[GRASS-dev] Re: [GRASS-user] low radiance values after i.atcor and i.topo.corr

[Glynn Clements]

Markus Neteler wrote:

> >> > ... without an update to i.gensigset I cannot test the FP capabilities of
> >> > i.smap. It seems that it was cloned from i.gensig.
> >>
> >> In retrospect, it appears that it's only the training map which is
> >> inherently restricted to integers. E.g. i.gensigset reads the group
> >> maps as CELL then immediately promotes the values to "double".
> >>
> >> Unless there's something else I have overlooked, I should have an FP
> >> version ready shortly.
> 
> Great!
> 
> > Done. Again, they compiles without warning, but need testing.
> 
> I have made a test (NC data set):
> 
>  r.mapcalc b10=lsat7_2000_10/10000.

[snip]

> Error: unreliable clustering
> try a smaller initial number of clusters
> Warning: Removed a singular subsignature; number 1; -1 remain
> [..]
> 
> I fear that it still reads in CELL types somewhere and that it loses precision
> through this (leading to 0.0000 stuff and singular subsignatures).

Could it be that the definition of ZERO in subcluster.c:

	#define ZERO 1e-10

is too high when the inputs are scaled down?

For an NxN matrix, scaling all of the the elements down by a factor of
K will scale the determinant by K^N.

AFAICT[1], the matrix elements are roughly proportional to the squares
of the input values, so if you're scaling the values down by 1e-5, the
matrix elements are scaled down by 1e-10, and with 7 bands the
determinant would be scaled down by 1e-70.

[1] Which isn't saying much; it's not like I actually understand the
algorithm being used here.

Can you try dividing by 10, 100, etc rather than 10000, to see if
there's a threshold at which the accuracy starts to be an issue?

And/or make ZERO much smaller.

-- 
Glynn Clements <glynn at gclements.plus.com>