[ome-users] Two issues with ND2 (Nikon) files

Fri Sep 8 13:18:35 BST 2017

Hi Maxime,

Thats great news that we were able to find a solution. I took a look at the ImageJ forum and found a similar issue also being reported recently and it looks like they have already worked on a fix also.
I have left a reply on the forum thread to link with this discussion and the associated Trello card for future reference.
http://forum.imagej.net/t/bfconvert-command-line-tool/5872/31

With Thanks,
David Gault

On 7 Sep 2017, at 19:03, Maxime Woringer <maxime.woringer at berkeley.edu<mailto:maxime.woringer at berkeley.edu>> wrote:

Hi David,

It works! It works! It works! This is awesome, I'm so happy! (this is a
different story, but the file opens correctly only if I do File>Open, I opens
wrongly if I do File>Import>TIFF virtual stack)

Thank you so much about all the help, I'm very glad you took so much
time to troubleshoot the issue!

If I understand correctly, there is an issue with ImageJ that prevents
it from importing the files properly when it does not use SCIFIO. Would
you advice that I report this issue to the ImageJ team?

Thanks again,
Maxime

David Gault (Staff) <d.gault at dundee.ac.uk<mailto:d.gault at dundee.ac.uk>> writes:

Hi Maxime,

I have carried out some further tests on this and I was reproducing the issue with clean installs of the latest 32 and 64bit Windows packages on a separate machine. It seems that the difference between this and my previous tests was an option to use SCIFIO when opening files.

In FIJI if you set the option below that should hopefully resolve your issue:
Edit > Options > ImageJ2 > Use SCIFIO when opening files (BETA!)

With Thanks,
David Gault

On 5 Sep 2017, at 17:04, Maxime Woringer <maxime.woringer at berkeley.edu<mailto:maxime.woringer at berkeley.edu>> wrote:

Hi David,

Thank you very much for these precisions and investigations, I
appreciate your efforts. I tested the 'converted_with_bfconvert.tif'
file with a newly downloaded Fiji (with ImageJ 1.51n), and I am still
encountering the issue. I tried this on two different computers, and I
am attaching screenshots of the full version number.

<Capture du 2017-09-05 08-42-33.png><Capture du 2017-09-05 08-53-36.png>

Would you have any pointer about what could be wrong?

Thanks again,
Maxime

David Gault (Staff) <d.gault at dundee.ac.uk<mailto:d.gault at dundee.ac.uk>> writes:

Hi Maxime,

The converted TIFF absolutely should be able to be read natively so this is not expected and you haven’t missed anything. My question was purely to help me narrow done the cause of the problem.

I have carried out some further testing this morning using various older FIJI installs and I was able to reproduce what you were seeing with some of the older packages. From the June 2014 install (ImageJ 1.49b) the image was corrupted after import, however using the November 2014 install (ImageJ 1.49k) and onwards the image imports correctly without any issues. Although I don’t yet know the precise cause it looks like updating your version of ImageJ  should be enough to resolve this import issue.

With Thanks,
David Gault

On 4 Sep 2017, at 17:53, Maxime Woringer <maxime.woringer at berkeley.edu<mailto:maxime.woringer at berkeley.edu>> wrote:

Hi David,

Thank you very much for this follow-up. I will follow the progress on
the Trello card.

I confirm that the two bugs still occur in BF5.7.0. I can't tell how
excited I am to read that you identified where the issue comes from!
Actually, I found out that people have been annoyed by these ND2
conversion "issues" since at least 2013 (no idea why nobody reported it,
though).

Regarding the second issue, I am getting more and more confused:
- When I import the resulting TIFF file into Fiji with File -> Open
(without the Bio-Formats importer), all the frame, starting at frame 2,
appear corrupted (I am attaching a screenshot of frame 2)
- When I use the Bio-Formats importer, then the TIFF file imports just
fine.

This suggest that I am missing something obvious... Can't the TIFF
files produced by bfconvert be read natively by ImageJ?

<5.7.0-frame2.png>
Thanks a lot,
Maxime

David Gault (Staff) <d.gault at dundee.ac.uk<mailto:d.gault at dundee.ac.uk>> writes:

Hi Maxime,

We have been able to further debug the first issue to identify how the problem occurs but we do not yet have a fix in place for it.
You can follow the progress of that issue on the Trello card here: https://trello.com/c/RyXt2BAg/177-nd2-incorrect-channel-count

With regards the second issue I have still been unable to reproduce it. If I open the 2 converted files you had originally attached (one converted with bfconvert, the other with FIJI) both open and display the same when using the latest FIJI install.
When you converted the TIFF image using FIJI, and when you read the image back, was that using the Bio-Formats exporter and importer?

With Thanks,
David Gault

On 1 Sep 2017, at 18:59, Maxime Woringer <maxime.woringer at berkeley.edu<mailto:maxime.woringer at berkeley.edu><mailto:maxime.woringer at berkeley.edu>> wrote:

Hi David,

Do you have any update about the two bugs mentioned earlier? Were you
able to reproduce the second bug mentioned in my previous email?

Thanks a lot,
Maxime

David Gault (Staff) <d.gault at dundee.ac.uk<mailto:d.gault at dundee.ac.uk><mailto:d.gault at dundee.ac.uk>> writes:

Hi Maxime,

Thank for reporting this issue and for providing sample files to test with.
I can certainly reproduce the first the issue using the sample files provided and this does indeed look to be a bug.
For the second issue I was unable to reproduce using bfconvert from the latest Bio-Formats release, and in theory the conversion should be the same as the FIJI result.
Would you be able to run bfconvert -version to confirm that bftools is using the latest jars?

I have created a Trello card on the Bio-Formats inbox for further investigation into these issues: https://trello.com/c/RyXt2BAg/177-nd2-incorrect-channel-count

With Thanks,
David Gault

On 14 Aug 2017, at 04:01, Maxime Woringer <maxime.woringer at berkeley.edu<mailto:maxime.woringer at berkeley.edu><mailto:maxime.woringer at berkeley.edu><mailto:maxime.woringer at berkeley.edu>> wrote:

Dear Bio-Formats users/developers,

I am trying to convert ND2 files (acquired with NIkon NIS elements) to
TIFF files, ideally in an automated fashion, but I am encountering
issues that seem to me like bugs (but I'd be happy to be proven wrong).

1. Some single images appear corrupted when I try to convert them to
TIFF. Below is an example. The two files (BF007 and BF008) were created
on the same machine, same software, at a 30' interval. I am not aware of
any differing setting except the ROI size.
- BF008.nd2 opens properly in Fiji and converts properly using bfconvert
(laters version). I am attaching the .nd2 file, the resulting TIFF file,
and the convert log (.txt files) and a screenshot of how Fiji opens it.
- BF008.nd2 opens corrupted in Fiji, and the same result happens using
bfconvert. It seems that BF thinks it is a 3-channel image (it is not).
<BF007.nd2><BF007.txt><BF007-bf.tiff><BF007-Capture du 2017-08-13 18-50-59.png><BF008.nd2><BF008.txt><BF008-bf.tiff><BF008-Capture du 2017-08-13 18-50-53.png>
2. Now moving to ND2 TIFF series (one channel, multiple timepoints),
they open fine in Fiji using BioFormats, however, if I use the following
command:

bftools/bfconvert in.nd2 out.tiff

The resulting TIFF file opens totally corrupted in Fiji/ImageJ (note
that the first frame is fine).

Below are an example source file (nd2), the conversion result using
bfconvert, and the conversion result using Fiji. I am attaching the
bfconvert log (convert.log)
<convert.log>- Source ND2 file: https://cloud.biologie.ens.fr/index.php/s/H7OPHzLvJ3MPUkd
- Corrupted TIFF generated by bfconvert: https://cloud.biologie.ens.fr/index.php/s/FFOIjHUg6Dl7B1I
- Correct TIFF generated by Fiji: https://cloud.biologie.ens.fr/index.php/s/dyl1hZha3pGiP50

Finally, I should mention that at least the latter issue happens with
files generated with several versions of NIS elements.

Am I the only person experiencing those issues? Is there any kind of
additional information I should provide?

- Version information: bf 5.5.3 and/or Fiji (latest update)
- OS: I encounter the issue on Ubuntu 14.04 and Debian Jessie (8.9)
- java version: I believe the one used is packaged by BioFormats, but in
case, java -version returns
Ubuntu machine:
  java version "1.8.0_144"
  Java(TM) SE Runtime Environment (build 1.8.0_144-b01)
  Java HotSpot(TM) 64-Bit Server VM (build 25.144-b01, mixed mode)
Debian machine:
  java version "1.7.0_131"
  OpenJDK Runtime Environment (IcedTea 2.6.9) (7u131-2.6.9-2~deb8u1)
  OpenJDK 64-Bit Server VM (build 24.131-b00, mixed mode)

Thanks a lot!
Maxime

PS: I tried to upload the files to the QA systemn, but it complains that
my version of Flash is not up-to-date. I apologize about that and hope
that you will be able to retrieve the files linked/attached.
PPS: I have checked the "Common issues to check" page before sending
this email, and I believe that none of them apply.
_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk><mailto:ome-users at lists.openmicroscopy.org.uk><mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

The University of Dundee is a registered Scottish Charity, No: SC015096
_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

The University of Dundee is a registered Scottish Charity, No: SC015096
_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

The University of Dundee is a registered Scottish Charity, No: SC015096
_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

The University of Dundee is a registered Scottish Charity, No: SC015096
_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

_______________________________________________
ome-users mailing list
ome-users at lists.openmicroscopy.org.uk<mailto:ome-users at lists.openmicroscopy.org.uk>
http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users

The University of Dundee is a registered Scottish Charity, No: SC015096
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20170908/6d9c5cee/attachment.html>