ImageJ

JACoP problem

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

7 messages Options

Michael Cammer

JACoP problem

We are having problems with the JACoP plugin and would appreciate help.

The plugin worked for a while and then began giving the following error:

Duplicate keyword:
Command: "JACoP"
Keyword: "image"
Value: green
Add an underscore to the corresponding label
in the dialog to make the first word unique.

It then continues to run but always prints out the same r= value for
any given pair of images regardless of how we set the threshold.

We're using 1.38u with Java 1.5.0_09.

Thanks.
____________________________________________________________________________
Michael Cammer Analytical Imaging Facility Albert Einstein Coll. of Med.
URL: http://www.aecom.yu.edu/aif/

Michael Schmid

Re: JACoP problem

Hi Michael,

this looks like an error in the plugin.

replace
gd.addChoice("Image A", imgTitleList,imgTitleList[0]);
gd.addChoice("Image B", imgTitleList,imgTitleList[1]);
by
gd.addChoice("Image_A", imgTitleList,imgTitleList[0]);
gd.addChoice("Image_B", imgTitleList,imgTitleList[1]);

ImageJ uses only the first word of each label for recording
macros. With the current code it would record

run("JACoP", "image=[imagenameA] image=[imagenameB] Pearson's=true ...

and it would be unclear which "image" refers to "Image A" and
which to "Image B".

Michael
________________________________________________________________

On 20 Jun 2007, at 16:53, Michael Cammer wrote:

> We are having problems with the JACoP plugin and would appreciate
> help.
>
> The plugin worked for a while and then began giving the following
> error:
>
> Duplicate keyword:
> Command: "JACoP"
> Keyword: "image"
> Value: green
> Add an underscore to the corresponding label
> in the dialog to make the first word unique.
>
> It then continues to run but always prints out the same r= value
> for any given pair of images regardless of how we set the threshold.
>
> We're using 1.38u with Java 1.5.0_09.
>
> Thanks.
> ______________________________________________________________________
> ______
> Michael Cammer Analytical Imaging Facility Albert Einstein
> Coll. of Med.
> URL: http://www.aecom.yu.edu/aif/

Wayne Rasband

Re: JACoP problem

In reply to this post by Michael Cammer

This is a command recorder error. You can avoid it by closing the
Recorder window. The recorder is displaying this error because the
first dialog displayed by the JACoP plugin has two labels that start
with the word "Image". To fix the problem, change the labels to
"Image_A" and "Image_B" and recompile.

-wayne

On Jun 20, 2007, at 10:53 AM, Michael Cammer wrote:

> We are having problems with the JACoP plugin and would appreciate help.
>
> The plugin worked for a while and then began giving the following
> error:
>
> Duplicate keyword:
> Command: "JACoP"
> Keyword: "image"
> Value: green
> Add an underscore to the corresponding label
> in the dialog to make the first word unique.
>
> It then continues to run but always prints out the same r= value for
> any given pair of images regardless of how we set the threshold.
>
> We're using 1.38u with Java 1.5.0_09.
>
> Thanks.
> _______________________________________________________________________
> _____
> Michael Cammer Analytical Imaging Facility Albert Einstein Coll.
> of Med.
> URL: http://www.aecom.yu.edu/aif/

Michael Cammer

Re: JACoP problem

Thanks.

I did this which got rid of the error message, but the software
appears to disregard a manual change in the threshold settings.

-Michael

At 12:05 PM 06/20/07, you wrote:

>This is a command recorder error. You can avoid it by closing the
>Recorder window. The recorder is displaying this error because the
>first dialog displayed by the JACoP plugin has two labels that start
>with the word "Image". To fix the problem, change the labels to
>"Image_A" and "Image_B" and recompile.
>
>-wayne
>
>On Jun 20, 2007, at 10:53 AM, Michael Cammer wrote:
>
>>We are having problems with the JACoP plugin and would appreciate help.
>>
>>The plugin worked for a while and then began giving the following
>>error:
>>
>>Duplicate keyword:
>>Command: "JACoP"
>>Keyword: "image"
>>Value: green
>>Add an underscore to the corresponding label
>>in the dialog to make the first word unique.
>>
>>It then continues to run but always prints out the same r= value for
>>any given pair of images regardless of how we set the threshold.
>>
>>We're using 1.38u with Java 1.5.0_09.
>>
>>Thanks.
>>_______________________________________________________________________ _____
>>Michael Cammer Analytical Imaging Facility Albert Einstein Coll.
>>of Med.
>>URL: http://www.aecom.yu.edu/aif/

____________________________________________________________________________
Michael Cammer Analytical Imaging Facility Albert Einstein Coll. of Med.
URL: http://www.aecom.yu.edu/aif/

Michael Cammer

Re: JACoP problem

I just realized that the Pearson's may not use the threshold. Should
we be subtracting the background to zero before running this to avoid
getting correlation from bg pixels?

Thanks.

At 12:32 PM 06/20/07, you wrote:

>Thanks.
>
>I did this which got rid of the error message, but the software
>appears to disregard a manual change in the threshold settings.
>
>-Michael
>
>At 12:05 PM 06/20/07, you wrote:
>>This is a command recorder error. You can avoid it by closing the
>>Recorder window. The recorder is displaying this error because the
>>first dialog displayed by the JACoP plugin has two labels that start
>>with the word "Image". To fix the problem, change the labels to
>>"Image_A" and "Image_B" and recompile.
>>
>>-wayne
>>
>>On Jun 20, 2007, at 10:53 AM, Michael Cammer wrote:
>>
>>>We are having problems with the JACoP plugin and would appreciate help.
>>>
>>>The plugin worked for a while and then began giving the following
>>>error:
>>>
>>>Duplicate keyword:
>>>Command: "JACoP"
>>>Keyword: "image"
>>>Value: green
>>>Add an underscore to the corresponding label
>>>in the dialog to make the first word unique.
>>>
>>>It then continues to run but always prints out the same r= value for
>>>any given pair of images regardless of how we set the threshold.
>>>
>>>We're using 1.38u with Java 1.5.0_09.
>>>
>>>Thanks.
>>>_______________________________________________________________________
>>>_____
>>>Michael Cammer Analytical Imaging Facility Albert Einstein Coll.
>>>of Med.
>>>URL: http://www.aecom.yu.edu/aif/
>
>____________________________________________________________________________
>Michael Cammer Analytical Imaging Facility Albert Einstein Coll. of Med.
>URL: http://www.aecom.yu.edu/aif/

____________________________________________________________________________
Michael Cammer Analytical Imaging Facility Albert Einstein Coll. of Med.
URL: http://www.aecom.yu.edu/aif/

Fabrice Cordelières

Re: JACoP problem

In reply to this post by Michael Cammer

Dear Michael,

Thanks for reporting this not ergonomic conception of the plugin ;-) and
thanks to Wayne for answering on the workaround. I'll keep this in mind as
I'll be soon working on a new improved version of JACoP.
About the unability to change manually the threshold, I found this bug to
occur on version 1.38t for instance, but not an older version (1.38l) of
IJ... I didn't determined the precise version where the function was lost
but I guess this might come from some kind of problems between the way JACoP
was programmed and newer versions of IJ. However, I noticed that with
another plugin (3D object counter), the threshold slider and numerical field
were behaving just fine with both version of IJ !? Anyway, I'll check my
code and will let you know about that off-list.

Cheers,

Fabrice

Fabrice Cordelières, PhD

Institut Curie - Section de recherche/ CNRS UMR 146
Plateforme d'Imagerie Cellulaire et Tissulaire
Bâtiment 112 - Centre universitaire
91405 Orsay Cedex
FRANCE

Tél. : +33 1 69 86 31 30
Fax. : +33 1 69 86 17 03

-----Message d'origine-----
De : ImageJ Interest Group [mailto:[hidden email]] De la part de
Michael Cammer
Envoyé : mercredi 20 juin 2007 18:32
À : [hidden email]
Objet : Re: JACoP problem

Thanks.

I did this which got rid of the error message, but the software
appears to disregard a manual change in the threshold settings.

-Michael

At 12:05 PM 06/20/07, you wrote:

_____
>>Michael Cammer Analytical Imaging Facility Albert Einstein Coll.
>>of Med.
>>URL: http://www.aecom.yu.edu/aif/

____________________________________________________________________________
Michael Cammer Analytical Imaging Facility Albert Einstein Coll. of Med.
URL: http://www.aecom.yu.edu/aif/

Fabrice Cordelières

Re: JACoP problem

In reply to this post by Michael Cammer

Hi Michael,
In the original calculation of Pearson's coefficient, a threshold value is
not supposed to be set. This is surely due to the fact that its first use
was on confocal images where the offset value is supposed to have been well
set to get rid of background pixels. As JACoP applies the original formula,
it doesn't take into account any threshold value. However, as you suggested,
you may subtract background on both images before launching JACoP.

Cheers,

Fab

Fabrice Cordelières, PhD

Institut Curie - Section de recherche/ CNRS UMR 146
Plateforme d'Imagerie Cellulaire et Tissulaire
Bâtiment 112 - Centre universitaire
91405 Orsay Cedex
FRANCE

Tél. : +33 1 69 86 31 30
Fax. : +33 1 69 86 17 03

-----Message d'origine-----
De : ImageJ Interest Group [mailto:[hidden email]] De la part de
Michael Cammer
Envoyé : mercredi 20 juin 2007 19:02
À : [hidden email]
Objet : Re: JACoP problem

I just realized that the Pearson's may not use the threshold. Should
we be subtracting the background to zero before running this to avoid
getting correlation from bg pixels?

Thanks.

At 12:32 PM 06/20/07, you wrote:

_
>Michael Cammer Analytical Imaging Facility Albert Einstein Coll. of
Med.
>URL: http://www.aecom.yu.edu/aif/

____________________________________________________________________________
Michael Cammer Analytical Imaging Facility Albert Einstein Coll. of Med.
URL: http://www.aecom.yu.edu/aif/