Grid Stitching - no output image

[Rossi, Sharyn (NIH/NIA/IRP) [C]]

Hello,

I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?

Thanks,

Sharyn

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

Hi, could you please send the messages from the log window and possible exceptions?

Thanks a lot,
Stephan
---

Dr. Stephan Preibisch
Group Leader

Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
Building 89, 1.08b

email: [hidden email]<mailto:[hidden email]>
web: http://preibischlab.mdc-berlin.de

On 27 Mar 2017, at 16:43, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]>> wrote:

Hello,

I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?

Thanks,

Sharyn

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


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

[Rossi, Sharyn (NIH/NIA/IRP) [C]]

Hello Stephan,

Please see attached. I also wrote to your email.  

Thanks for your help,

Sharyn

On 3/27/17, 11:12 AM, "[hidden email]" <[hidden email]> wrote:

    Hi, could you please send the messages from the log window and possible exceptions?
   
    Thanks a lot,
    Stephan
    ---
   
    Dr. Stephan Preibisch
    Group Leader
   
    Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
    Building 89, 1.08b
   
    email: [hidden email]<mailto:[hidden email]>
    web: http://preibischlab.mdc-berlin.de
   
    On 27 Mar 2017, at 16:43, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]>> wrote:
   
    Hello,
   
    I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?
   
    Thanks,
   
    Sharyn
   
    --
    ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   
   
    --
    ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   


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

[Rossi, Sharyn (NIH/NIA/IRP) [C]]

Hi,

Figured I would try this once again to see if anyone else has encountered a problem like this. I am attaching the log and console info.

Thanks,

Sharyn

On 3/27/17, 10:43 AM, "Rossi, Sharyn (NIH/NIA/IRP) [C]" <[hidden email]> wrote:

    Hello,
   
    I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?
   
    Thanks,
   
    Sharyn
   
    --
    ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   


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

Hi, your images are too big for the current version (>2^31 pixels) … which parameters do you use to run the Stitching? It seems to me you might not specify an approximate overlap between tiles?

Cheers,
Stephan
---

Dr. Stephan Preibisch
Group Leader

Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
Building 89, 1.08b

email: [hidden email]<mailto:[hidden email]>
web: http://preibischlab.mdc-berlin.de

On 5 Apr 2017, at 21:26, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]>> wrote:

Hi,

Figured I would try this once again to see if anyone else has encountered a problem like this. I am attaching the log and console info.

Thanks,

Sharyn

On 3/27/17, 10:43 AM, "Rossi, Sharyn (NIH/NIA/IRP) [C]" <[hidden email]<mailto:[hidden email]>> wrote:

   Hello,

   I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?

   Thanks,

   Sharyn

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



--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html
<console.txt><Log.txt><Exception.txt>


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

[Rossi, Sharyn (NIH/NIA/IRP) [C]]

Hi Stephan,

Thanks for your reply. Yes – there are many images that were obtained using a light-sheet ultramicroscope so there is that in addition to the size of each individual image. I did select down sampling and cut the pixel size in half and I did specify the 20% overlap (although I also tried it without and obtained the same result). Do you think I should down-sample even more? I have 2 filters in the dataset so if you think it would help, I could separate the filters – stitch one color at a time and then overlay.

Thanks for your help,

Sharyn

On 4/6/17, 1:28 AM, "[hidden email]" <[hidden email]> wrote:

    Hi, your images are too big for the current version (?>2^31 pixels) … which parameters do you use to run the Stitching? It seems to me you might not specify an approximate overlap between tiles?
   
    Cheers,
    Stephan
    ---
   
    Dr. Stephan Preibisch
    Group Leader
   
    Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
    Building 89, 1.08b
   
    email: [hidden email]<mailto:[hidden email]>
    web: http://preibischlab.mdc-berlin.de
   
    On 5 Apr 2017, at 21:26, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]>> wrote:
   
    Hi,
   
    Figured I would try this once again to see if anyone else has encountered a problem like this. I am attaching the log and console info.
   
    Thanks,
   
    Sharyn
   
    On 3/27/17, 10:43 AM, "Rossi, Sharyn (NIH/NIA/IRP) [C]" <[hidden email]<mailto:[hidden email]>> wrote:
   
       Hello,
   
       I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?
   
       Thanks,
   
       Sharyn
   
       --
       ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   
   
   
    --
    ImageJ mailing list: http://imagej.nih.gov/ij/list.html
    <console.txt><Log.txt><Exception.txt>
   
   
    --
    ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   


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

Hi, no, the channels will not help. Downsampling should, but the final fused image will most likely not be displayable either at full resolution. We are actually implementing a new version of the Stitching (the BigStitcher) that will be released in May. This will address exactly those problems. How urgent is your reconstruction, we could give you a preliminary version that already works to some extent (still requires to replace some jars manually in Fiji).

All the best,
Stephan
---

Dr. Stephan Preibisch
Group Leader

Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
Building 89, 1.08b

email: [hidden email]<mailto:[hidden email]>
web: http://preibischlab.mdc-berlin.de

On 6 Apr 2017, at 16:55, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]>> wrote:

Hi Stephan,

Thanks for your reply. Yes – there are many images that were obtained using a light-sheet ultramicroscope so there is that in addition to the size of each individual image. I did select down sampling and cut the pixel size in half and I did specify the 20% overlap (although I also tried it without and obtained the same result). Do you think I should down-sample even more? I have 2 filters in the dataset so if you think it would help, I could separate the filters – stitch one color at a time and then overlay.

Thanks for your help,

Sharyn

On 4/6/17, 1:28 AM, "[hidden email]<mailto:[hidden email]>" <[hidden email]<mailto:[hidden email]>> wrote:

   Hi, your images are too big for the current version (?>2^31 pixels) … which parameters do you use to run the Stitching? It seems to me you might not specify an approximate overlap between tiles?

   Cheers,
   Stephan
   ---

   Dr. Stephan Preibisch
   Group Leader

   Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
   Building 89, 1.08b

   email: [hidden email]<mailto:[hidden email]><mailto:[hidden email]>
   web: http://preibischlab.mdc-berlin.de

   On 5 Apr 2017, at 21:26, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]><mailto:[hidden email]>> wrote:

   Hi,

   Figured I would try this once again to see if anyone else has encountered a problem like this. I am attaching the log and console info.

   Thanks,

   Sharyn

   On 3/27/17, 10:43 AM, "Rossi, Sharyn (NIH/NIA/IRP) [C]" <[hidden email]<mailto:[hidden email]><mailto:[hidden email]>> wrote:

      Hello,

      I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?

      Thanks,

      Sharyn

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



   --
   ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   <console.txt><Log.txt><Exception.txt>


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



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


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

[Rossi, Sharyn (NIH/NIA/IRP) [C]]

Hi Stephan,

Thanks for your reply. It can wait until May – I will keep an eye out for BigStitcher.

Cheers,

Sharyn

On 4/7/17, 12:41 AM, "[hidden email]" <[hidden email]> wrote:

    Hi, no, the channels will not help. Downsampling should, but the final fused image will most likely not be displayable either at full resolution. We are actually implementing a new version of the Stitching (the BigStitcher) that will be released in May. This will address exactly those problems. How urgent is your reconstruction, we could give you a preliminary version that already works to some extent (still requires to replace some jars manually in Fiji).
   
    All the best,
    Stephan
    ---
   
    Dr. Stephan Preibisch
    Group Leader
   
    Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
    Building 89, 1.08b
   
    email: [hidden email]<mailto:[hidden email]>
    web: http://preibischlab.mdc-berlin.de
   
    On 6 Apr 2017, at 16:55, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]>> wrote:
   
    Hi Stephan,
   
    Thanks for your reply. Yes – there are many images that were obtained using a light-sheet ultramicroscope so there is that in addition to the size of each individual image. I did select down sampling and cut the pixel size in half and I did specify the 20% overlap (although I also tried it without and obtained the same result). Do you think I should down-sample even more? I have 2 filters in the dataset so if you think it would help, I could separate the filters – stitch one color at a time and then overlay.
   
    Thanks for your help,
   
    Sharyn
   
    On 4/6/17, 1:28 AM, "[hidden email]<mailto:[hidden email]>" <[hidden email]<mailto:[hidden email]>> wrote:
   
       Hi, your images are too big for the current version (?>2^31 pixels) … which parameters do you use to run the Stitching? It seems to me you might not specify an approximate overlap between tiles?
   
       Cheers,
       Stephan
       ---
   
       Dr. Stephan Preibisch
       Group Leader
   
       Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
       Building 89, 1.08b
   
       email: [hidden email]<mailto:[hidden email]><mailto:[hidden email]>
       web: http://preibischlab.mdc-berlin.de
   
       On 5 Apr 2017, at 21:26, Rossi, Sharyn (NIH/NIA/IRP) [C] <[hidden email]<mailto:[hidden email]><mailto:[hidden email]>> wrote:
   
       Hi,
   
       Figured I would try this once again to see if anyone else has encountered a problem like this. I am attaching the log and console info.
   
       Thanks,
   
       Sharyn
   
       On 3/27/17, 10:43 AM, "Rossi, Sharyn (NIH/NIA/IRP) [C]" <[hidden email]<mailto:[hidden email]><mailto:[hidden email]>> wrote:
   
          Hello,
   
          I have run a large dataset obtained from an ultra-microscope through the grid-stitch plugin and although it generates an output.txt file and processes each of the images in the mosaic (as seen via console), I get no output image. I have tried writing to disk as well as the live stitch option to no avail. Any comments or suggestions as to what may be the problem?
   
          Thanks,
   
          Sharyn
   
          --
          ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   
   
   
       --
       ImageJ mailing list: http://imagej.nih.gov/ij/list.html
       <console.txt><Log.txt><Exception.txt>
   
   
       --
       ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   
   
   
    --
    ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   
   
    --
    ImageJ mailing list: http://imagej.nih.gov/ij/list.html
   


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

[MChapman]

Hi Stephan,

Are there any other changes that you are making to the stitching program for "BigStitcher" besides the ability to stitch larger mosaics?

aka - Are you looking into fixing the other issues that were present with the grid collection stitching program?

I have been trying to get a hold of you with the data set that we had talked about around Christmas on the mailing list.  I sent a few emails to the one you have listed on your site and a PM through the ImageJ forum but never got any response.  The mailing list has been not letting me send out anything for a few months now.

Here is the link to the dataset if you are still interested:
https://drive.google.com/open?id=0B9llW4iht0nZSW1ZUXl5eWdZZDQ

I really like your stitching program (and use it everyday), and I would like to help to improve it any way that I can.  

Thanks!
Mike Chapman

Hi Mike,

great to hear that you like the Stitching. Apart from supporting larger, bigger and more images, it will offer:

- much better memory efficiency
- faster
- manual interventions for determining the right overlap between tiles
- as correct as possible placement of “empty” tiles and “islands of tiles with content” relative to each other
- multi-view alignment (rotations in 3d)
- interactive 3d browsing using the BigDataViewer
- instantenous fusion (into ImagePlus) of the final dataset at various resolution levels and fields of view as defined by the user

And maybe a few more gimmicks. We are confident to release a stable beta end of May!

For now it will live in parallel to the current Stitching, but sooner or later things are supposed to converge.

Thanks for you support and all the best,
Stephan
---

Dr. Stephan Preibisch
Group Leader

Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
Building 89, 1.08b

email: [hidden email]<mailto:[hidden email]>
web: http://preibischlab.mdc-berlin.de

On 13. Apr 2017, at 16:52, MChapman <[hidden email]<mailto:[hidden email]>> wrote:

Hi Stephan,

Are there any other changes that you are making to the stitching program for
"BigStitcher" besides the ability to stitch larger mosaics?

aka - Are you looking into fixing the other issues that were present with
the grid collection stitching program?

I have been trying to get a hold of you with the data set that we had talked
about around Christmas on the mailing list.  I sent a few emails to the one
you have listed on your site and a PM through the ImageJ forum but never got
any response.  The mailing list has been not letting me send out anything
for a few months now.

Here is the link to the dataset if you are still interested:
https://drive.google.com/open?id=0B9llW4iht0nZSW1ZUXl5eWdZZDQ

I really like your stitching program (and use it everyday), and I would like
to help to improve it any way that I can.

Thanks!
Mike Chapman



--
View this message in context: http://imagej.1557.x6.nabble.com/Grid-Stitching-no-output-image-tp5018380p5018536.html
Sent from the ImageJ mailing list archive at Nabble.com.

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


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

Hi Mike,

in this email I wanted to address your problem. It does not find overlapping tiles since the determined correlation between both:


10000.tif[1] <- 10001.tif[1]: (1037.8536, -1.7621746) correlation (R)=0.29953933 (144 ms)

is below the correlation threshold of 0.3 that you set in the parameters, thus it trashes this connection leaving you with only the first input tile.

So to keep it just try lowering the threshold - but I assume it might come up with a wrong solution.

Hope this helps anyways a bit?

All the best,
Stephan
---

Dr. Stephan Preibisch
Group Leader

Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
Building 89, 1.08b

email: [hidden email]<mailto:[hidden email]>
web: http://preibischlab.mdc-berlin.de

On 13. Apr 2017, at 16:52, MChapman <[hidden email]<mailto:[hidden email]>> wrote:

Hi Stephan,

Are there any other changes that you are making to the stitching program for
"BigStitcher" besides the ability to stitch larger mosaics?

aka - Are you looking into fixing the other issues that were present with
the grid collection stitching program?

I have been trying to get a hold of you with the data set that we had talked
about around Christmas on the mailing list.  I sent a few emails to the one
you have listed on your site and a PM through the ImageJ forum but never got
any response.  The mailing list has been not letting me send out anything
for a few months now.

Here is the link to the dataset if you are still interested:
https://drive.google.com/open?id=0B9llW4iht0nZSW1ZUXl5eWdZZDQ

I really like your stitching program (and use it everyday), and I would like
to help to improve it any way that I can.

Thanks!
Mike Chapman



--
View this message in context: http://imagej.1557.x6.nabble.com/Grid-Stitching-no-output-image-tp5018380p5018536.html
Sent from the ImageJ mailing list archive at Nabble.com.

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


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

[MChapman]

Stephan,

Thanks for the responses, and I look forward to trying out BigStitcher!

I recognize that the reason that stitch failed was because there was a low correlation between the two images, and I have found ways fix those issues.

The reason that I sent this dataset to you though is that I think there is something peculiar going on when grid collection/stitching is initializing the correlations which is causing the first tiles to drop out in the larger stitches.  

I literally have tens of thousands of examples of this error, and there error will go away by eliminating a portion of the images in the stitch (cutting out everything but the first few rows) and running the same stitching parameters.  

It doesn't make sense to me why it would behave differently in the initial correlation measurements with the exact same first two images and stitching parameters.

Mike

[MChapman]

Stephan,

One other question - Are you planning to post the source code for the Grid Collection/Stitching or BigStitcher (or is it already posted somewhere)?  I know most of ImageJ is open source but I haven't been able to find the code for your stitching algorithms.

Thanks again!
Mike

[MChapman]

Stephan,

Any idea when Big Stitcher will be released?

Thanks again!
Mike

[Stephan Collins]

Sorry, wrong stephan here... You are probably after Stephan Preibisch?


2017-06-12 16:23 GMT+02:00 MChapman <[hidden email]>:

> Stephan,
>
> Any idea when Big Stitcher will be released?
>
> Thanks again!
> Mike
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.
> com/Grid-Stitching-no-output-image-tp5018380p5018878.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
Dr Stephan Collins
MCF - lecturer

IGBMC, CBI
room 3014
1 Rue Laurent Fries, 67400 Illkirch-Graffenstaden
Tel: [phone]

Université de Bourgogne
CSGA
9E Boulevard Jeanne d'Arc 21000 Dijon

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

Hi, it is now released as alpha. We will make a bigger announcement once we put some documentation up. If you want to try it already, activate the Big Stitcher update site in Fiji (Help > Update … > Manage Update Sites > Click BigStitcher ), after a Fiji restart it is then available under Plugins > BigStitcher > BigStitcher.

All the best,
Stephan
---

Dr. Stephan Preibisch
Group Leader

Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center for Molecular Medicine (MDC)
Building 89, 1.08b

email: [hidden email]<mailto:[hidden email]>
web: http://preibischlab.mdc-berlin.de

On 12. Jun 2017, at 22:23, MChapman <[hidden email]<mailto:[hidden email]>> wrote:

Stephan,

Any idea when Big Stitcher will be released?

Thanks again!
Mike



--
View this message in context: http://imagej.1557.x6.nabble.com/Grid-Stitching-no-output-image-tp5018380p5018878.html
Sent from the ImageJ mailing list archive at Nabble.com<http://Nabble.com>.

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


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