Re: how to smooth a data set with proper window [message #81406] |
Mon, 17 September 2012 07:51 |
David Fanning
Messages: 11724 Registered: August 2001
|
Senior Member |
|
|
Kris writes:
> I think we should make a balance between making up missing data and keeping truth.
Yes, but I wonder if the "truth" we find in, say,
a work of fiction, is the same as the "truth" we
try to convey in science. Certainly, I appreciate
and respect both. But, I think I would hold the
scientist to a higher standard.
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
Re: how to smooth a data set with proper window [message #81412 is a reply to message #81407] |
Sun, 16 September 2012 19:09  |
David Fanning
Messages: 11724 Registered: August 2001
|
Senior Member |
|
|
Kris writes:
> When I try to make the contour 'beautiful', I have to smooth the data set.
> But I have two problems as follows:
> 1) the smoothed data with SMOOTH function is different from
> the origin one?
Yes, this is pretty much the point, I think. :-)
> Does it make sense to create some data?
It makes sense to create new data if you only
care about money and not your scientific reputation.
> 2) the data smoothed with MIN_CURVE_SURF function dose not change very much, how to adjust the parameters?
Wait, first you complain about it changing too much.
And now it doesn't change enough? Are you sure you
understand what you are looking for? Have you tried
the Goldilocks function?
Maybe your data isn't smooth. Maybe you just want
to interpolate your data and make it larger in size,
with more pixels with interpolated values between
points. You could do this with the Congrid function
with the INTERP keyword set.
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|