Combining Images to Stack

Hi,

I am using Olympus ScanR system to acquire Z-Stack Images. However with the ScanR acquisition software, I do not get a Stack image at the end but rather an individual image for each layer. These images are the start point of my workflow, I want to combine all the images that represents the different Z-layers at the same position in the same well to a Stack image ( Images are named as (A1--W00001--P00001--Z00000--T00000--mCherry1), where W is well, P is position, Z is Z-layer and T is time point). I was not able to locat a node that does that. Also I was not able to use the ImageJ Snippet node and just copy the code from ImageJ...I am sure you faced this problem or similar before, so please let me know if you have any solution for this????

Best,

Mohamed

Hi Mohamed,

yes there is a solution: the Merger-Node. To be honest, the merger node is not very intuitive to configure, so far, but quite flexible. For your problem first of all tranpose your column to a row (e.g. using the Transpose node, or you can also use the "Column to Grid" or "Pivot" to better specify which images to put in a row and, hence, to merge). Than in the Merger-Node you can just select the column you want to merge and specify the dimensions of the resulting image (you can also use the dimension labels as place holders). In short: put "X,Y" in the "Result dimension"-field and "X,Y,Z" in the "Result dimension names"-field.

If that doesn't work for you, you can also send me your images to merge and I can compose the workflow for you.

Alternatively, in the new "1.0" version of the image processing plugin (it will be released with the release of KNIME 2.6 end of July), the GroupBy-node can also be used to merge images to an "image stack".

Hope that helps,

Best,

Martin

Hi Martin,

Thank you for your suggestion, it was really useful. So, I used "Column to Grid" node to put the images representing the different layers of the same position in one row. Then I used the Merger node as u suggested. I put "1000,1000" in the "Result dimension"-field ( these are the dimensions of my images in pixels) and "X,Y,Z" in the "Result dimension names"-field.  And, I think it did work  :) 

Thank you again.

Mohamed

1 Like