[ome-users] possible bioformats bug--zeiss tiff files

[Roger Leigh]

On 04/01/2017 13:36, 李长林 wrote:
>   I am new to the bio-formats. I have inherited a workflow that calls
> bioformats through FIJI. I find one possible bug using Bio-formats for
> reading Zeiss AxioVision TIFF and Zeiss AxioVision ZVI microscopy images.
>
>   1. *.tif or *.tiff ?
>
>    The extensions in the plugin is different form the website
> （http://www.openmicroscopy.org/site/support/bio-formats5.3/supported-
> formats.html）（Fig.1 & Fig.2）

The extension for all the Zeiss AxioVision TIFF samples we have, and the
one that AxioVision uses is ".tif".  So the Zeiss TIFF reader only
handles .tif extensions.  If there are any real-world examples of
AxioVision creating .tiff files, then we can update the reader to use
them as well.

The documentation which has ".tiff" as a valid extension is incorrect,
and we will updated it to use ".tif".  Thanks for bringing the
discrepancy to our attention.

>  2. I obtain microscope images with extensions of *.zvi or *.tif
> using Zeiss AxioVision, no  *.tiff. Therr is an error when reading
> series of Zeiss AxioVision TIFF (*.tif) images using Bio-
> formats（Fig.4）. If I change the  extensions of images to *.tiff
> manually, it will work, but sometimes error, too（Fig. 5）

I think this is likely because the Zeiss TIFF reader internally does its
own file grouping, since Zeiss TIFF is itself a collection of one or
more TIFF files with associated XML metadata.

The file grouping options here are probably only going to work correctly
with simple plain TIFF files (with no extra metadata which will cause
the Zeiss TIFF reader to be used in place of the basic TIFF reader).
This could result in the error you saw if the stitching process
determines that the channel or sample counts are inconsistent between
TIFF files.

>  3. Reading *.zvi will change my data(Fig.6)

Is this a result of reading in with the Zeiss TIFF reader, or was it set
this way when exporting from AxioVision?

>   Can you please provide some assistance - this is part of a high
> content assay so I really need to be able to run this in batch mode.

If you could possibly create a zip file containing the ZVI and TIFF
files, and upload it using http://qa.openmicroscopy.org.uk/qa/upload/
then I'll be happy to look at them in more detail to see what's
happening for point (3) above.


Kind regards,
Roger

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