No subject

[Net.Noise owner]

Newsgroups: info.grass.user
Path: zorro.cecer.army.mil!shapiro
From: shapiro at zorro.cecer.army.mil (Michael Shapiro)
Subject: Re: scaling imagery
Message-ID: <C800Dp.E5u at news.cecer.army.mil>
Sender: news at news.cecer.army.mil (Net.Noise owner)
Organization: US Army Corps of Engineers Construction Engineering Research Labs
References: <9305291454.AA04430 at wilson>
Date: Wed, 2 Jun 1993 14:32:13 GMT
Lines: 25

In <9305291454.AA04430 at wilson> kelly at crseo.ucsb.edu (Kelly Elder) writes:


>In the past I have used ipw for image processing, and
>before using clustering and classifying routines I have
>scaled the input values from the imagery (i.e. mean=0,
>std.dev=1).

>I would like to use the grass routines, i.cluster and
>i.maxlik for future imagery work and for classification
>schemes of nonimagery data. The obvious problem is the
>inability of grass to handle floating point numbers,
>which means scaling by conventional methods is impossible.
>Is there a statistically sound way around this problem?

>thank you, kelly

GRASS will handle 32 bit integers raster maps. It treats images
as raster maps. So scale them up to large integers (provided ipw
can do this).
-- 
Michael Shapiro               shapiro at zorro.cecer.army.mil
U.S. Army CERL                (217) 373-7277
P.O. Box 9005 
Champaign, Ill. 61826-9005