2

I need to calculate line intersections for building footprints. Line intersections are easy in planar systems and I am aware that I normally cannot apply the same formulas for long lat.

However, I am wondering if it is OK in my situation since the spatial extent of my data is very small (buildings). I also know that I could just project to e.g. UTM and then calculate in the planar system, but I would like to work directly with the WGS84 coordinates.

whyzar
  • 12,053
  • 23
  • 38
  • 72
Chris
  • 291
  • 1
  • 10
  • 1
    You can work with wgs84 without any problem. Which GIS software do you use ? – BBG_GIS Nov 24 '16 at 18:54
  • @wetland I do not use a GIS in this case, but rather implement it myself in my software. That's the kind of calculation I was thinking about: --- stackoverflow.com/questions/385305/… --- x and y would correspond to lon and lat. – Chris Nov 25 '16 at 07:32
  • @Wetland There are huge problems associated with treating lat-lon coordinates as if they were Euclidean: away from the Equator, distances will vary with direction and most angle calculations will be wrong. Whether these affect finding line intersections depends on how the lines are represented. For instance, if they are given in terms of starting points and bearings, the answers will be incorrect. Using simple adjustments, as described at http://gis.stackexchange.com/a/30037, Euclidean calculations will be fine for building footprints. – whuber Nov 25 '16 at 21:02

0 Answers0