i.rectify

[Jim Aanstoos (919) 541-6890]

>  What is the magnitude of errors introduced by the i.rectify
> algorithm? We are comparing two methods of coastal erosion analysis,
> ...

i.rectify uses nearest-neighbor resampling, so the the maximum error
will be half a pixel width.  This is with respect to the grid locations
determined in your i.points session, however, which may add additional
error that depends on a number of factors (number and precision of
control points, suitability of transformation polynomial order for
the type of distortion, etc).  All i.rectify does is resample the image
based on the control points you specified in i.points.  Note that the
new i.rectify2 in Grass 4.1 allows use of a first, second, or third
order transformation matrix while i.rectify is only first order.
            /\
      jim  /--\
          /    \ 
______________________________________________________________________
Jim Aanstoos                      Internet: aanstoos%louie at rcc.rti.org
                                         or aanstoos%louie at rti.rti.org
Research Triangle Institute                                           
PO Box 12194                                       Phone: 919-541-6890
Research Triangle Park, NC 27709-2194                Fax: 919-541-6515
----------------------------------------------------------------------