[ome-devel] accessing ome-metadata from matlab.

[Melissa Linkert]

Hi Heinrich,

> > I am building a LabView-controlled fluorescence microscope. As a data format for saving the acquired images, I chose ome.tif, and therefore wrote a dll putting the image and metadata into a tif file. It all seems to be working out quite fine by now, the command line tool tells me I have a valid ome.tif, and using the loci plugin for imageJ, I can display everything OK.
> > There are some issues left, though. I have written some matlab code for image evaluation, and some of that access metadata. Lately, I cannot access that any more, though. I have made some minor changes to the OME-xml, and updated to a new loci_tools.jar and bio-formats.jar, but changing back does not help which confuses me. So I wouldlike to ask you guys whether you might find anything wrong in my ome-xml that the command line tool does not, and whether you can access metadata via matlab with the script I wrote to check.
> > The script returns sensible values only for getPixelsSizeX, it returns the current date and time when asking for the acquiredDate, always returns one for getImageCount and getChannelCount and otherwise, it returns nothing.
> > 
> > I attach the metadata checking matlab script and a sample image Any 
> > help is highly appreciated
> 
> Thank you for the report.  Just so that you know, we do have a fix in the works, it just hasn't quite made it into the trunk builds yet.  If you're interested, you can see the discussion of the fix here:
> 
> https://github.com/openmicroscopy/openmicroscopy/pull/21
> 
> In short, the problem was that the OME-XML failed to be parsed, which caused the OME-TIFF file to be read as a plain TIFF file - that's why the metadata looked completely wrong.  The fix proposed above will prevent the XML parsing from failing, allowing the file to be properly read as an OME-TIFF.
> 
> We'll let you know once a new trunk build is available with the fix.

Updating to the latest trunk build now should fix the problem.  Of
course, please let us know if you find that it does not work for you.

Regards,
-Melissa

On Fri, Nov 04, 2011 at 07:26:27PM +0000, Grabmayr, Heinrich wrote:
> Hi Melissa,
> 
> thanks for letting me know!
> 
> Best Heinrich
> 
> -----Ursprüngliche Nachricht-----
> Von: Melissa Linkert [mailto:melissa.linkert at gmail.com] Im Auftrag von Melissa Linkert
> Gesendet: Freitag, 4. November 2011 20:23
> An: Grabmayr, Heinrich
> Cc: ome-devel at lists.openmicroscopy.org.uk
> Betreff: Re: [ome-devel] accessing ome-metadata from matlab.
> 
> Hi Heinrich,
> 
> > I am building a LabView-controlled fluorescence microscope. As a data format for saving the acquired images, I chose ome.tif, and therefore wrote a dll putting the image and metadata into a tif file. It all seems to be working out quite fine by now, the command line tool tells me I have a valid ome.tif, and using the loci plugin for imageJ, I can display everything OK.
> > There are some issues left, though. I have written some matlab code for image evaluation, and some of that access metadata. Lately, I cannot access that any more, though. I have made some minor changes to the OME-xml, and updated to a new loci_tools.jar and bio-formats.jar, but changing back does not help which confuses me. So I wouldlike to ask you guys whether you might find anything wrong in my ome-xml that the command line tool does not, and whether you can access metadata via matlab with the script I wrote to check.
> > The script returns sensible values only for getPixelsSizeX, it returns the current date and time when asking for the acquiredDate, always returns one for getImageCount and getChannelCount and otherwise, it returns nothing.
> > 
> > I attach the metadata checking matlab script and a sample image Any 
> > help is highly appreciated
> 
> Thank you for the report.  Just so that you know, we do have a fix in the works, it just hasn't quite made it into the trunk builds yet.  If you're interested, you can see the discussion of the fix here:
> 
> https://github.com/openmicroscopy/openmicroscopy/pull/21
> 
> In short, the problem was that the OME-XML failed to be parsed, which caused the OME-TIFF file to be read as a plain TIFF file - that's why the metadata looked completely wrong.  The fix proposed above will prevent the XML parsing from failing, allowing the file to be properly read as an OME-TIFF.
> 
> We'll let you know once a new trunk build is available with the fix.
> 
> Regards,
> -Melissa
> 
> On Wed, Nov 02, 2011 at 04:33:41PM +0000, Grabmayr, Heinrich wrote:
> > Hey everyone,
> > 
> > I am building a LabView-controlled fluorescence microscope. As a data format for saving the acquired images, I chose ome.tif, and therefore wrote a dll putting the image and metadata into a tif file. It all seems to be working out quite fine by now, the command line tool tells me I have a valid ome.tif, and using the loci plugin for imageJ, I can display everything OK.
> > There are some issues left, though. I have written some matlab code for image evaluation, and some of that access metadata. Lately, I cannot access that any more, though. I have made some minor changes to the OME-xml, and updated to a new loci_tools.jar and bio-formats.jar, but changing back does not help which confuses me. So I wouldlike to ask you guys whether you might find anything wrong in my ome-xml that the command line tool does not, and whether you can access metadata via matlab with the script I wrote to check.
> > The script returns sensible values only for getPixelsSizeX, it returns the current date and time when asking for the acquiredDate, always returns one for getImageCount and getChannelCount and otherwise, it returns nothing.
> > 
> > I attach the metadata checking matlab script and a sample image Any 
> > help is highly appreciated
> > 
> > Best
> >    Heinrich
> 
> 
> 
> > _______________________________________________
> > ome-devel mailing list
> > ome-devel at lists.openmicroscopy.org.uk
> > http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-devel
>