visibility: Computes the visibility from a single location in a 3D point...
In viewshed3d: Compute Viewshed in 3D Point Clouds of Ecosystems

Description Usage Arguments Details Value Note References Examples

Computes visibility from a user-defined location with user-defined sightline angles and returns the visibility as function of distance and, optionally, the 3D point cloud classified as visible and non-visible points.

1	visibility(data, position, angular_res, scene_radius, store_points)

`data`	LAS class object containing the xyz coordinates of a 3D point cloud
`position`	vector of length 3 containing the xyz coordinates of the animal location. Default = c(0,0,0).
`angular_res`	numeric. The angular resolution of a single sightline. Default = 1.
`scene_radius`	(optional) numeric. Defines the radius of the scene relative to the animal position. Can be used to apply a cut-off distance to visibility analyses.
`store_points`	logical. If `TRUE`, the 3D point cloud is returned with visible and not visible points classified (see details).

Sightline directions are computed from the method described by Malkin (2016). This ensures that every sightline explores a similar portion of the 3d space.

If store_points = FALSE, a list containing the data.table of the visibility (Visibility) as a function of distance to the animal location (r) is returned and the viewshed coefficient (the area under the curve of visibility as function of distance). If store_points = TRUE, a LAS containing the point cloud (X, Y, Z), the distance of each point to the animal position (r) and the class of each point: visible or not visible from the animal position (Visibility = 2 or 1, respectively) is added to the output.

In most cases, a ground reconstruction should be performed before visibility computation to avoid sightlines passing through the ground. This can be done with the reconstruct_ground function.

Malkin, Z. (2016). A new method to subdivide a spherical surface into equal-area cells. arXiv:1612.03467.

#- import the tree_line_plot dataset
file <- system.file("extdata", "tree_line_plot.laz", package="viewshed3d")
tls <- lidR::readLAS(file)

center <- c(0,0,2) # defines the scene center for the entire process
angle <- 1 # defines the angular resolution for the entire process

#- remove noise to avoid visibility estimates error
tls_clean <- viewshed3d::denoise_scene(tls,method="sd",
                                       filter=6)


#- class ground and vegetation points
class <- lidR::classify_ground(tls_clean, lidR::csf(rigidness = 1L,
                                              class_threshold = 0.2,
                                              sloop_smooth = FALSE))

#- reconstruct the ground with the optimal resolution
recons <- viewshed3d::reconstruct_ground(data=class,
                                         position = center,
                                         ground_res = 0.05,
                                         angular_res = angle,
                                         method="knnidw",
                                         full_raster = TRUE)

#- compute the visibility and store the output point cloud.
#- As the input file is a LAS object, the output
#- point cloud is also stored in a LAS file.
view.data <- viewshed3d::visibility(data = recons,
                                    position = center,
                                    angular_res = angle,
                                    scene_radius = 17, # apply cut_oof distance
                                    store_points = TRUE)

#- viewshed coefficient
view.data$viewshed_coeffecient

#- 3D plot with visible points in white and non-visible points in darkgrey
x=lidR::plot(view.data$points,color="Visibility",colorPalette = c("grey24","white"))

#- add animal position to the plot
position=data.frame(X=center[1],Y=center[2],Z=center[3])
lidR::add_treetops3d(x,sp::SpatialPointsDataFrame(position,position),
                     radius=0.2,col="red")

#- plot the visibility as function of distance
plot(view.data$visibility$r,view.data$visibility$visibility,
     type="l",ylim=c(0,100),lwd=4)