Commit: r2477: Avoid that multipolygon cutting algorithm creates very small polygons
Version 2477 was committed by wanmil on Wed, 06 Feb 2013 Avoid that multipolygon cutting algorithm creates very small polygons The algorithm has been changed so that cuts with a high aspect ratio are preferred to those that cut multiple inner polygons. This avoids that the cut algorithm creates narrow but wide polygons that are filtered out when zoomed out. This avoids white stripes in sea areas.
Are new sea polygons needed for this? On 06.02.2013 20:08, svn commit wrote:
Version 2477 was committed by wanmil on Wed, 06 Feb 2013
Avoid that multipolygon cutting algorithm creates very small polygons
The algorithm has been changed so that cuts with a high aspect ratio are preferred to those that cut multiple inner polygons. This avoids that the cut algorithm creates narrow but wide polygons that are filtered out when zoomed out. This avoids white stripes in sea areas. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
Felix Hartmann-2 wrote
Are new sea polygons needed for this?
If you want to benefit from the new algo: yes Gerd -- View this message in context: http://gis.19327.n5.nabble.com/Commit-r2477-Avoid-that-multipolygon-cutting-... Sent from the Mkgmap Development mailing list archive at Nabble.com.
So (@Wanmil) are theese alreay created using the new algo? http://www.navmaps.eu/wanmil/sea_20130205.zip On 07.02.2013 11:14, GerdP wrote:
Felix Hartmann-2 wrote
Are new sea polygons needed for this? If you want to benefit from the new algo: yes
Gerd
-- View this message in context: http://gis.19327.n5.nabble.com/Commit-r2477-Avoid-that-multipolygon-cutting-... Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
okay tried it out. Seems to be the new algo. I must say, it looks a bit nicer, and what is more important it saves data / means faster display on the GPS. What I still not really understand is why the following is happening: See the two screenshots taken with Mapsource and details=normal 50km -- correct: http://openmtbmap.org/norway1.jpg 70km -- see disappears mostly: http://openmtbmap.org/norway2.jpg I don't really understand why the see is left away so much. On 07.02.2013 11:19, Felix Hartmann wrote:
So (@Wanmil) are theese alreay created using the new algo?
http://www.navmaps.eu/wanmil/sea_20130205.zip On 07.02.2013 11:14, GerdP wrote:
Felix Hartmann-2 wrote
Are new sea polygons needed for this? If you want to benefit from the new algo: yes
Gerd
-- View this message in context: http://gis.19327.n5.nabble.com/Commit-r2477-Avoid-that-multipolygon-cutting-... Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
Well, I just tried it out with min-size-polygon=12 instead of 18, then it works quite nicely. So the sea polygon parser is still cutting rather small sea polygons. It should rather be the opposite way around I suppose though. Through away small islands instead of throwing away the sea. This is using: --generate-sea --precomp-sea=\maps\sea On 07.02.2013 11:58, Felix Hartmann wrote:
okay tried it out. Seems to be the new algo. I must say, it looks a bit nicer, and what is more important it saves data / means faster display on the GPS. What I still not really understand is why the following is happening: See the two screenshots taken with Mapsource and details=normal 50km -- correct: http://openmtbmap.org/norway1.jpg
70km -- see disappears mostly: http://openmtbmap.org/norway2.jpg
I don't really understand why the see is left away so much. On 07.02.2013 11:19, Felix Hartmann wrote:
So (@Wanmil) are theese alreay created using the new algo?
http://www.navmaps.eu/wanmil/sea_20130205.zip On 07.02.2013 11:14, GerdP wrote:
Felix Hartmann-2 wrote
Are new sea polygons needed for this? If you want to benefit from the new algo: yes
Gerd
-- View this message in context: http://gis.19327.n5.nabble.com/Commit-r2477-Avoid-that-multipolygon-cutting-... Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
-- View this message in context: http://gis.19327.n5.nabble.com/Commit-r2477-Avoid-that-multipolygon-cutting-... Sent from the Mkgmap Development mailing list archive at Nabble.com.
Felix Hartmann-2 wrote
Well, I just tried it out with min-size-polygon=12 instead of 18, then it works quite nicely. So the sea polygon parser is still cutting rather small sea polygons. It should rather be the opposite way around I suppose though. Through away small islands instead of throwing away the sea.
This is using: --generate-sea --precomp-sea=\maps\sea
It is the filter that throws away small shapes. The cut algo just creates sea and land shapes which cover the whole area. Later, the shape size filter throws away small shapes, no matter what they are. So, until WanMil has finished the work on the new mp-cut branch you should see better results if you use the default for min-size-polygon (8) or even a smaller value. Gerd -- View this message in context: http://gis.19327.n5.nabble.com/Commit-r2477-Avoid-that-multipolygon-cutting-... Sent from the Mkgmap Development mailing list archive at Nabble.com.
On 07.02.2013 13:31, GerdP wrote:
Felix Hartmann-2 wrote
Well, I just tried it out with min-size-polygon=12 instead of 18, then it works quite nicely. So the sea polygon parser is still cutting rather small sea polygons. It should rather be the opposite way around I suppose though. Through away small islands instead of throwing away the sea.
This is using: --generate-sea --precomp-sea=\maps\sea It is the filter that throws away small shapes. The cut algo just creates sea and land shapes which cover the whole area. Later, the shape size filter throws away small shapes, no matter what they are. So, until WanMil has finished the work on the new mp-cut branch you should see better results if you use the default for min-size-polygon (8) or even a smaller value.
Gerd
My thinking was more, I want to have rather sea than land. So throw away small islands, instead of throwing away sea. So run a filter throwing away islands first IF the area is covered by precompiled sea. I'ld have to give it a try again, but I'm pretty sure that the old sea-generation not using precompiled sea, worked much more like this and did rather drop islands than sea (or rather flood things, than miss sea).
participants (3)
-
Felix Hartmann -
GerdP -
svn commit