[GRASS-dev] GSoC ideas - vector legend, medial axis

[Markus Metz]

On Thu, Feb 20, 2014 at 10:47 PM, Anna Petrášová <kratochanna at gmail.com> wrote:
>
>
>
> On Thu, Feb 20, 2014 at 4:06 PM, Markus Metz <markus.metz.giswork at gmail.com>
> wrote:
>>
>> On Thu, Feb 20, 2014 at 9:08 AM, Tereza Fiedlerová
>> <tfiedlerova at gmail.com> wrote:
>> > Hello,
>> >
>> > I'm thinking about participating in GSoC this year, if I find some
>> > interesting topic. I went through GRASS GSoC ideas. The two most
>> > interesting
>> > for me are Vector legend and Extracting the medial axis of polygon. I
>> > would
>> > appreciate if anyone can give me more information to these topics.
>> >
>> > The first one is maybe described enough and I just need to study that
>> > more
>> > to understand it better, but still anything more regarding that would be
>> > appreciated.
>> >
>> > The medial axis is not really clear to me. I know what is medial axis,
>> > but I
>> > would like to know more about desired results. How it should be
>> > integrated
>> > to GRASS (some new module?). Is it necessary to write it in C (why not
>> > python)? Should the result algorithm extract the exact medial axis or
>> > the
>> > approximation? ...
>>
>> A good approximation would be fine. It would be necessary to write in
>> C for efficiency reasons and because C library functions need to be
>> heavily used. Additionally, search structures like balanced binary
>> search trees and min heaps are needed which are already implemented in
>> GRASS as C code. I have already written 2 different C modules trying
>> to extract the medial axis, both are incredibly complex and none of
>> them works properly. I am familiar with GRASS, C, and Python and like
>> to use the easiest approach. In this case I am pretty sure that the
>> easiest and most appropriate approach is using C, not Python.
>>
>> Therefore I would recommend a good knowledge of C and some knowledge
>> of the GRASS C API for this project.
>>
>> Markus M
>
>
>
> Just in case it would be useful (at least for Tereza, it's mostly in Czech):
>
> http://web.natur.cuni.cz/~bayertom/Adk/adk7.pdf
>
> this is a presentation on straight skeleton algorithm (if medial axis is the
> same or at least related?).

A straight skeleton is not the same like a medial axis: only the
medial axis is equidistant to the polygon boundaries.

Markus M

>
> Anna
>
>>
>> >
>> > And one more question. Is any of these topics suitable for me if I am
>> > not
>> > experienced in GRASS at all?
>> >
>> > Thank you
>> >
>> > Tereza Fiedlerová
>> >
>> > _______________________________________________
>> > grass-dev mailing list
>> > grass-dev at lists.osgeo.org
>> > http://lists.osgeo.org/mailman/listinfo/grass-dev
>> _______________________________________________
>> grass-dev mailing list
>> grass-dev at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/grass-dev
>
>