of them, mentioned in the quote. That is why I think it might be dangerous to quote that part of the spec without mentioning the newer SubIFD scheme, and would like to refer once again to http://partners.adobe.com/public/developer/en/tiff/TIFFPM6.pdf.

Joris Van Damme
info@awaresystems.be
http://www.awaresystems.be/
Download your free TIFF tag viewer for windows here:
http://www.awaresystems.be/imaging/tiff/astifftagviewer.html