[GRASS5] MLC for sub-pixel classification.

Hello all,
    Grass classifier (i.maxlik) is well tested I think.
But grass laks a sub-pixel classifier.
But adding sub-pixel classification support to existing
Maximum Likelihood classifier seems to be easy(at least
theoretically).
Can we add this facility to i.maxlik
--
Change the rules, or the rules will change you
     ---------Kumaranasan

On Fri, Jul 08, 2005 at 10:20:13AM +0530, Sajith VK wrote:

Hello all,
    Grass classifier (i.maxlik) is well tested I think.
But grass laks a sub-pixel classifier.
But adding sub-pixel classification support to existing
Maximum Likelihood classifier seems to be easy(at least
theoretically).
Can we add this facility to i.maxlik

I have written a spectral unmixing tool for GRASS years
ago and still have to finish the update for 'gmath' lib.
On the list for this summer...

Ref: M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
     C. Furlanello, 2005. An integrated toolbox for image registration,
     fusion and classification. International Journal of
     Geoinformatics. Special Issue on FOSS/GRASS 2004 & GIS-IDEAS 2004,
     1(1), pp. 51-61, March 2005.

Coming (hopefully) soon,

Markus

Good news... I hope it will be good for me(at least).

I checked i.maxlink's source code. I think converting
it into sub-pixel classifier is quite simple.
We are calculating probability vector(Or whatever it may be called),
this can direvtily give class proportions.
Should we try for this?

On 7/8/05, Markus Neteler <neteler@itc.it> wrote:

On Fri, Jul 08, 2005 at 10:20:13AM +0530, Sajith VK wrote:
> Hello all,
> Grass classifier (i.maxlik) is well tested I think.
> But grass laks a sub-pixel classifier.
> But adding sub-pixel classification support to existing
> Maximum Likelihood classifier seems to be easy(at least
> theoretically).
> Can we add this facility to i.maxlik

I have written a spectral unmixing tool for GRASS years
ago and still have to finish the update for 'gmath' lib.
On the list for this summer...

Ref: M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
     C. Furlanello, 2005. An integrated toolbox for image registration,
     fusion and classification. International Journal of
     Geoinformatics. Special Issue on FOSS/GRASS 2004 & GIS-IDEAS 2004,
     1(1), pp. 51-61, March 2005.

Coming (hopefully) soon,

Markus

--
Change the rules, or the rules will change you
     ---------Kumaranasan

On 7/9/05, Sajith VK <sajithvk@gmail.com> wrote:

Good news... I hope it will be good for me(at least).

I checked i.maxlink's source code. I think converting
it into sub-pixel classifier is quite simple.
We are calculating probability vector(Or whatever it may be called),
this can direvtily give class proportions.
Should we try for this?

On 7/8/05, Markus Neteler <neteler@itc.it> wrote:
> On Fri, Jul 08, 2005 at 10:20:13AM +0530, Sajith VK wrote:
> > Hello all,
> > Grass classifier (i.maxlik) is well tested I think.
> > But grass laks a sub-pixel classifier.
> > But adding sub-pixel classification support to existing
> > Maximum Likelihood classifier seems to be easy(at least
> > theoretically).
> > Can we add this facility to i.maxlik
>
> I have written a spectral unmixing tool for GRASS years
> ago and still have to finish the update for 'gmath' lib.
> On the list for this summer...
>
> Ref: M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
> C. Furlanello, 2005. An integrated toolbox for image registration,
> fusion and classification. International Journal of
> Geoinformatics. Special Issue on FOSS/GRASS 2004 & GIS-IDEAS 2004,
> 1(1), pp. 51-61, March 2005.
>
> Coming (hopefully) soon,
>
> Markus
>

--
Change the rules, or the rules will change you
     ---------Kumaranasan

--
Change the rules, or the rules will change you
     ---------Kumaranasan