[Insight-developers] JPEG2000ImageIO moved to Review : Input images too large...

[Bradley Lowekamp]

Hello,

Well it was very useful to get these tests running! They are failing on some systems in different ways now.

http://www.cdash.org/CDash/testSummary.php?project=2&name=itkJPEG2000Test00&date=2010-08-26
http://www.cdash.org/CDash/testSummary.php?project=2&name=itkJPEG2000Test01&date=2010-08-26
http://www.cdash.org/CDash/testSummary.php?project=2&name=itkJPEG2000Test02&date=2010-08-26
http://www.cdash.org/CDash/testSummary.php?project=2&name=itkJPEG2000Test03&date=2010-08-26

So this is the sun OS on a AMD64 with either the gcc or SunCC compiler where the test is failing.

Then on my own mini building for a G5, and then having the code run in emulation PowerPC code. I believe this is the only big endian system on the dashboard, I will build an experimental actually on a G5, to see if it fails the same way. Perhaps the JPEG2000 isn't dealing with correcting for endianness?


Additionally we are getting some linking errors on  many of the windows systems:

http://www.cdash.org/CDash/viewBuildError.php?buildid=704316

Brad

On Aug 25, 2010, at 4:17 PM, Lowekamp, Bradley (NIH/NLM/LHC) [C] wrote:

> Luis,
> 
> I just did the following to get the JPEG2000 test working:
> 
> victoria:ITK blowek1$ cd Testing/Data/
> victoria:Data blowek1$ git checkout 1eb70beedde43
> Previous HEAD position was bc9550f... added baseline images for HessianToObjectnessMeasure
> HEAD is now at 1eb70be... Added input images for JPEG2000ImageIO tests.
> victoria:Data blowek1$ cd ../..
> victoria:ITK blowek1$ git status
> # On branch exp
> # Changed but not updated:
> #   (use "git add <file>..." to update what will be committed)
> #   (use "git checkout -- <file>..." to discard changes in working directory)
> #
> #	modified:   Testing/Data (new commits)
> #
> # Untracked files:
> #   (use "git add <file>..." to include in what will be committed)
> #
> #	Code/Review/itkJPEG2000ImageIO.cxx.orig
> #	Code/Review/itkSliceBySliceImageFilter.h.orig
> #	Code/Review/itkSliceBySliceImageFilter.txx.orig
> no changes added to commit (use "git add" and/or "git commit -a")
> victoria:ITK blowek1$ git commit Testing/Data
> [exp 113fc37] Adding reference to newly commited JPEG2000 in data submodule
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> Here is to a greener dashboard tomorrow!
> 
> Brad
> 
> On Aug 21, 2010, at 8:48 PM, Luis Ibanez wrote:
> 
>> 
>> Hi Kishore,
>> 
>> Now that we have updated openjpeg  to Monday's 
>> version of openjpeg-v2, we were also able to move 
>> the JPEG2000ImageIO classes from the 
>> Insight Journal paper:
>> 
>> http://www.insight-journal.org/browse/publication/741
>> http://hdl.handle.net/10380/3187
>> 
>> 
>> BTW: Thanks for submitting the paper     :-)
>> 
>> 
>> We are however, struggling with the input images,
>> since they are rather large.
>> 
>> So far we have committed only the images 
>> 
>> Bretagne1.j2k ---- 90Kb
>> Cevennes2.jp2 ---- 18Kb
>> 
>> The others are larger than 100Kb.... 
>> 
>> 
>> The conundrum is that, we actually should test 
>> with very large images, given that this is one 
>> of the purposes of introducing JPEG2000.
>> 
>> 
>> Any ideas are welcome....
>> 
>> In the meantime, if you have a chance, please
>> give it a try at the new code in ITK/Code/Review.
>> 
>> 
>>    Thanks
>> 
>> 
>>         Luis
>> 
>> 
>> <ATT00001..txt>
> 
> ========================================================
> Bradley Lowekamp  
> Lockheed Martin Contractor for
> Office of High Performance Computing and Communications
> National Library of Medicine 
> blowekamp at mail.nih.gov
> 
> 
> <ATT00001..txt>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20100826/4f0bb404/attachment.htm>