comp.lang.idl-pvwave archive: archive

Home » Public Forums » archive » curve fiting issue

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

Re: curve fiting issue [message #64266 is a reply to message #64173]

Wed, 10 December 2008 07:21

Wasit.Weather
Messages: 62
Registered: February 2008

Member

On Dec 10, 7:54 am, Jeremy Bailin <astroco...@gmail.com> wrote:
> On Dec 10, 12:12 am, Elkunn <Wasit.Weat...@gmail.com> wrote:
>
>
>
>
>
>> Hello,
>> I have a data array like this. The 1st value a[0] is somewhat
>> contaminated and need to be removed.
>
>> a=[0.0382000, 0.3919000, 0.3843000, 0.3880000, 0.3720000, 0.4221000,
>> 0.5966000, 0.8063000,0.7955000
>> 0.8022000,0.7941000,0.8149000,0.8170000,0.7212000,0.7299000, 0.7644000,0.7738000,0.7574000
>> 0.6756000, 0.6122000,0.5646000,0.5595000,0.5151000]
>
>> I want to remove 1st pixel and replace it by a predicted value from
>> curve fitting and smooth the overll data, return them back into
>> spatial domain. I do not have error values. Is there any simpler
>> method to do that?
>
>> Thanks a lot!
>
>> Elkuun
>
> And do you expect whatever the curve you fit to have compact support
> (probably a good idea for most applications, but I have no clue about
> in your case) or to depend on all of the other data?
>
> -Jeremy.- Hide quoted text -
>
> - Show quoted text -

Thanks for your reply!
This is NDVI data of one pixel over one year. I think Gussian curve
fit works for that. Some pixels has no clouds, then I just need to
smooth the curve, but one for like this, I want to remove the cloud
pixel, then predict its value from least-square fitting, then smooth
the whole curve.

Thank you!

Report message to a moderator

[Message index]

		curve fiting issue By: Wasit.Weather on Tue, 09 December 2008 21:12
		Re: curve fiting issue By: Wasit.Weather on Thu, 11 December 2008 22:32
		Re: curve fiting issue By: Jeremy Bailin on Thu, 11 December 2008 08:02
		Re: curve fiting issue By: Wasit.Weather on Wed, 10 December 2008 07:21
		Re: curve fiting issue By: Jeremy Bailin on Wed, 10 December 2008 05:54
		Re: curve fiting issue By: Jeremy Bailin on Wed, 10 December 2008 05:53

Previous Topic:	Re: Catalyst library: pixmap object use.
Next Topic:	Hypothesis testing

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Oct 11 04:51:12 PDT 2025

Total time taken to generate the page: 1.43859 seconds