[Insight-developers] RE: NoiseImageFilter : renaming suggestion

[Miller, James V (Research)]

Luis,

I struggled with what to call this filter. I avoided calling it a 
StandardDeviationImageFilter because I want to allow extensions
to the filter (activated via modes) to change how the level of the 
noise is calculated. For instance, right now it calculates the
standard deviation relative to the sample mean in a neighborhood
of a pixel.  But the filter could be setup to perform a least squares
fit to the intensities in a neighborhood of the pixel and calculate 
the noise as the standard deviation of the intensities relative to 
a plane or quadric fit to the intensities in a neighborhood.

So something like "NoiseEstimatingImageFilter" might be a better name.

I use these types of techniques for determining how to set thresholds 
for various algorithms and to measure the effectiveness of a smoothing
filter.  For the latter, I'll run the noise estimating filter on the 
input and output of a smoothing filter and then compare the mean values
of the output of the noise estimating filter.


Jim



-----Original Message-----
From: Luis Ibanez [mailto:luis . ibanez at kitware . com]
Sent: Tuesday, June 17, 2003 10:43 AM
To: Miller, James V (Research)
Cc: Insight Developers List
Subject: NoiseImageFilter : renaming suggestion


Jim,

Could we consider renaming this filter ?

          " NoiseImageFilter "

actually sounds like a filter that adds noise
to the input image, rather than estimating
the standard deviation of the input.

Should it be:

     NoiseEstimatorImageFilter ?
or
     StandardDeviationImageFilter ?
       (so it matches with "MeanImageFilter")




   Luis



----------------------------------
Miller, James V (Research) wrote:
> Zein,
> 
> It sounds like NoiseImageFilter is what you want.  
> 
> http://www . itk . org/Insight/Doxygen/html/classitk_1_1NoiseImageFilter . html
> 
> It runs across an image an estimates the standard deviation of the pixel
> intensities in a small neighborhood around each pixel.  
> 
> 
> 
> 
> -----Original Message-----
> From: Luis Ibanez [mailto:luis . ibanez at kitware . com]
> Sent: Monday, June 16, 2003 4:10 PM
> To: salah
> Cc: Insight-users at public . kitware . com
> Subject: [Insight-users] Re: image Homogeniety
> 
> 
> 
> Hi Zein,
> 
> You could easily take the MeanImageFilter filter
> http://www . itk . org/Insight/Doxygen/html/classitk_1_1MeanImageFilter . html
> and modify its loop for computing the standard
> deviation of its neighborhood, instead of computing
> the mean.
> 
> This should require to modify less than ten lines of code.
> (between 131 and 145).
> 
> The filter is in
> Insight/Code/BasicFilters/
> 
>    itkMeanImageFilter.h
>    itkMeanImageFilter.txx
> 
> This filter is using internally the Neighborhood
> iterator.
> 
> You could even use the
> 
>    itkVarianceImageFunction.h
>
http://www . itk . org/Insight/Doxygen/html/classitk_1_1VarianceImageFunction . ht
> ml
> 
> to help you in the computation of the standard
> deviation so you don't even have to write the
> loop for visiting all the neighbors.
> 
> 
> 
> 
> Regards,
> 
> 
>    Luis
> 
> 
> 
> ------------
> 
> salah wrote:
> 
>>Hi Luis,
>>
>>I need a Filter that calculates the the voxel-wise 
>>homogeniety of an image. 
>>
>>1. Does Itk have sucha filter?
>>2. If not I would develop it myself, but I have a question
>>about this. I need a filter that scans all voxels of the
>>image, and at each voxel, do some calculations on the
>>surrounding (kernel region) of this voxel (say for ex. in a 
>>surroundign of 3x3x3 voxel). 
>>what kind of iterator would you suggest me to use for
>>the main iterator, and for the region-iterator?
>>
>>thanks,
>>Zein
>>
>>
>>
> 
>
->8<------------->8<------------->8<------------->8<------------->8<--------
> ----->8<-
> 
>>Zein I. Salah 
>>University of Tubingen, WSI-GRIS, Sand 14, 72076 Tubingen 
>>Email: salah at gris . uni-tuebingen . de
>>Tel.: (07071) 29 75465 (GRIS),           Fax: (07071) 29 54 66
>>
>>
> 
> 
> 
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at www . itk . org
> http://www . itk . org/mailman/listinfo/insight-users
> _______________________________________________
> Insight-users mailing list
> Insight-users at www . itk . org
> http://www . itk . org/mailman/listinfo/insight-users
>