[GRASSLIST:2672] Re: i.points and composite images

[Kirk R. Wythers]

Thanks Paul,

I plead temporary insanity. I thought I had already added the composite 
image to the group. Upon double checking, I found that I had not.

Kirk
On Feb 17, 2004, at 3:10 PM, Paul Kelly wrote:

> On Tue, 17 Feb 2004, Kirk R. Wythers wrote:
>
>> I am trying to rectify a series of scanned color images. The images
>> were read into an xy location with r.in.gdal, Then the individual rgb
>> bands were recombined with r.composite (for simplicities sake). I ran
>> i.target to assign each group to the target location (a utm location).
>> I was going to remove the individual rgb images (with g.mremove)  and
>> rectify the composite image only, however, upon running i.points, I
>> don't see the composite image as a selection option (only the rgb 
>> bands
>> are present).
>>
>> Is this a strong hint to not remove the rgb bands until after the
>> rectification process is complete?
>
> r.in.gdal automatically creates an imagery group and adds the three 
> bands
> to it. You should run i.group to add the composite image as well. All 4
> images should be in the same group as they cover the same area. You can
> mark the points on one and then rectify them all simultaneously. 
> Probably
> you already know this.
>
> Really what you should do is use the composite image as a backdrop for
> digitising the points in i.points and then rectify the 3 bands. You 
> should
> keep them because you are losing information when you combine them.
>
> Paul
>
>
------------------------------------------------------------------------
Kirk R. Wythers	                                     tel: 612.625.2261
Dept. of Forest Resources                 fax: 612.625.5212
University of Minnesota      email: kwythers at umn.edu
------------------------------------------------------------------------