[GRASS-dev] New georectifying module in TclTk

[Michael Barton]

I¹ve nearly finished a new georectifying module in TclTk and want to get it
into the CVS where it can be tested and I can get some help on RMS error
calculation.

Here is a brief rundown on how it works.

Start out in the location/mapset where you want the georectified map to END
UP (i.e., the TARGET location). Open any map display (or displays) that you
want to use for getting geographic coordinates

Select georectify from the file menu

In the georectify startup, you must:
    1) select the mapset of the file to be georectified (the location and
gisdbase are automatically noted)
    2) run i.group to create a group to georectify. Target is automatically
set to the current location and mapset.
    3) select the group to georectify
    4) select the raster map to display for interactive georectification
(normally a raster in the group you want to georectify)
    5) start the georectification

You will have a map display with the raster to georectify and a GCP (ground
control point) manager window.
    1) Click in an xy entry box in the GCP manager
    2) Click on the xy map you want to georectify to put xy coordinates in
the entry box. The focus will shift to the corresponding geographic entry
box in the GCP manager.
    3) Click on a point in a georectified map to get geographic coordinates.
    4) Repeat until you have enough xy and geographic coordinate pairs.

    NOTE: you can edit/enter any  point by typing. You can also edit
existing points by clicking on a new spot in the display.

Click the button to save the points to a POINTS file

Click on the georectify button to select the type of georectification you
want (Is 3rd order georectification still broken?)

That's it.

RMS error calculation does not work yet. There may be some issues still
remaining with zooming in the xy map display.

***************

Anyone who is familiar with how i.rectify works, I need to ask for help on
RMS error calculation. Somehow, I need to actually rectify the GCP's so I
will get meainingful RMS error results. I can't find where to do this in the
C code (semi-incomprehensible to me).

Try it out and let me know what you think.

Michael
__________________________________________
Michael Barton, Professor of Anthropology
School of Human Evolution & Social Change
Center for Social Dynamics and Complexity
Arizona State University

phone: 480-965-6213
fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton