Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It only takes a minute to sign up.
Sign up to join this community
I am using UAV acquired imagery to identify trees at a peat land in the UK. I have a DEM and an Orthomosaic which i have created which can bee seen here (note: the Orthomosaic is at a much higher resolution when viewed in ArcMap than online).
Manually i can identify trees by looking at the high points on the DEM and by visually spotting them on the Orthomosaic and placing a point on them.
How would i start going about getting ArcMap to do this automatically for me?
Lots of ways to try to get at this:
Topographic Position Index http://www.jennessent.com/downloads/tpi-poster-tnc_18x22.pdf
Elevation residuals (DEM minus a smoothed version of DEM)
Local maxima (Focal Statistics with "Maximum".
All of these will give a resulting raster. Threshold the output raster appropriately (should be fairly standard on peat lands), then convert those raster zones of interest to points or generate centroids of those zones. All of it can be automated in Model Builder or Python.
In ArcMap, assuming you have access to the Spatial Analyst extension, use the Raster Calculator tool and enter: "DEM" - FocalStatistics("DEM", NbrCircle(5, "MAP"), "MEAN", "DATA"). Where "DEM" is your elevation raster and 5 in this case being a 5 m radius on which to generate mean values. This equation will subtract the mean raster value of a 5m radius around each cell, from the actual value of that cell. This gives a relative slope position for each cell where negative values are lower than the surrounding average and positive are higher than the surrounding average. Tweak the radius to generalize at different scales.
