Received: by 10.66.77.101 with SMTP id r5mr181555paw.27.1345129467050;
        Thu, 16 Aug 2012 08:04:27 -0700 (PDT)
Path: s8ni2203pbk.0!nntp.google.com!news2.google.com!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail
From: CR <rog...@googlemail.com>
Newsgroups: comp.lang.idl-pvwave
Subject: Re: Extraction phenological events
Date: Thu, 16 Aug 2012 08:04:26 -0700 (PDT)
Organization: http://groups.google.com
Lines: 60
Message-ID: <f20d6b6d-d06a-4b29-8367-7d0d908269d9@googlegroups.com>
References: <ae820005-0eb8-447f-9a09-632dd5766b1c@d6g2000pbt.googlegroups.com>
NNTP-Posting-Host: 139.17.67.210
Mime-Version: 1.0
X-Trace: posting.google.com 1345129466 29977 127.0.0.1 (16 Aug 2012 15:04:26 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Thu, 16 Aug 2012 15:04:26 +0000 (UTC)
In-Reply-To: <ae820005-0eb8-447f-9a09-632dd5766b1c@d6g2000pbt.googlegroups.com>
Complaints-To: groups-abuse@google.com
Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=139.17.67.210;
 posting-account=6VKJPgoAAACOR76ZRp-kNlggBhRFR6aj
User-Agent: G2/1.0
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Am Donnerstag, 7. Juni 2012 19:52:38 UTC+2 schrieb Bandrush Barda:
> Dear all?
>=20
> I=92m doing phenological study using threshold method in IDL. I have 36
>=20
> images with 10 days time interval in 2005
>=20
> (ENVI format). They all have been stacked. I know how to set threshold
>=20
> value on a single image. But I don=92t know how to set threshold value
>=20
> on multiband image to determine the date when grown season start and
>=20
> end. Start of season is determined a day when NDVI value is return to
>=20
> threshold value in a given year. So what I=92d like to ask:
>=20
> 1. How can I set threshold value on multiband image to determine the
>=20
> date when grown season start and end.
>=20
> 2. What kind of procedure and function should I use to extract start
>=20
> and end of season using threshold method on multiband image? The dates
>=20
> when start/end of season occurs are different in every vegetation
>=20
> types. Basically I want to extract one image showing different places
>=20
> are different phenological date. Is there anyone who can help me for
>=20
> this?
>=20
> Thanks a lot

Hi,
as far as I understood you have a stack of NDVI images. If then I would tra=
nsfer them to IDL by the export function in the ENVI file menu. then I woul=
d apply this:

;stack is the name of the varaible you have used for export
sz=3Dsize(stack,/dim)
threshold=3D0.3; you must set this
stack ge=3D threshold
wh=3Dwhere(~stack,c)
if c gt 0 then stack[wh]=3D!values.f_nan
stack*=3Drebin(lindgen(1,1,sz[2]),sz)
startperiod=3Dmin(stack,max=3Dendperiod,/nan,dimension=3D3)


startperiod and endperiod are then two matrices of your spatial extent cont=
aining the frame number per pixel when the period hast started and ended du=
e to the selected threshold. All elements of the matrices having NAN didn't=
 pass the selected threshold.

Cheers

CR