What is the center line of a complex polygon? New approach.
Today, I just show pictures of my process for filtering the medial axis of extraneous elements. I’ll let the pics tell the story for now:
Original Geometry, with ST_Segmentize run to densify the verticesMedial Axis (Calculated outside PostGIS... at this time)Density of vertex points in Medial Axis-- proportional to how closely it matches the direction of the local geometryMedial Axis filtered based on density of vertices. Red segments are the removed ones, blue are the retained. It's not perfect, but it's very cheap computationally... .
3 thoughts on “What is the center line of a complex polygon? New approach.”
I found your post today and I’m very interested in your techniques. I’m conducting some research on river depth probabilities from belief networks but one of my main inputs is the centerline of the river polygon. I already have geoprocessing tools to derive the river polygon quickly but now I need a way to derive the centerline. I found some GP tools but they require ArcInfo and I only have acess to Editor. I’m not that familiar with other GIS softwares do most of my work through the geoprocessor or python. Any help or suggestions would be greatly appreciated.
I found your post today and I’m very interested in your techniques. I’m conducting some research on river depth probabilities from belief networks but one of my main inputs is the centerline of the river polygon. I already have geoprocessing tools to derive the river polygon quickly but now I need a way to derive the centerline. I found some GP tools but they require ArcInfo and I only have acess to Editor. I’m not that familiar with other GIS softwares do most of my work through the geoprocessor or python. Any help or suggestions would be greatly appreciated.
Thanks,
Nate
Hi Nathan,
If you’re thinking of using PostGIS, you can do it in R or pl/python, e.g.
http://www.bostongis.com/PrinterFriendly.aspx?content_name=postgresql_plr_tut02
http://geogeek.garnix.org/2012/04/faster-voronoi-diagrams-in-postgis.html
Best,
Steve