[GRASS5] Re: v.clean rmsa option - what does it do ?

[Moritz Lennert]

Radim Blazek wrote:
> On 4/28/06, Moritz Lennert <mlennert at club.worldonline.be> wrote:
> 
>>Radim Blazek wrote:
>>
>>>On 4/27/06, Moritz Lennert <mlennert at club.worldonline.be> wrote:
>>>
>>>
>>>>Hello Radim,
>>>>
>>>>Could you (or someone else who knows this) give a more thorough
>>>>explanation of what the 'rmsa' option of v.clean does (ideally in some
>>>>graphical form) ?
>>>>
>>>>The man page says "remove small angles between lines at nodes", but I
>>>>don't really understand what this means and the French translators of
>>>>the manuals are trying to understand...
>>>
>>>
>>>If angle between 2 lines is smaller than threshold the lines
>>>are broken and the partialy merged.
>>
>>Thanks for the explanation.
>>
>>So if I understand correctly it does what is shown here:
>>
>>http://geog-pc40.ulb.ac.be/grass/rmsa.pdf
> 
> 
> No, this can be don with 'prune' but tolerance is distance.
> 
> I have attached picture of what rmsa does.


Thanks, that makes it much clearer. However, one more thing. In the 
source code you write:


"It may happen that even if the angle between 2 boundaries at node is 
very small, the calculated angle is 0 because of representation error."

And when one tries to set a threshold, the module says "WARNING: 
Threshold for tool 1 may not be > 0, set to 0".

So, does this mean that this tool only works automatically for those 
nodes where the calculation ends up with an angle of 0 ? The user cannot 
change this threshold.

Should the description of the tool thus read something like this:

rmsa: remove small angles between lines at nodes (Only concerns angles 
which are so small that the calculated angle is 0. The user cannot 
change this threshold.)

Moritz

> 
> Radim
> 
> 
>>i.e. converting the black line with a small angle into the new red line.
>>
>>However, when I try to apply this to this line:
>>
>>http://geog-pc40.ulb.ac.be/grass/rmsa_example.jpg
>>
>>defined as follows in v.build:
>>
>> > v.build brol
>>Building topology ...
>>3 primitives registered
>>Building areas:  100%
>>0 areas built
>>0 isles built
>>Attaching islands:
>>Attaching centroids:  100%
>>Topology was built.
>>Number of nodes     :   4
>>Number of primitives:   3
>>Number of points    :   0
>>Number of lines     :   3
>>Number of boundaries:   0
>>Number of centroids :   0
>>Number of areas     :   0
>>Number of isles     :   0
>>
>>
>>
>>in the following projection:
>>
>>  > g.proj -p
>>-PROJ_INFO-------------------------------------------------
>>name       : Lambert Conformal Conic
>>proj       : lcc
>>datum      : bel72
>>a          : 6378388
>>es         : 0.0067226700
>>lat_1      : 49.83333333333334
>>lat_2      : 51.16666666666666
>>lat_0      : 90
>>lon_0      : 4.356939722222222
>>x_0        : 150000.01256
>>y_0        : 5400088.4378
>>no_defs    : defined
>>-PROJ_UNITS------------------------------------------------
>>unit       : metre
>>units      : metres
>>meters     : 1
>>
>>Then whatever the threshhold value I set (going up to ridiculous values,
>>I continue to get 4 nodes, with  3 lines which look identical to the
>>original.
>>
>>Am I still wrong in my understanding of rmsa, or is it something else ?
>>
>>Moritz
>>
>>
>>
>> ------------------------------------------------------------------------
>>