[ImageJ-devel] [fiji-devel] [Bug 258] Opening multiposition LSM files by Drag&Drop throws exception (fwd)

Curtis Rueden ctrueden at wisc.edu
Thu May 19 13:40:41 CDT 2011 

Hi Dan,

Sorry for the long delay in my reply to your questions.

LSM files can be 1D line scans over multiple data channels (lambda or
> spectral scan) over time.
> We have no way of easily displaying that in imageJ right now without
> jumping through many hoops...
> but we should aim to make a good general and trivial (for the user)
>  solution for the 2D and 3D display of n dimensional pixels?
>
> Whats the plan in imageJ2 for this Curtis?
> Now the 1st alpha is out... its certainly the right time to get this stuff
> nailed down right?
> Is there a design for the new "hyperstack" pixel data viewer?
>

The current status of the multidimensional viewer is that it displays one
plane at a time, just like IJ1 does, with optional color compositing across
the Channel axis (if any). So RGB images display as expected with red, green
and blue channels simultaneously. The compositing options are more flexible
than in IJ1, in that there is no 7-channel limit, and you can have a
separate color table per plane, not just per channel.

I imagine a display configuration window gadget... something like the image
> disply properties in imaris, where you can choose which dimension to show as
> x and whoch as y on the 2D display, and which extra dimensions with sliders,
> and which as different colour channels (like composite, still with a
> slider?). LUTs could also be handled in there....?
>

Unfortunately, we are currently limited to the first two dataset dimensions
as XY. I agree completely that eventually, we want to allow the user to
select which two dimensions are displayed as XY in space, but there are
various implementation challenges to that. I am not sure whether we can have
something like that in place for the 2.0-final release, but if not it will
be a goal for the next major release after that.

This is all tied into the ongoing discussion about and OME/imglib work on
> n-dimensional ROIs I guess...
>

True, and N-dimensional ROIs get really complicated very quickly. For ImageJ
2.0-final we are shooting for 2D ROIs only tied to specific image planes,
like ImageJ1 has, but with an extensible architecture which will allow 3D+
ROIs later. We are making good progress on that now, and should have
something to show for the 2.0-alpha3 release at the end of the month.

Right now there is the imglib1/2 stuff to sort out, so better not get too
> distracted from that...
> but anyway, these are ideas for the future.
>

Fortunately, as of 2.0-alpha2, we are completely switched over to ImgLib2.
Development continues!

-Curtis

On Tue, Apr 5, 2011 at 4:32 AM, Daniel James White <white at mpi-cbg.de> wrote:

> Hi Jason and all,
>
> On Apr 5, 2011, at 9:58 AM, Jason Swedlow wrote:
>
> > Hi Dan-
> >
> > <large scale tooting of own horn>
>
> you should and you must!!!
>
> >
> > In fact, this type of data collection is not strange at all-- we, and
> many others do it alot.  It's really very helpful for gathering many
> examples in timelapse, and also for tiling.  Another example is the concept
> of a WellSample in HCS-- multiple images in different locations of the same
> Well.
>
> Exactly...
> I just wonder how we are going to get imglib and these higher dimensions to
> play nice?
>
> >
> > In OME this is handled explicitly (
> http://www.openmicroscopy.org/Schemas/OME/2010-06/ome.xsd).  Look under
> the Plane Element for details.
>
> Good (phew!)
>
> >
> > I understand this approach is somewhat at odds with imglib. As we move
> forward towards a similar concept (
> http://trac.openmicroscopy.org.uk/ome/ticket/3678), we'll definitely keep
> an eye on what the Fiji/Imagejdev team is doing there.
> >
> > Obviously, implementing this in software is technically not challenging,
> but it's not always clear how user wants data displayed.  View 10 20 GB
> images at once?  Toggle them?  Montage? Tile?
>
> I guess the way familiar to most imageJ users will be to have a single
> "hyperstack"  window, which only ever shows one image at a time (or maybe 2
> images in magenta / green, or 3 images as  RGB or CYM) with a slider for
> every dimension other than x and y.
>
> I suppose there is no need to say that x and y must be the 2 dimensions
> plotted to the screen.... once could choose which 2 to plot,
> and relegate the others to having sliders.
>
> In the 3D viewer.... same story, choose which 3 dimensions to render as a
> volume, and have sliders for the others?
>
> As you suggest, once can also imagine a "'google maps" type viewer for
> large tiled image data sets, where you can see and over view of the whole
> structure when zoomed out, and also zoom in to see the details of one
> area... and of course Stephan Saalfeld already did exactly that right here:
> http://fly.mpi-cbg.de/
> so we have a good idea how to approach that i guess?
> I think I remember that this required pre computation of the over view,
> zoomed out images at various scales.. and this is exactly why imaris takes
> forever to open large images... its calculating the down sampled
> representations.... so once its loaded the zooming in and out is fast...
>
> LSM files can be 1D line scans over multiple data channels (lambda or
> spectral scan) over time.
> We have no way of easily displaying that in imageJ right now without
> jumping through many hoops...
> but we should aim to make a good general and trivial (for the user)
>  solution for the 2D and 3D display of n dimensional pixels?
>
> Whats the plan in imageJ2 for this Curtis?
> Now the 1st alpha is out... its certainly the right time to get this stuff
> nailed down right?
> Is there a design for the new "hyperstack" pixel data viewer?
>
> I imagine a display configuration window gadget... something like the image
> disply properties in imaris, where you can choose which dimension to show as
> x and whoch as y on the 2D display, and which extra dimensions with sliders,
> and which as different colour channels (like composite, still with a
> slider?). LUTs could also be handled in there....?
>
> This is all tied into the ongoing discussion about and OME/imglib work on
> n-dimensional ROIs I guess...
>
> Right now there is the imglib1/2 stuff to sort out, so better not get too
> distracted from that...
> but anyway, these are ideas for the future.
>
> cheers
>
> Dan
>
>
>
>
>
> >
> > </large scale tooting of own horn>
> >
> > Cheers,
> >
> > jason
> >
> > On 5 Apr 2011, at 08:45, Daniel James White wrote:
> >
> >> Hi all .lsm sufferers,
> >>
> >>
> >> On Apr 4, 2011, at 11:46 PM, Johannes Schindelin wrote:
> >>
> >>> Hi Mark & other .lsm experts,
> >>>
> >>>> ---------- Forwarded message ----------
> >>>> [...]
> >>>> http://pacific.mpi-cbg.de/cgi-bin/bugzilla/show_bug.cgi?id=258
> >>>>
> >>>> [...]
> >>>> ------- Comment #9 from johannes.schindelin at gmx.de  2011-04-04 23:31
> -------
> >>>> I pushed the changes to the 'lsm-reader-4.0g' branch, and hope to
> merge
> >>>> to 'master' and to upload to the Updater soon.
> >>>
> >>> May I ask you for a quick sanity check? It's actually very small:
> >>>
> >>>
> http://pacific.mpi-cbg.de/cgi-bin/gitweb.cgi?p=fiji.git;a=commitdiff;h=7e70bc5762abf993b729910b4c3b0414e5a97ae6
> >>
> >> Hmmm, looking at the diffs... it seems that tile position is stuffed
> into the time dimension.
> >>
> >> So whats gonna happen if there is a time series tile scan?
> >> "There may be trouble ahead, but while there's moonlight, and music, and
> love and romance..."
> >>
> >> Do we have a strategy in imageJ2 for dealing with multi stage xy
> position data sets
> >> that also allows time series?
> >>
> >> Imglib is the answer guess? But hoiw?
> >> Does imglib have a concept for this rather strange pair of extra
> dimensions: x and y stage position?
> >> Some kind of x and y offset (in theory also z could be "tiled" eg. when
> a z piezo stepper has a limited range, and you do several scans in z to
> cover a larger range)
> >>
> >> ???
> >>
> >> Dan
> >>
> >>
> >>>
> >>> Ciao,
> >>> Dscho
> >>>
> >>> --
> >>> You received this message because you are subscribed to the Google
> Groups "Fiji-devel" group.
> >>> To post to this group, send email to fiji-devel at googlegroups.com.
> >>> To unsubscribe from this group, send email to
> fiji-devel+unsubscribe at googlegroups.com.
> >>> For more options, visit this group at
> http://groups.google.com/group/fiji-devel?hl=en.
> >>>
> >>
> >> Dr. Daniel James White BSc. (Hons.) PhD
> >> Senior Microscopist / Image Visualisation, Processing and Analysis
> >> Light Microscopy and Image Processing Facilities
> >> Max Planck Institute of Molecular Cell Biology and Genetics
> >> Pfotenhauerstrasse 108
> >> 01307 DRESDEN
> >> Germany
> >>
> >> +49 (0)15114966933 (German Mobile)
> >> +49 (0)351 210 2627 (Work phone at MPI-CBG)
> >> +49 (0)351 210 1078 (Fax MPI-CBG LMF)
> >>
> >> http://www.bioimagexd.net    BioImageXD
> >> http://pacific.mpi-cbg.de            Fiji -  is just ImageJ (Batteries
> Included)
> >> http://www.chalkie.org.uk            Dan's Homepages
> >> https://ifn.mpi-cbg.de                       Dresden Imaging Facility
> Network
> >> dan (at) chalkie.org.uk
> >> ( white (at) mpi-cbg.de )
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >>
> >> _______________________________________________
> >> ImageJ-devel mailing list
> >> ImageJ-devel at imagejdev.org
> >> http://imagejdev.org/mailman/listinfo/imagej-devel
> >
> >
> >
> > **************************
> > Wellcome Trust Centre for Gene Regulation & Expression
> > College of Life Sciences
> > MSI/WTB/JBC Complex
> > University of Dundee
> > Dow Street
> > Dundee  DD1 5EH
> > United Kingdom
> >
> > phone (01382) 385819
> > Intl phone:  44 1382 385819
> > FAX   (01382) 388072
> > email: jason at lifesci.dundee.ac.uk
> >
> > Lab Page: http://gre.lifesci.dundee.ac.uk/staff/jason_swedlow.html
> > Open Microscopy Environment: http://openmicroscopy.org
> > **************************
> >
> > The University of Dundee is a Scottish Registered Charity, No. SC015096.
> >
> >
> >
>
> Dr. Daniel James White BSc. (Hons.) PhD
> Senior Microscopist / Image Visualisation, Processing and Analysis
> Light Microscopy and Image Processing Facilities
> Max Planck Institute of Molecular Cell Biology and Genetics
> Pfotenhauerstrasse 108
> 01307 DRESDEN
> Germany
>
> +49 (0)15114966933 (German Mobile)
> +49 (0)351 210 2627 (Work phone at MPI-CBG)
> +49 (0)351 210 1078 (Fax MPI-CBG LMF)
>
> http://www.bioimagexd.net       BioImageXD
> http://pacific.mpi-cbg.de               Fiji -  is just ImageJ (Batteries
> Included)
> http://www.chalkie.org.uk               Dan's Homepages
> https://ifn.mpi-cbg.de                  Dresden Imaging Facility Network
> dan (at) chalkie.org.uk
> ( white (at) mpi-cbg.de )
>
>
>
>
>
>
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20110519/8ea82c20/attachment.html>

More information about the ImageJ-devel mailing list