Hi List

I need to create a statement of the probable error of the area of a
polygon where the coordinates have a standard deviation of 2m (SD=2
say).

Given the tools available in MI, how far wrong is buffering the object
by 3 x SD (99% confidence), subtracting off the MI computed area of the
original and saying that this is a 99% confidence value of the area?

A further tweak on the problem is overlapping a region with coordinate
of SD 2 against a region of coordinate of SD  5 (2 different data sets)
and giving an estimate of the quality of the overlap?

Any help would be appreciated

Tony




Given below is the C code snippet for area calculation. Classically I
should determine the expectations for each sub polygon.


=====================================================================
//
//      Measure the polygon area
//

//
//             A[i]=(poly[i].x - poly[i-1].x)*(poly[i].y +
poly[i-1].y)/2.F;
//             Differentiate twice to get expectations
//
long    Measure_Polygon_Area(double *poly_area,dPoint2D *poly,long
poly_count)
{
long    i;
double  total=0.F;

        *poly_area=0.F;
        if (poly_count < 3) return 0L;

        for(i=1;i<poly_count;i++)
                {
                total += (poly[i].x - poly[i-1].x)*
                         (poly[i].y + poly[i-1].y)/2.F;
                }

        total += (poly[0].x - poly[poly_count-1].x)*
                         (poly[0].y+poly[poly_count-1].y)/2.F;

        if(total < 0.) total *= (-1.);
        *poly_area=total;
        return 1L;
}




The e-mail and any files sent with it are private and intended solely for the 
use of the person or body to whom they are addressed. If you are not the 
intended recipient, you have received this in error and use of this information 
is prohibited.

Nothing in the e-mail amounts to a legal commitment on our part unless 
confirmed by a signed communication.

English Nature will make every effort to keep its network free of viruses. 
However, you will need to scan this message/files for viruses as we can take no 
responsibility for any computer virus that might be transferred.
_______________________________________________
MapInfo-L mailing list
[email protected]
http://www.directionsmag.com/mailman/listinfo/mapinfo-l

Reply via email to