Image stitching advice

[PEARSON Matthew]

Hi all,

Quite often in our facility we capture "large images", stitching  
multiple fields of view together to produce an image of a larger  
area.  This is done on a confocal microscope and the cells we're  
looking at are motile hence the need to tile as they often move out of  
the field of view of the objective lens.  In terms of capturing the  
data there is an option for the capture software to stitch or not.  If  
we don't tell the software to stitch then we're relying on the  
accuracy of the stage to join the tiles.  If we stitch we use  
something like 15% overlap.  The stitching is not always accurate  
though and i understand that cells moving around the field could be  
problematic from a stitching point of view.

So basically i'm wondering if we do not stitch whether we can look to  
other applications to carry out the stitching or at least try them  
out.  Does anyone have any recommendations on ImageJ/FIJI plugins to  
use to stitch tiles from confocal time lapse data?  There seems to be  
quite a few plugins related to this.  I've tried StackReg and this  
does work on the data but it seems quite slow and we're capturing  
quite large data sets at different xy stage positions.

Many thanks,

Matt











--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Stephan Preibisch]

Hi Matt,

for exactly that purpose we developed the Stitching plugin for ImageJ/Fiji (http://fiji.sc/Image_Stitching). It should be able to reconstruct your acquisition. If you encounter any problem just let me know.

The plugin is included in Fiji (http://fiji.sc/) - Fiji is an ImageJ distribution, or available as download for imageJ on my website: http://fly.mpi-cbg.de/~preibisch/software.html#Stitching
I do not maintain the ImageJ plugin very well though.

Please note that the stitching is based on a publication, so if it is useful to you it would be very nice to cite it: http://bioinformatics.oxfordjournals.org/content/25/11/1463.abstract

Have a great day,
Stephan
---

Dr. Stephan Preibisch
HFSP Fellow
Robert H. Singer / Eugene Myers lab

Albert Einstein College of Medicine / HHMI Janelia Farm / MPI-CBG

email: [hidden email] / [hidden email] / [hidden email]
web: http://www.preibisch.net/

On Apr 18, 2014, at 16:16 , Matthew Pearson wrote:

> Hi all,
>
> Quite often in our facility we capture "large images", stitching multiple fields of view together to produce an image of a larger area.  This is done on a confocal microscope and the cells we're looking at are motile hence the need to tile as they often move out of the field of view of the objective lens.  In terms of capturing the data there is an option for the capture software to stitch or not.  If we don't tell the software to stitch then we're relying on the accuracy of the stage to join the tiles.  If we stitch we use something like 15% overlap.  The stitching is not always accurate though and i understand that cells moving around the field could be problematic from a stitching point of view.
>
> So basically i'm wondering if we do not stitch whether we can look to other applications to carry out the stitching or at least try them out.  Does anyone have any recommendations on ImageJ/FIJI plugins to use to stitch tiles from confocal time lapse data?  There seems to be quite a few plugins related to this.  I've tried StackReg and this does work on the data but it seems quite slow and we're capturing quite large data sets at different xy stage positions.
>
> Many thanks,
>
> Matt
>
>
>
>
>
>
>
>
>
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html