comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » particle detection - a way to speed up things?
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Return to the default flat view Create a new topic Submit Reply
Re: particle detection - a way to speed up things? [message #57114 is a reply to message #57067] Mon, 03 December 2007 08:30 Go to previous messageGo to previous message
Dan Larson is currently offline  Dan Larson
Messages: 21
Registered: March 2002
Junior Member
Hi,

Depending on the types of blobs you are detecting, FFT filtering can
often work much faster for the initial segmentation. My approach for
particle detection:
1. spatial bandpass to identify the regions of interest.
2. cut out the regions of interest to pass to the fitting routine (as
David suggested).
3. find the center using the algorithm of choice.

This last step can be any number of things, but if you choose a
centroid approach, these lines work well for me, although I think they
are similar to what you have:

a=size(pic)
x_dim = a[1]
y_dim = a[2]
array=lindgen(x_dim, y_dim)
xarr=array mod x_dim
yarr=array/x_dim
sum=double(total(pic))
xcenter=total(xarr*pic)/sum
ycenter=total(yarr*pic)/sum
return, [xcenter, ycenter, sum]
end

For fourier filtering, I particulary like a routine from Crocker and
Grier: www.physics.emory.edu/~weeks/idl/kit/bpass.pro

If you are interested in detecting and fitting diffraction-limited
spots in a fluorescence microscope, I have software which is
specifically designed for that purpose. I could send that to you
directly if you want to try it out.

best,
dan
[Message index]
 
Read Message
Read Message
Read Message
Read Message
Previous Topic: reshaping figure
Next Topic: How to modify my IDL library path?

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Fri Oct 10 12:52:28 PDT 2025

Total time taken to generate the page: 1.04112 seconds