Problem with getting stats from trainable weka segmentation plugin

[Mr Patrick Fortuna]

Hi everyone,


I'm hoping someone can help with my small problem. I've been using the trainable weka segmentation plugin to quantify some images, although havn't used it for a while since the latest update and the macro I've been using doesn't seem to do what I'm after anymore. I train with classifiers and then run this macro to get percentages of the selected classes in the image (text continues after code):


// id values of each class in the classified image

classLabels = newArray( 0, 64, 128, 191, 255 );



setBatchMode( true );

run("Set Measurements...", "area_fraction redirect=None decimal=3");



for( i=0; i<classLabels.length; i ++ )

{

minTh = classLabels[ i ];

maxTh = classLabels[ i ];

 if( i < classLabels.length - 1 )

 maxTh = classLabels[ i+1 ] -1;

else // last value is assumed to be 255

 {

minTh = classLabels[ i-1 ] +1;

 maxTh = classLabels[ i ];

}

 selectWindow("Classified image");

 run("Duplicate...", "title=[to-be-thresholded]");

selectWindow("to-be-thresholded");

 setThreshold( minTh, maxTh );

run("Convert to Mask");

 run("Measure");

close();

}



Although now the problem is the result comes back as 100% for class 1 and 0% for the rest of the classes.

I'm not proficient at all at scripting, so I'd really appreciate if anyone could debug my problem.

Thanks before hand.


Regards,

Patrick

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

[Ignacio Arganda-Carreras]

Hello Patrick,

Since the last version of the Trainable Weka Segmentation, the labels are
set to more meaningful values: 0, 1, 2... So your macro should work if you
change the first line of code to:

classLabels = newArray( 0, 1, 2, 3, 4 );

Let me know if you have more problems.

ignacio


On Thu, May 1, 2014 at 3:10 AM, Mr Patrick Fortuna <
[hidden email]> wrote:

> Hi everyone,
>
>
> I'm hoping someone can help with my small problem. I've been using the
> trainable weka segmentation plugin to quantify some images, although havn't
> used it for a while since the latest update and the macro I've been using
> doesn't seem to do what I'm after anymore. I train with classifiers and
> then run this macro to get percentages of the selected classes in the image
> (text continues after code):
>
>
> // id values of each class in the classified image
>
> classLabels = newArray( 0, 64, 128, 191, 255 );
>
>
>
> setBatchMode( true );
>
> run("Set Measurements...", "area_fraction redirect=None decimal=3");
>
>
>
> for( i=0; i<classLabels.length; i ++ )
>
> {
>
> minTh = classLabels[ i ];
>
> maxTh = classLabels[ i ];
>
>  if( i < classLabels.length - 1 )
>
>  maxTh = classLabels[ i+1 ] -1;
>
> else // last value is assumed to be 255
>
>  {
>
> minTh = classLabels[ i-1 ] +1;
>
>  maxTh = classLabels[ i ];
>
> }
>
>  selectWindow("Classified image");
>
>  run("Duplicate...", "title=[to-be-thresholded]");
>
> selectWindow("to-be-thresholded");
>
>  setThreshold( minTh, maxTh );
>
> run("Convert to Mask");
>
>  run("Measure");
>
> close();
>
> }
>
>
>
> Although now the problem is the result comes back as 100% for class 1 and
> 0% for the rest of the classes.
>
> I'm not proficient at all at scripting, so I'd really appreciate if anyone
> could debug my problem.
>
> Thanks before hand.
>
>
> Regards,
>
> Patrick
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
Ignacio Arganda-Carreras, Ph.D.
Seung's lab, 46-5065
Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology
43 Vassar St.
Cambridge, MA 02139
USA

Phone: [phone]
Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html

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