plot fails with lasmetrics3d objects and boolean attribute

[TedwardErker]

Thank you for creating this package. It's truly incredible.

I found a small issue with plotting lasmetrics3d objects. I ran the example from "grid_metrics3d", the part that is not run fails to plot the colors.

## Not run:

# Define your own metric function
myMetrics = function(i)
{
   ret = list(
      npoints = length(i),
      imean   = mean(i))
   
   return(ret)
}
     
voxels = grid_metrics3d(las, ~myMetrics(Intensity), 3)
     
plot(voxels, color = "imean")
#etc.
## End(Not run) 

When I run the code above I get no error, but I also don't get any color. When I run with my own function I get this error: Error in col2rgb(colors) : invalid color name 'tree'.

Not a big deal I think I'll be able to find a work around, but wanted to put it up there. Thanks again for this amazing tool. I've only just started using it and I'm excited to find all it can do.

[Jean-Romain]

Could you please make a fully reproducible example. It works for me. You don't have any color because the color palette is broken by outliers. You can use trim = 100. The following do works as expected. I will update the doc to add trim = 100. However I'm not able to reproduce the error you mentionned.

library(lidR)

LASfile <- system.file("extdata", "Megaplot.laz", package="lidR")
las = readLAS(LASfile)

myMetrics = function(i) {
  ret = list(
    npoints = length(i),
    imean   = mean(i)
  )
  
  return(ret)
}

voxels = grid_metrics3d(las, ~myMetrics(Intensity), 3)
plot(voxels, color = "imean", trim = 100)

[TedwardErker]

Thank you. Setting trim fixed my visualization problem. I tracked down my error and think it was from my function not returning numeric (it was boolean).

For posterity, here's the reproducible example:

library(lidR)
Modes <- function(x) {
  ux <- unique(x)
  tab <- tabulate(match(x, ux))
  return(list(mode = ux[tab == max(tab)]))
}

LASfile <- system.file("extdata", "Megaplot.laz", package="lidR")
las = readLAS(LASfile)

# works great
gmclass <- grid_metrics3d(las, ~Modes(Classification), 3)
plot(gmclass, color = "mode", trim = 3)


lp <- lasdetectshape(las, shp_plane(th1 = 15, th2 = 4, k = 20), "Coplanar")
plot(lp, color = "Coplanar")

gmcoplanar <- grid_metrics3d(lp, ~Modes(Coplanar), 3)
# doesn't work
plot(gmcoplanar, color = "mode")


Modes_numeric<- function(x) {
  ux <- unique(x)
  tab <- tabulate(match(x, ux))
  return(list(mode = as.numeric(ux[tab == max(tab)])))
}

gmcoplanar <- grid_metrics3d(lp, ~Modes_numeric(Coplanar), 3)
# works now
plot(gmcoplanar, color = "mode", trim = 3)

[Jean-Romain]

Ok this is a bug. Boolean data should be displayed. They are displayed for LAS objects.

[Jean-Romain]

TODO

• add trim = 100 in the example of the documentation of grid_metrics3d()
• fix plot.lasmetrics3d() with boolean

[TedwardErker]

Great, thanks for addressing this. Sorry for closing too soon. :)