Leica LIF from RGB problem in OMERO v5.1

[mattc]

We are having problems with Leica LIF files coming from a colour camera (RGB). When a single colour image is displayed BioFormats is separating red/green/blue into individual channel layers then creating a composite, which works for single images. When you have 2 colour images (fluorescent + bright field) it gets confused and makes a big mess. We've also noticed that BioFormat when opening an RGB tif file it separates the colours into channel layers (composite) whereas opening using ImageJ the image file is RGB (no channel layers).

[rleigh]

Would it be possible to upload a sample file here: http://qa.openmicroscopy.org.uk/qa/upload/ so we can take a look at how the channels are specified?

Which software are you using to view the image? ImageJ, showinf/ImageInfo or some other program or your own custom code?


Many thanks,
Roger Leigh

[mattc]

Hi Roger,

thanks for your reply. I've uploaded an image for you. I'm using ImageJ to view the image.

Regards,

Matt.

[rleigh]

Dear Matt,

Thanks for the image. Looking at the metadata, image 0 has 3 channels (RGB), images 1-3 have 6 channels each. Most software won't be able to cope with the sample count being more than three. It should however be possible to separate the channels on import into ImageJ and then reassemble them into two RGB images as a workaround.

The question I have is regarding the original image file is whether the Leica software treats this as a 1 channel image containing 6 samples, or as a 2 channel image, each containing three samples (i.e. 2 RGB images)? Is this something that the Bio-Formats LIFReader is mis-interpreting, leading to the metadata being incorrect? If so, that's something we can look at correcting.


Thanks,
Roger

[mattc]

Hi Roger,

the original image should be two RGB images (some with Zs in them) and it does look like it is BioFormats is misinterpreting it the metadata to create an image with 6 channels but additionally it is converting it into a tiled collage of the 6 channels. I should have given the filename that I uploaded but I've uploaded another (DM6000_colour-camera-8bit.lif) to be sure we're looking at the same problem .

Thanks for your help,

Matt.

[bramalingam]

Hi,

Thank you for submitting your files.
We were able to reproduce the issue locally, with Bio-Formats 5.1.6.

A ticket has been created on this regard,
https://trac.openmicroscopy.org/ome/ticket/13112#ticket

and will get looked into, at the earliest.

Best,
Balaji