Closed Fonsan closed 7 years ago
Hi Eric, sorry for the late reply. Let me know if you can find a test case. This is something that I've thought might be problematic, but haven't been able to reproduce other than due to the noise caused by "deflecting" segments as filtered by filter_segment_deflections.py.
Closing this one for now, never found a anything to back it up with
Intuitively anything that has a high density of points will almost naturally be more curvy since if the segment was straighter fewer points would be needed to represent the actual shape. The OSM data seems to handle gpx tracks quite well and filter points that are to dense and are exactly in between the preceding and following point
Looking at https://github.com/adamfranco/curvature/blob/c93f7047ec0b9c4359872820fdba433d4d221669/curvature/post_processors/add_segment_curvature.py#L31
It seems that given high enough density of the underlying data would weight the radii into the lower bonus bracket of levels. I will try to produce a test case proving the bug.