20001024: IMGREMAP: MAG= vs RES= (cont.)

>From: Jason Allard <address@hidden>
>Organization: PSU
>Keywords: 200010240345.e9O3jC425238 McIDAS-X IMGREMAP MAG RES

Jason,

>Thank you for the response, but I guess my understanding of mcidas is
>still a bit weak because I'm still confused.

No problem.  I am most of the time also :-)

>I understand that mag doesn't change the size of the image.

OK, good.

>A negative mag value will reduce the resolution of an image (e.g., -2
will make 4km into 8km)?

Right.

>How is that different than using RES?

RES= is only used when you are specifying the output dataset element
navigation/projection.  It specifies the resolution of the resultant
pixels in the output dataset element.

MAG= controls how the pixels from the source dataset are sampled.
Again, a positive MAG= value replicates pixels; a negative MAG= value
samples pixels.  It seems to me, therefore, that you want to leave MAG=
as one, and adjust the RES= of the output image.

Quick guide:

MAG= is done on the input side; RES= is done on the output side.

>I do need to a pixel-by-pixel comparison of VIS and IR images.
>I'm not sure why you want me to copy the IR to a new dataset element
>and then remap the VIS image to the same location.  Perhaps because I
>don't understand why this step is done,

The best way to compare pixels from two images to leave one of the
images in its native projection (since you would then minimize the
massaging of one part of the data being compared) and remap the other
into it.  After doing this, you are guaranteed to have colocated pixel
values that you can then compare.

You can also remap one image into a different projection (like lambert
conformal as you mention below) specifying the output RES= resolution.
Instead of remapping into the same projection for the other image,
I would copy the first remapped image and then remap into it.  This
way you are guaranteed of both pixel colocation and pixel magnification.

My first example had the implicit assumption that the VIS and IR images
were either not the same size, or did not have the same registration.
In this case, you _must_ transform (remap) one image into the
projection of the other so that you can do a pixel-by-pixel
comparison.  After all, the objective has to be the comparison of pixel
values at the same geolocation for both images.  This is why I
recommended that you make a copy of the IR image and remap the VIS one
into it.

>I also don't understand why I would
>have to use the IMGCHA command afterwards.

When you do an IMGREMAP, the image header information is not
preserved.  The pixel values are what you want, but when you go to
display the image it will have the wrong band (i.e., wavelength
channel), etc. in its file header.  When you let this image sit for
awhile and then go back to it, you may get fooled by the header
information since it may not be what you expect.

>I do know that I do want
>both images to be co-located (same resolution and same location for
>each pixel).

Right, so the best thing to do is remap one into a copy of the other
and then compare the remapped copy with the original.  This was the
example I sent you.

>Perhaps if I describe a little more of what I'd like to do, it'll
>help.

OK.

>I've obtained a program that will detect shallow cumulus clouds,
>but it needs to read both the VIS and IR images to do this (and have
>the same resolution and be co-located).

Do the images have to be in Lambert Conformal projection as well?  If
not, I would leave the images in their native satellite projection.

>I also only want to do this for the midwest region of the US.

Subsecting a portion out of an image is easily done with the IMGCOPY
command;  you can choose the section of an image to copy into a new
image.  You could then clone (copy) this new, smaller image and remap
the other image into the clone (copy).  Then you can do your
pixel-by-pixel comparisons on exactly the regions you want.

>Since the program doesn't have a way
>of limiting what section of the complete GOES image, I need to create a
>subset of the image (for both the VIS and IR).

Use IMGCOPY.

>For images from the
>early 90s, the resolution is the same, so I don't have to worry about
>that, but for the late 90s, the VIS and IR resolutions are different.

I understand.  Will you eventually be doing comparisons between images
from the early 90s to ones from the late 90s?

>My intention was to create a subset of the images using the IMGREMAP
>command since I'd also like to change the projection to a lambert
>projection.

By doing this, you are going to modify the data values in both images
and then compare the modified values.  As long as you realize this,
you will not get into interpretation difficulties.

>With that command, I could also choose my upper left
>corner location and the size of the image. For example, for a VIS image
>from May 10, 1991 (18:00 hours), pretending it has a resolution of 4km
>(its actually 8km):
>
>IMGREMAP WF/051091VIS.8 WF/SUB051091VIS.8 LAT=55 100 PLA=CENTER PRO=LAMB SIZ=2
> 60 260 MAG=-2
>
>For the IR image, same day and time, but with 8km resolution:
>
>IMGREMAP WF/051091IR.19 WF/SUB051091IR.19 LAT=55 100 PLC=CENTER PRO=LAMB SIZ=2
> 60 260

This will work, but if I were going to remap into a Lambert Conformal
projection, I would:

IMGREMAP WF/051091IR.19 WF/SUB051091IR.19 LAT=55 100 PLA=CENTER PRO=LAMB 
SIZ=260 260 MAG=1 RES=8
IMGCOPY WF/SUB051091IR.19 MYDATA/IMAGES.1234 SIZE=ALL
IMGREMAP WF/051091VIS.8 MYDATA/IMAGES.1234

Now, the images that are comparable are WF/SUB051091IR.19 and
MYDATA/IMAGES.1234.

>I guess I thought that this would create a VIS and IR image that would
>be of the same resolution and be co-located.  Am I wrong about this?

No, you are not wrong.  As you can see, there are a number of ways to
accomplish the same objective.  I always recommend, however, to do as
little data massaging as possible.  This is why I immediately went into
not modifying one image while remapping the other into a copy of the
first.

>I guess I'm still not certain why you remaped the VIS image into the same
>location as the IR.

So that their pixels would be guaranteed to be colocated, and so that
the data in the IR image would be as "pristine" as possible.

>Thanks for the help,

I hope that this helped.

Tom