[ome-users] Problem reading czi files of line scans

[Roger Leigh]

On 30/09/2013 12:48, Ponti Aaron wrote:
> Dear list,
>
> we run line scans on Zeiss hardware using Zen Black. I attach a czi file that contains an X = 2048, Y = 1, Z = 25 dataset. Bio-formats only reads the first 2048 pixels for Z = 1. I attach a TIFF export as well for comparison.
>
> The OME metadata reports:
>
> <Pixels DimensionOrder="XYCZT" ID="Pixels:0" PhysicalSizeX="0.10378322968147374" PhysicalSizeY="0.10378322968147374" PhysicalSizeZ="1.4733031502239864" SizeC="2" SizeT="1" SizeX="2048" SizeY="1" SizeZ="1" Type="uint16">
>
> Note the SizeZ = 1 entry. Interestingly, however, deep into the ImageJ metadata one can find the correct Z value:
>
> Information|Image|SizeZ|0        25

Many thanks for providing the sample image.  I have tested with our most
current 4.4 development version, and this also shows exactly the same
behaviour you observed above.  I have opened a ticket for the issue in
our bug tracker (https://trac.openmicroscopy.org.uk/ome/ticket/11513)
and you should be CC'd on any changes to it.

Just to confirm, am I correct that the correct channel count of the
image is 2, and both channels are 16-bit greyscale?

> I tried overriding the value of Z in my loadGeneric.m MATLAB function that uses bio-formats (4.4.8) in the backhand but this causes bio-formats to fail with following error.
>
> Error using loadGeneric (line 305)
> Java exception occurred:
> loci.formats.FormatException: Invalid image number: 2 (series=0, imageCount=2)
> at loci.formats.FormatTools.checkPlaneNumber(FormatTools.java:737)
> at loci.formats.ChannelSeparator.openBytes(ChannelSeparator.java:167)
> at loci.formats.ChannelSeparator.openBytes(ChannelSeparator.java:159)
> at loci.formats.ChannelSeparator.openBytes(ChannelSeparator.java:143)
>
> The problem is therefore NOT just a wrong interpretation of the Z value of the dataset.

I'm afraid I'm not too familiar with the MATLAB bindings, but I would
suspect that given that the dimensionality is incorrect, this will
result in the array indexing being incorrect as well.  As soon as the
CZI reader is fixed, it should work correctly in MATLAB.


Kind reagards,
Roger Leigh

--
Dr Roger Leigh -- Open Microscopy Environment
Wellcome Trust Centre for Gene Regulation and Expression,
College of Life Sciences, University of Dundee, Dow Street,
Dundee DD1 5EH Scotland UK   Tel: (01382) 386364

The University of Dundee is a registered Scottish Charity, No: SC015096