comp.lang.idl-pvwave archive: archive

Home » Public Forums » archive » correlation question

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: correlation question [message #92238 is a reply to message #92237]

Wed, 04 November 2015 09:27

Helder Marchetto
Messages: 520
Registered: November 2011

Senior Member

Hi,
Sorry I didn't explain myself properly.
I'm looking for a IDL way to do this:

Similarity = fltarr(xs,us)
RefCurve = {array of n elements}
data = {array of xs,us,n elements}
for i=0,xs-1 do begin
for j=0,ys-1 do begin
Similarity = "correlation between reform(data[i,j,*]) and RefCurve"
endfor
endfor

With correlation I mean a 1 if the two curves (arrays) are similar (e.g.two lines with slope 1), a zero (or -1) if the lines are perpendicular.

[Sorry, but I'm writing from a mobile device...]

By thresholding (e.g. at gt 0.9) the resulting Similarity image I can "see" where the evolution of the [xs,ys] slices in data resembles that of RefCurve.

Did I make myself clear? If not, I have to come up with a practical example... Sorry if my explanation is so poor.

Cheers, Helder

Report message to a moderator

[Message index]

		correlation question By: Helder Marchetto on Wed, 04 November 2015 01:36
		Re: correlation question By: Jeremy Bailin on Wed, 04 November 2015 08:38
		Re: correlation question By: Helder Marchetto on Wed, 04 November 2015 09:27
		Re: correlation question By: Helder Marchetto on Thu, 05 November 2015 03:27
		Re: correlation question By: Helder Marchetto on Thu, 05 November 2015 04:37
		Re: correlation question By: lecacheux.alain on Thu, 05 November 2015 07:05

Previous Topic:	I am trying to install shana packge, got the following error
Next Topic:	Dates conversions

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Oct 11 10:13:46 PDT 2025

Total time taken to generate the page: 0.62998 seconds