[GRASS-user] A problem with i.gensigset.

Joel1 · April 6, 2010, 2:15pm

Hello to all,

I trying to classify remote sensing imagery, and
I thought to try the i.smap module, but
when I run the i.gensigset, I always
get the same problem:

…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
Clustering class 12 (184 pixels)…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
i.gensigset complete.

And afterwards the signature file doesn’t have any
spectral information. I really don’t understand whats happening
and the cause to this outcome.

Can anyone solve this “enigma”?

Ps: I don’t know if this information is important: I collected
about 450 samples distributed by 10 classes. I may also
add that the average number of pixels per sample is around 50.

And thanks list, for the previous tips: they were very useful.

Nikos_Alexandris1 · April 6, 2010, 3:04pm

joel dinis wrote:

I trying to classify remote sensing imagery, and
I thought to try the i.smap module, but
when I run the i.gensigset,

Please, could you copy-paste the exact command you used?
Does the "maxsig=" (e.g. maxsig=20) make any difference?

Nikos

I always
get the same problem:
...
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
Clustering class 12 (184 pixels)...
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
i.gensigset complete.

And afterwards the signature file doesn't have any
spectral information. I really don't understand whats happening
and the cause to this outcome.

Can anyone solve this "enigma"?

Ps: I don't know if this information is important: I collected
about 450 samples distributed by 10 classes. I may also
add that the average number of pixels per sample is around 50.

And thanks list, for the previous tips: they were very useful.

Joel1 · April 6, 2010, 3:09pm

Hello Nikos,

In fact, I put maxsig=30.
i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo signaturefile=assinaturas maxsig=30

JD

On Tue, Apr 6, 2010 at 4:04 PM, Nikos Alexandris <nikos.alexandris@felis.uni-freiburg.de> wrote:

joel dinis wrote:

I trying to classify remote sensing imagery, and
I thought to try the i.smap module, but
when I run the i.gensigset,

Please, could you copy-paste the exact command you used?
Does the “maxsig=” (e.g. maxsig=20) make any difference?

Nikos

I always
get the same problem:
…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
Clustering class 12 (184 pixels)…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
i.gensigset complete.

And afterwards the signature file doesn’t have any
spectral information. I really don’t understand whats happening
and the cause to this outcome.

Can anyone solve this “enigma”?

Ps: I don’t know if this information is important: I collected
about 450 samples distributed by 10 classes. I may also
add that the average number of pixels per sample is around 50.

And thanks list, for the previous tips: they were very useful.

Nikos_Alexandris1 · April 6, 2010, 3:33pm

(sorry for re-sorting the message, just to keep it easy to follow-up)

joel dinis:

        > I trying to classify remote sensing imagery, and
        > I thought to try the i.smap module, but
        > when I run the i.gensigset,

Nikos Alexandris:

Please, could you copy-paste the exact command you used?
Does the "maxsig=" (e.g. maxsig=20) make any difference?

joel dinis:

In fact, I put maxsig=30.
i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
signaturefile=assinaturas maxsig=30

What about smaller maxsig's, like:

- maxsig=25
- maxsig=20
- maxsig=15

?

Nikos

        > I always
        > get the same problem:
        > ...
        > Removed a singular subsignature number 1 (9 remain)
        > Removed a singular subsignature number 1 (8 remain)
        > Removed a singular subsignature number 1 (7 remain)
        > Removed a singular subsignature number 1 (6 remain)
        > Removed a singular subsignature number 1 (5 remain)
        > Removed a singular subsignature number 1 (4 remain)
        > Removed a singular subsignature number 1 (3 remain)
        > Removed a singular subsignature number 1 (2 remain)
        > Removed a singular subsignature number 1 (1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Removed a singular subsignature number 1 (-1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Number of subclasses is 0
        > Clustering class 12 (184 pixels)...
        > Removed a singular subsignature number 1 (9 remain)
        > Removed a singular subsignature number 1 (8 remain)
        > Removed a singular subsignature number 1 (7 remain)
        > Removed a singular subsignature number 1 (6 remain)
        > Removed a singular subsignature number 1 (5 remain)
        > Removed a singular subsignature number 1 (4 remain)
        > Removed a singular subsignature number 1 (3 remain)
        > Removed a singular subsignature number 1 (2 remain)
        > Removed a singular subsignature number 1 (1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Removed a singular subsignature number 1 (-1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Number of subclasses is 0
        > i.gensigset complete.
        >
        > And afterwards the signature file doesn't have any
        > spectral information. I really don't understand whats
        happening
        > and the cause to this outcome.
        >
        > Can anyone solve this "enigma"?
        >
        > Ps: I don't know if this information is important: I
        collected
        > about 450 samples distributed by 10 classes. I may also
        > add that the average number of pixels per sample is around
        50.
        >
        > And thanks list, for the previous tips: they were very
        useful.

Nikos_Alexandris1 · April 6, 2010, 3:38pm

joel dinis:

> In fact, I put maxsig=30.
> i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
> signaturefile=assinaturas maxsig=30

NikosA:

What about smaller maxsig's, like:

- maxsig=25
- maxsig=20
- maxsig=15

and even smaller than 10 perhaps since the algorithm indicates "9
remain" after the first removal.

Are the "sub-signature removals of the same amount for all classes"?

Nikos

> > I always
> > get the same problem:
> > ...
> > Removed a singular subsignature number 1 (9 remain)
> > Removed a singular subsignature number 1 (8 remain)
> > Removed a singular subsignature number 1 (7 remain)
> > Removed a singular subsignature number 1 (6 remain)
> > Removed a singular subsignature number 1 (5 remain)
> > Removed a singular subsignature number 1 (4 remain)
> > Removed a singular subsignature number 1 (3 remain)
> > Removed a singular subsignature number 1 (2 remain)
> > Removed a singular subsignature number 1 (1 remain)
> > Unreliable clustering. Try a smaller initial number of
> clusters
> > Removed a singular subsignature number 1 (-1 remain)
> > Unreliable clustering. Try a smaller initial number of
> clusters
> > Number of subclasses is 0
> > Clustering class 12 (184 pixels)...
> > Removed a singular subsignature number 1 (9 remain)
> > Removed a singular subsignature number 1 (8 remain)
> > Removed a singular subsignature number 1 (7 remain)
> > Removed a singular subsignature number 1 (6 remain)
> > Removed a singular subsignature number 1 (5 remain)
> > Removed a singular subsignature number 1 (4 remain)
> > Removed a singular subsignature number 1 (3 remain)
> > Removed a singular subsignature number 1 (2 remain)
> > Removed a singular subsignature number 1 (1 remain)
> > Unreliable clustering. Try a smaller initial number of
> clusters
> > Removed a singular subsignature number 1 (-1 remain)
> > Unreliable clustering. Try a smaller initial number of
> clusters
> > Number of subclasses is 0
> > i.gensigset complete.
> >
> > And afterwards the signature file doesn't have any
> > spectral information. I really don't understand whats
> happening
> > and the cause to this outcome.
> >
> > Can anyone solve this "enigma"?
> >
> > Ps: I don't know if this information is important: I
> collected
> > about 450 samples distributed by 10 classes. I may also
> > add that the average number of pixels per sample is around
> 50.
> >
> > And thanks list, for the previous tips: they were very
> useful.

Micha_Silver · April 6, 2010, 3:39pm

Nikos Alexandris wrote:

(sorry for re-sorting the message, just to keep it easy to follow-up)

joel dinis:


        > I trying to classify remote sensing imagery, and
        > I thought to try the i.smap module, but
        > when I run the i.gensigset,

Nikos Alexandris:

        Please, could you copy-paste the exact command you used?
        Does the "maxsig=" (e.g. maxsig=20) make any difference?

joel dinis:

In fact, I put maxsig=30.
i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
signaturefile=assinaturas maxsig=30

What about smaller maxsig's, like:

- maxsig=25
- maxsig=20
- maxsig=15

?

Nikos

Also it might help to know what is contained in the trainingmap. i.e:
r.info -r amostras

                > I always
        > get the same problem:
        > ...
        > Removed a singular subsignature number 1 (9 remain)
        > Removed a singular subsignature number 1 (8 remain)
        > Removed a singular subsignature number 1 (7 remain)
        > Removed a singular subsignature number 1 (6 remain)
        > Removed a singular subsignature number 1 (5 remain)
        > Removed a singular subsignature number 1 (4 remain)
        > Removed a singular subsignature number 1 (3 remain)
        > Removed a singular subsignature number 1 (2 remain)
        > Removed a singular subsignature number 1 (1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Removed a singular subsignature number 1 (-1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Number of subclasses is 0
        > Clustering class 12 (184 pixels)...
        > Removed a singular subsignature number 1 (9 remain)
        > Removed a singular subsignature number 1 (8 remain)
        > Removed a singular subsignature number 1 (7 remain)
        > Removed a singular subsignature number 1 (6 remain)
        > Removed a singular subsignature number 1 (5 remain)
        > Removed a singular subsignature number 1 (4 remain)
        > Removed a singular subsignature number 1 (3 remain)
        > Removed a singular subsignature number 1 (2 remain)
        > Removed a singular subsignature number 1 (1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Removed a singular subsignature number 1 (-1 remain)
        > Unreliable clustering. Try a smaller initial number of
        clusters
        > Number of subclasses is 0
        > i.gensigset complete.
        >
        > And afterwards the signature file doesn't have any
        > spectral information. I really don't understand whats
        happening
        > and the cause to this outcome.
        >
        > Can anyone solve this "enigma"?
        >
        > Ps: I don't know if this information is important: I
        collected
        > about 450 samples distributed by 10 classes. I may also
        > add that the average number of pixels per sample is around
        50.
        >
        > And thanks list, for the previous tips: they were very
        useful.

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

This mail was received via Mail-SeCure System.

--
Micha Silver
Arava Development Co. +972-52-3665918
http://www.surfaces.co.il

Joel1 · April 6, 2010, 3:42pm

It remains.

I tried, as you sugested, and the problem
remains.

…
Clustering class 9 (4994 pixels)…
Removed a singular subsignature number 1 (24 remain)
Removed a singular subsignature number 1 (23 remain)
Removed a singular subsignature number 1 (22 remain)
Removed a singular subsignature number 1 (21 remain)
Removed a singular subsignature number 1 (20 remain)
Removed a singular subsignature number 1 (19 remain)
Removed a singular subsignature number 1 (18 remain)
Removed a singular subsignature number 1 (17 remain)
Removed a singular subsignature number 1 (16 remain)
Removed a singular subsignature number 1 (15 remain)
Removed a singular subsignature number 1 (14 remain)
Removed a singular subsignature number 1 (13 remain)
Removed a singular subsignature number 1 (12 remain)
Removed a singular subsignature number 1 (11 remain)
Removed a singular subsignature number 1 (10 remain)
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
Clustering class 10 (2147 pixels)…
Removed a singular subsignature number 1 (24 remain)
Removed a singular subsignature number 1 (23 remain)
Removed a singular subsignature number 1 (22 remain)
Removed a singular subsignature number 1 (21 remain)
Removed a singular subsignature number 1 (20 remain)
Removed a singular subsignature number 1 (19 remain)
Removed a singular subsignature number 1 (18 remain)
Removed a singular subsignature number 1 (17 remain)
Removed a singular subsignature number 1 (16 remain)
Removed a singular subsignature number 1 (15 remain)
Removed a singular subsignature number 1 (14 remain)
Removed a singular subsignature number 1 (13 remain)
Removed a singular subsignature number 1 (12 remain)
Removed a singular subsignature number 1 (11 remain)
Removed a singular subsignature number 1 (10 remain)
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
Clustering class 11 (266 pixels)…
Removed a singular subsignature number 1 (24 remain)
Removed a singular subsignature number 1 (23 remain)
Removed a singular subsignature number 1 (22 remain)
Removed a singular subsignature number 1 (21 remain)
Removed a singular subsignature number 1 (20 remain)
Removed a singular subsignature number 1 (19 remain)
Removed a singular subsignature number 1 (18 remain)
Removed a singular subsignature number 1 (17 remain)
Removed a singular subsignature number 1 (16 remain)
Removed a singular subsignature number 1 (15 remain)
Removed a singular subsignature number 1 (14 remain)
Removed a singular subsignature number 1 (13 remain)
Removed a singular subsignature number 1 (12 remain)
Removed a singular subsignature number 1 (11 remain)
Removed a singular subsignature number 1 (10 remain)
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
Clustering class 12 (184 pixels)…
Too many subclasses for class index 12
Number of subclasses set to 21
Removed a singular subsignature number 1 (20 remain)
Removed a singular subsignature number 1 (19 remain)
Removed a singular subsignature number 1 (18 remain)
Removed a singular subsignature number 1 (17 remain)
Removed a singular subsignature number 1 (16 remain)
Removed a singular subsignature number 1 (15 remain)
Removed a singular subsignature number 1 (14 remain)
Removed a singular subsignature number 1 (13 remain)
Removed a singular subsignature number 1 (12 remain)
Removed a singular subsignature number 1 (11 remain)
Removed a singular subsignature number 1 (10 remain)
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of clusters
Number of subclasses is 0
i.gensigset complete.

Yes, this happen to all the classes…

Maybe is something prior to the i.gensigset run.
I’ll investigate.

On Tue, Apr 6, 2010 at 4:33 PM, Nikos Alexandris <nikos.alexandris@felis.uni-freiburg.de> wrote:

(sorry for re-sorting the message, just to keep it easy to follow-up)

joel dinis:

I trying to classify remote sensing imagery, and
I thought to try the i.smap module, but
when I run the i.gensigset,

Nikos Alexandris:

Please, could you copy-paste the exact command you used?
Does the “maxsig=” (e.g. maxsig=20) make any difference?

joel dinis:

In fact, I put maxsig=30.
i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
signaturefile=assinaturas maxsig=30

What about smaller maxsig’s, like:

maxsig=25

maxsig=20

maxsig=15

?

Nikos

I always
get the same problem:
…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Number of subclasses is 0
Clustering class 12 (184 pixels)…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Number of subclasses is 0
i.gensigset complete.

And afterwards the signature file doesn’t have any
spectral information. I really don’t understand whats
happening
and the cause to this outcome.

Can anyone solve this “enigma”?

Ps: I don’t know if this information is important: I
collected
about 450 samples distributed by 10 classes. I may also
add that the average number of pixels per sample is around

And thanks list, for the previous tips: they were very
useful.

neteler · April 6, 2010, 3:46pm

On Tue, Apr 6, 2010 at 5:42 PM, joel dinis <dinis.joel@gmail.com> wrote:

It remains.

I tried, as you sugested, and the problem
remains.

...
Clustering class 9 (4994 pixels)...

...

Number of subclasses is 0
i.gensigset complete.

Yes, this happen to all the classes...

Maybe is something prior to the i.gensigset run.
I'll investigate.

It this point it might be the best to make the relevant parts
of the location available (even privately).

Markus

Joel1 · April 6, 2010, 3:47pm

Hello Micha,

I put r.info -r amostras
as sugested, get the following:
min=112
max=512

(This values are index to my classes.)

JD

On Tue, Apr 6, 2010 at 4:39 PM, Micha Silver <micha@arava.co.il> wrote:

Nikos Alexandris wrote:

(sorry for re-sorting the message, just to keep it easy to follow-up)

joel dinis:

I trying to classify remote sensing imagery, and
I thought to try the i.smap module, but
when I run the i.gensigset,

Nikos Alexandris:

Please, could you copy-paste the exact command you used?
Does the “maxsig=” (e.g. maxsig=20) make any difference?

joel dinis:

In fact, I put maxsig=30.
i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
signaturefile=assinaturas maxsig=30

What about smaller maxsig’s, like:

maxsig=25

maxsig=20

maxsig=15

?

Nikos

Also it might help to know what is contained in the trainingmap. i.e:
r.info -r amostras

I always
get the same problem:
…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Number of subclasses is 0
Clustering class 12 (184 pixels)…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial number of
clusters
Number of subclasses is 0
i.gensigset complete.

And afterwards the signature file doesn’t have any
spectral information. I really don’t understand whats
happening
and the cause to this outcome.

Can anyone solve this “enigma”?

Ps: I don’t know if this information is important: I
collected
about 450 samples distributed by 10 classes. I may also
add that the average number of pixels per sample is around

And thanks list, for the previous tips: they were very
useful.

grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

This mail was received via Mail-SeCure System.

–
Micha Silver
Arava Development Co. +972-52-3665918
http://www.surfaces.co.il

Nikos_Alexandris1 · April 6, 2010, 4:42pm

--%<---

...
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial number of clusters

--%<---

As a side note:

while reading the manual of i.gensigset, I think it would be better to
"rephrase" either the above output message or the the manual directly to
make it clear that the "initial number of clusters" refers to... (what
is it?) "maximum number of (initial) subclasses" ?

Nikos

Micha_Silver · April 6, 2010, 5:44pm

joel dinis wrote:

Hello Micha,

I put r.info <http://r.info/> -r amostras
as sugested, get the following:
min=112
max=512

So, if I understand correctly, you are trying to classify into 400 separate classes?!
I'm a beginner with image classification, and others with more experience will know better, but I'll bet that's your problem. There must be many separate training areas (with different cat values) with a very similar spectral signature, and the algorithm is not able to differentiate between them.

(This values are index to my classes.)

JD

On Tue, Apr 6, 2010 at 4:39 PM, Micha Silver <micha@arava.co.il <mailto:micha@arava.co.il>> wrote:

    Nikos Alexandris wrote:

        (sorry for re-sorting the message, just to keep it easy to
        follow-up)

        joel dinis:

                   > I trying to classify remote sensing imagery, and
                   > I thought to try the i.smap module, but
                   > when I run the i.gensigset,

        Nikos Alexandris:

                   Please, could you copy-paste the exact command you
            used?
                   Does the "maxsig=" (e.g. maxsig=20) make any
            difference?

        joel dinis:

            In fact, I put maxsig=30.
            i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
            signaturefile=assinaturas maxsig=30

        What about smaller maxsig's, like:

        - maxsig=25
        - maxsig=20
        - maxsig=15

        ?

        Nikos

    Also it might help to know what is contained in the trainingmap. i.e:
    r.info <http://r.info> -r amostras

                           > I always
                   > get the same problem:
                   > ...
                   > Removed a singular subsignature number 1 (9 remain)
                   > Removed a singular subsignature number 1 (8 remain)
                   > Removed a singular subsignature number 1 (7 remain)
                   > Removed a singular subsignature number 1 (6 remain)
                   > Removed a singular subsignature number 1 (5 remain)
                   > Removed a singular subsignature number 1 (4 remain)
                   > Removed a singular subsignature number 1 (3 remain)
                   > Removed a singular subsignature number 1 (2 remain)
                   > Removed a singular subsignature number 1 (1 remain)
                   > Unreliable clustering. Try a smaller initial
            number of
                   clusters
                   > Removed a singular subsignature number 1 (-1 remain)
                   > Unreliable clustering. Try a smaller initial
            number of
                   clusters
                   > Number of subclasses is 0
                   > Clustering class 12 (184 pixels)...
                   > Removed a singular subsignature number 1 (9 remain)
                   > Removed a singular subsignature number 1 (8 remain)
                   > Removed a singular subsignature number 1 (7 remain)
                   > Removed a singular subsignature number 1 (6 remain)
                   > Removed a singular subsignature number 1 (5 remain)
                   > Removed a singular subsignature number 1 (4 remain)
                   > Removed a singular subsignature number 1 (3 remain)
                   > Removed a singular subsignature number 1 (2 remain)
                   > Removed a singular subsignature number 1 (1 remain)
                   > Unreliable clustering. Try a smaller initial
            number of
                   clusters
                   > Removed a singular subsignature number 1 (-1 remain)
                   > Unreliable clustering. Try a smaller initial
            number of
                   clusters
                   > Number of subclasses is 0
                   > i.gensigset complete.
                   >
                   > And afterwards the signature file doesn't have any
                   > spectral information. I really don't understand whats
                   happening
                   > and the cause to this outcome.
                   >
                   > Can anyone solve this "enigma"?
                   >
                   > Ps: I don't know if this information is important: I
                   collected
                   > about 450 samples distributed by 10 classes. I
            may also
                   > add that the average number of pixels per sample
            is around
                   50.
                   >
                   > And thanks list, for the previous tips: they were
            very
                   useful.

        _______________________________________________
        grass-user mailing list
        grass-user@lists.osgeo.org <mailto:grass-user@lists.osgeo.org>
        http://lists.osgeo.org/mailman/listinfo/grass-user

        This mail was received via Mail-SeCure System.

    -- Micha Silver
    Arava Development Co. +972-52-3665918
    http://www.surfaces.co.il

This mail was received via Mail-SeCure System.

--
Micha Silver
Arava Development Co. +972-52-3665918
http://www.surfaces.co.il

Nikos_Alexandris1 · April 6, 2010, 6:14pm

On Tue, 2010-04-06 at 20:44 +0300, Micha Silver wrote:

joel dinis wrote:

> Hello Micha,
>
> I put r.info <http://r.info/> -r amostras
> as sugested, get the following:
> min=112
> max=512
>
So, if I understand correctly, you are trying to classify into 400
separate classes?!
I'm a beginner with image classification, and others with more
experience will know better, but I'll bet that's your problem. There
must be many separate training areas (with different cat values) with a
very similar spectral signature, and the algorithm is not able to
differentiate between them.

I 've also thought about that Micha (I was about to suggest Joel to use
less training classes), but I was hesitant to say it :-p

Nikos

> (This values are index to my classes.)
>
> JD
>
> On Tue, Apr 6, 2010 at 4:39 PM, Micha Silver <micha@arava.co.il
> <mailto:micha@arava.co.il>> wrote:
>
> Nikos Alexandris wrote:
>
> (sorry for re-sorting the message, just to keep it easy to
> follow-up)
>
> joel dinis:
>
>
> > I trying to classify remote sensing imagery, and
> > I thought to try the i.smap module, but
> > when I run the i.gensigset,
>
>
>
> Nikos Alexandris:
>
>
> Please, could you copy-paste the exact command you
> used?
> Does the "maxsig=" (e.g. maxsig=20) make any
> difference?
>
>
>
> joel dinis:
>
>
>
> In fact, I put maxsig=30.
> i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
> signaturefile=assinaturas maxsig=30
>
>
>
> What about smaller maxsig's, like:
>
> - maxsig=25
> - maxsig=20
> - maxsig=15
>
> ?
>
> Nikos
>
>
>
> Also it might help to know what is contained in the trainingmap. i.e:
> r.info <http://r.info> -r amostras
>
>
>
> > I always
> > get the same problem:
> > ...
> > Removed a singular subsignature number 1 (9 remain)
> > Removed a singular subsignature number 1 (8 remain)
> > Removed a singular subsignature number 1 (7 remain)
> > Removed a singular subsignature number 1 (6 remain)
> > Removed a singular subsignature number 1 (5 remain)
> > Removed a singular subsignature number 1 (4 remain)
> > Removed a singular subsignature number 1 (3 remain)
> > Removed a singular subsignature number 1 (2 remain)
> > Removed a singular subsignature number 1 (1 remain)
> > Unreliable clustering. Try a smaller initial
> number of
> clusters
> > Removed a singular subsignature number 1 (-1 remain)
> > Unreliable clustering. Try a smaller initial
> number of
> clusters
> > Number of subclasses is 0
> > Clustering class 12 (184 pixels)...
> > Removed a singular subsignature number 1 (9 remain)
> > Removed a singular subsignature number 1 (8 remain)
> > Removed a singular subsignature number 1 (7 remain)
> > Removed a singular subsignature number 1 (6 remain)
> > Removed a singular subsignature number 1 (5 remain)
> > Removed a singular subsignature number 1 (4 remain)
> > Removed a singular subsignature number 1 (3 remain)
> > Removed a singular subsignature number 1 (2 remain)
> > Removed a singular subsignature number 1 (1 remain)
> > Unreliable clustering. Try a smaller initial
> number of
> clusters
> > Removed a singular subsignature number 1 (-1 remain)
> > Unreliable clustering. Try a smaller initial
> number of
> clusters
> > Number of subclasses is 0
> > i.gensigset complete.
> >
> > And afterwards the signature file doesn't have any
> > spectral information. I really don't understand whats
> happening
> > and the cause to this outcome.
> >
> > Can anyone solve this "enigma"?
> >
> > Ps: I don't know if this information is important: I
> collected
> > about 450 samples distributed by 10 classes. I
> may also
> > add that the average number of pixels per sample
> is around
> 50.
> >
> > And thanks list, for the previous tips: they were
> very
> useful.
>
>
>
>
>
> _______________________________________________
> grass-user mailing list
> grass-user@lists.osgeo.org <mailto:grass-user@lists.osgeo.org>
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
> This mail was received via Mail-SeCure System.
>
>
>
>
>
>
> --
> Micha Silver
> Arava Development Co. +972-52-3665918
> http://www.surfaces.co.il
>
>
>
>
> This mail was received via Mail-SeCure System.

Joel1 · April 6, 2010, 7:59pm

NO NO NO!

I have 30 CORINE land cover classes to classify.
The number 512, for example, means 5.1.2, is a
nominal number, IS NOT an classe id.

But may be right: may be the way the classes are
represented in the system cause that problem. I’ll test
that…

Thanks
JD

On Tue, Apr 6, 2010 at 6:44 PM, Micha Silver <micha@arava.co.il> wrote:

joel dinis wrote:

Hello Micha,

I put r.info <http://r.info/> -r amostras

as sugested, get the following:
min=112
max=512

So, if I understand correctly, you are trying to classify into 400 separate classes?!
I’m a beginner with image classification, and others with more experience will know better, but I’ll bet that’s your problem. There must be many separate training areas (with different cat values) with a very similar spectral signature, and the algorithm is not able to differentiate between them.

(This values are index to my classes.)

JD

On Tue, Apr 6, 2010 at 4:39 PM, Micha Silver <micha@arava.co.il mailto:[micha@arava.co.il](mailto:micha@arava.co.il)> wrote:

Nikos Alexandris wrote:

(sorry for re-sorting the message, just to keep it easy to
follow-up)

joel dinis:

I trying to classify remote sensing imagery, and
I thought to try the i.smap module, but
when I run the i.gensigset,

Nikos Alexandris:

Please, could you copy-paste the exact command you
used?
Does the “maxsig=” (e.g. maxsig=20) make any
difference?

joel dinis:

In fact, I put maxsig=30.
i.gensigset trainingmap=amostras group=grupo subgroup=subgrupo
signaturefile=assinaturas maxsig=30

What about smaller maxsig’s, like:

maxsig=25

maxsig=20

maxsig=15

?

Nikos

Also it might help to know what is contained in the trainingmap. i.e:

r.info <http://r.info> -r amostras

I always
get the same problem:
…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial
number of
clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial
number of
clusters
Number of subclasses is 0
Clustering class 12 (184 pixels)…
Removed a singular subsignature number 1 (9 remain)
Removed a singular subsignature number 1 (8 remain)
Removed a singular subsignature number 1 (7 remain)
Removed a singular subsignature number 1 (6 remain)
Removed a singular subsignature number 1 (5 remain)
Removed a singular subsignature number 1 (4 remain)
Removed a singular subsignature number 1 (3 remain)
Removed a singular subsignature number 1 (2 remain)
Removed a singular subsignature number 1 (1 remain)
Unreliable clustering. Try a smaller initial
number of
clusters
Removed a singular subsignature number 1 (-1 remain)
Unreliable clustering. Try a smaller initial
number of
clusters
Number of subclasses is 0
i.gensigset complete.

And afterwards the signature file doesn’t have any
spectral information. I really don’t understand whats
happening
and the cause to this outcome.

Can anyone solve this “enigma”?

Ps: I don’t know if this information is important: I
collected
about 450 samples distributed by 10 classes. I
may also
add that the average number of pixels per sample
is around

And thanks list, for the previous tips: they were
very
useful.

grass-user mailing list

grass-user@lists.osgeo.org mailto:[grass-user@lists.osgeo.org](mailto:grass-user@lists.osgeo.org)

http://lists.osgeo.org/mailman/listinfo/grass-user

This mail was received via Mail-SeCure System.

– Micha Silver
Arava Development Co. +972-52-3665918
http://www.surfaces.co.il

This mail was received via Mail-SeCure System.

–
Micha Silver
Arava Development Co. +972-52-3665918
http://www.surfaces.co.il

Nikos_Alexandris1 · April 6, 2010, 9:07pm

On Tue, 2010-04-06 at 20:59 +0100, joel dinis wrote:

NO NO NO!

I have 30 CORINE land cover classes to classify.
The number 512, for example, means 5.1.2, is a
nominal number, IS NOT an classe id.

But may be right: may be the way the classes are
represented in the system cause that problem. I'll test
that..

OK, I understand. I also use(d) this scheme but only for a few (mostly
major) classes (i.e. 1, 3, 332, 334 and 5).

What does for example "r.stats -cp amostras" say about your training
set?

[...]

Oh... maybe it is best to follow Markus Neteler advice and make part of
the data accessible for testing.

Greets, Nikos