Hey, thanks for your help. _I'm precisely looking for 2 axis (convex hull is excluded then) _using the line where the polygon is the thinner as one of the bbox axis is not going to work in several case
Cheers, Rémi-C 2014-10-23 15:46 GMT+02:00 Stephen Woodbridge <[email protected]>: > On 10/23/2014 9:33 AM, Rémi Cura wrote: > >> Hey list, >> somebody has a good idea about how efficiently computing an approximate >> oriented bbox for a polygon ? >> >> The simplest approaches I can think of are >> - iteratively rotate the geom by few degrees, take the envelop. Keep >> ration where envelope is min >> - using Principal Components Analysis : extract points regularly >> spaced on border of object, compute PCA . >> > > A couple of ideas: > > * Compute convex hull and use that as simplification of the geometry. > > * google: postgis minimum polygon width > > if you can get the line representing the minimum width, then rotate the to > be horizontal or vertical > > -Steve > _______________________________________________ > postgis-users mailing list > [email protected] > http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users >
_______________________________________________ postgis-users mailing list [email protected] http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
