[ome-devel] Multi-spectral imaging formats

[Curtis Rueden]

Hi Lee,

> For the IM3 reader, my naive idea was to package all 63 channels as
> one interleaved image, but I have a feeling that's 60 more than most
> software expects, but it would be very sad to run through the 200Mb
> file 63 times in order to get 63 planar images.

The way Bio-Formats handles this for the SDT file format, which has
interleaved data of 16 channels or more, is indeed to report the channels
as interleaved.

Return a dimension order of XYCZT, and return true for isInterleaved().

https://github.com/openmicroscopy/bioformats/blob/v5.0.3/components/formats-api/src/loci/formats/IFormatReader.java#L227-L239

Bio-Formats has built-in tools to separate out the channels by buffering
the relevant interleaved plane in memory. It's called ChannelSeparator:

https://github.com/openmicroscopy/bioformats/blob/v5.0.3/components/formats-bsd/src/loci/formats/ChannelSeparator.java

Regards,
Curtis

P.S. The SCIFIO API generalizes this much more, moving away from the
hardcoded 5D OME data model in favor of the ImgLib2 N-dimensional data
model. But unfortunately that API is still in flux.


On Tue, Aug 26, 2014 at 11:48 AM, Lee Kamentsky <leek at broadinstitute.org>
wrote:

> Hi all,
> I've been working on the .IM3 format for the Perkin-Elmer Nuance
> multi-spectral microscope. It has 63 channels per pixel for the spectral
> data and the data is interleaved (and I thought it would be the most
> efficient to keep it that way). For the IM3 reader, my naive idea was to
> package all 63 channels as one interleaved image, but I have a feeling
> that's 60 more than most software expects, but it would be very sad to run
> through the 200Mb file 63 times in order to get 63 planar images.
>
> Otherwise, it looks like an awesome machine. I am guessing that LOCI has
> been working with other multispectral imaging systems and might point me to
> a representative reader that I might use as a guide.
>
> --Lee
>
> _______________________________________________
> ome-devel mailing list
> ome-devel at lists.openmicroscopy.org.uk
> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-devel/attachments/20140826/eee373c1/attachment.html>