the algorithm of finding particle positions

[Stone Von]

Can anyone tell me what method or algorithm is used in measuring particle position?  I am using Analyze > Measure after choosing a threshold, and in "Set Measurements" I am not selecting Fit Ellipse.

I am guessing that the method has two steps:
1. choose contiguous pixels brighter than the threshold
2. second finding the center of these selected pixels -- this could be done several ways and I list three possibilities: (a) fitting the chosen pixels to a Gaussian spatial distribution of intensity, (b) calculating the first moment of the spatial distribution of intensity, or (c) simply just finding the point halfway between the outermost pixels that were selected - maybe that's how it was done in NIH Image.

Can anybody tell me which of these methods is used in ImageJ, or is it some other method?

Thank you very much!~~

[Wayne Rasband]

The Centroid (X and Y) and Center of Mass (XM and YM) parameters are  
described in the online manual at:

      http://rsb.info.nih.gov/ij/docs/menus/analyze.html#set

Look at the getCentroid() and calculateMoments() methods in the  
FloatStatistics class to see how these values are calculated. The  
source code is online at:

     
http://rsb.info.nih.gov/ij/developer/source/ij/process/ 
FloatStatistics.java.html

-wayne

On Sep 5, 2006, at 6:50 PM, Stone Von wrote:

> Can anyone tell me what method or algorithm is used in measuring
> particle position?  I am using Analyze > Measure after choosing a
> threshold, and in "Set Measurements" I am not selecting Fit Ellipse.
>
> I am guessing that the method has two steps:
> 1. choose contiguous pixels brighter than the threshold
> 2. second finding the center of these selected pixels -- this could be  
> done
> several ways and I list three possibilities: (a) fitting the chosen  
> pixels
> to a Gaussian spatial distribution of intensity, (b) calculating the  
> first
> moment of the spatial distribution of intensity, or (c) simply just  
> finding
> the point halfway between the outermost pixels that were selected -  
> maybe
> that's how it was done in NIH Image.
>
> Can anybody tell me which of these methods is used in ImageJ,
> or is it some other method?
>
> Thank you very much!~~
> --
> View this message in context:  
> http://www.nabble.com/the-algorithm-of-finding-particle-positions- 
> tf2223878.html#a6161922
> Sent from the ImageJ forum at Nabble.com.
>