Stitching / seams problem

[David Gutman]

Good question--- no good answer.

So it's definitely not an openslide problem--- but there are about 50
places it can be introduced....and I haven't systematically tried to
evaluate the problem.

So- for the case of that file... it is likely this was a "pre" openslide
conversion-- just using VIPS... However there are a number of tradeoff's
that have to be made due to compression / formatting/pipeline stuff... and
since I have like 12,000 slides on that site I haven't been motivated
enough to go back and make sure everything's fixed..

Anyway..

There are at least 2 JPEG decode/recode steps--- and each of those has a
choice I must make in terms of image compression or lack thereof.  At one
point I screwed up and didn't set compression, and each of my resulting
PYRAMIDS was 30 GBS..

Anyway... the VIPS command itself has a JPEG Compression step--- so I think
I default to 75...... and then I generate the 256x256 tile pyramids.


In addition... the IIPIMAGE server it self will take the pyramidal tiff
(aka the 256x256 tile) and again recompress it... as you can set the
iipimage server's output jpeg resolution in the server log-----  I think I
set it to 75 or 90... I can't remember.


I've asked Ruven @ IIPIMAGE about it... and in theory there should be a way
to basically have it check that if the output tile size requested matches
that of the PYRAMID tiff file itself (aka 256x256)... it should return the
"raw" JPEG and not decompress/recompress--- but I haven't had the time or
resources to look into it... it sounds like from Ruven it wouldn't be a
huge pain... but my knowledge of C++ is... not good.


There are a couple of files I have done at 95 or 100 for the initial JPEG
compression step, and also I believe I tweaked the output settings on
iipimage server and it looked better... but I haven't done any real
performance testing at this point..





On Wed, Oct 3, 2012 at 5:36 AM, Yves Sucaet <sucaet at histogenex.com> wrote:

> Hi David,
>
> I notice that you have a stitching / seams problem at high magnification.
> Is this a problem with OpenSlide? It seems similar to the problem that my
> team is trying to resolve with the MRXS 2.2 format. How are your users
> responding to this? Is it something they can live with or have they
> complained about it?
>
>
>
> Kind regards,
>
> Yves
>
>
>
> From:        David Gutman <dagutman at gmail.com>
> To:        "PD Dr. M. Weihrauch" <martin.weihrauch at uni-koeln.de>
> Cc:        openslide-users at lists.andrew.cmu.edu
> Date:        03-10-12 09:58
> Subject:        Re: Openlayers / Deepzoom format
> Sent by:        openslide-users-bounces+sucaet=
> histogenex.com at lists.andrew.cmu.edu
> ------------------------------
>
>
> *http://cancer.digitalslidearchive.net/*<http://cancer.digitalslidearchive.net/>
>
>
> David A Gutman, M.D. Ph.D.
> Assistant Professor of Biomedical Informatics
> Senior Research Scientist, Center for Comprehensive Informatics
> Emory University School of Medicine
>
> ------------------------------
>
> WARNING: This e-mail, including any attachments, may contain CONFIDENTIAL
> INFORMATION, including privileged and/or health information. It is for the
> sole use of the intended recipients. Any unauthorized copying, disclosure,
> distribution, reproduction, use or retention of this email or the
> information in it, is strictly FORBIDDEN. If you are not an intended
> recipient, please notify the sender immediately (REPLY this e-mail) and
> permanently DELETE the related e-mail. Please be aware that this email and
> replies to it may be monitored by the sender's company for quality
> assurance, policy compliance and/or security purposes.
>
> _______________________________________________
> openslide-users mailing list
> openslide-users at lists.andrew.cmu.edu
> https://lists.andrew.cmu.edu/mailman/listinfo/openslide-users
>
>


-- 
David A Gutman, M.D. Ph.D.
Assistant Professor of Biomedical Informatics
Senior Research Scientist, Center for Comprehensive Informatics
Emory University School of Medicine
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.andrew.cmu.edu/pipermail/openslide-users/attachments/20121003/b61be10e/attachment.html