Splitting Hyperstack in smaller images and reassembling after processing

Dear ImageJ community,

I am currently trying to perform Spectral Unmixing on large SPIM image datasets, as our current technical setup does not allow to filter out spectral bleed through. Working with smaller datasets has shown us that the Spectral Unmixing does indeed enhance our images, but on large datasets I am not able to perform Spectral Unmixing as the Computer runs out of memory before it completes the task (Our Imaging workstation has 128 GB RAM). So my idea was to somehow split the Hyperstack into smaller areas, perform the Spectral Unmixing on each of them and then reassemble them into the original large Hyperstack again. However, I am unaware of such a function in Fiji and so far have not been able to find a solution. Does anybody know about a method how this problem could be solved?

Thanks in Advance.

Have you tried to open and process the Hyperstack in virtual mode (disk resident)?

File->Import->Tiff Virtual Stack

run("TIFF Virtual Stack...", "open=C:\\HyperStack.tif");

See also for macro usage:


1 Like


thanks for the suggestion, but I quickly came to realize that it would be much easier to just split the timeseries into several smaller timeseries files.

The problem is solved :slight_smile: