road_kernel: Kernel density estimation on roads
In bmsasilva/roadHotspots: Interactive shiny app to visualize road hotspots

Description Usage Arguments Details Value Author(s) References Examples

View source: R/road_kernel.R

Kernel density estimation on roads using the function kde2d from package MASS.

1	road_kernel (count_path, roads_path, bandw = 500, group = "all")

`count_path`	Path for the .csv file with the location of the observations. The file must have three headed columns (ID, x coordinate and y coordinate) with one observation per row. The coordinates must have the same projection as the roads shapefile. The header names are indiferent but the columns position must be as described.
`roads_path`	Path for the .shp file with the roads
`bandw`	Vector of bandwidths for x and y directions. A scalar value will be taken to apply to both directions.
`group`	String used to subset species/groups from the observations dataframe. Can be a single string or multiple strings. For multiple strings use: c("spe1", "spe2"). By default all observations are used.

Observations are snaped to closest roads prior to estimation and the resulting kernel is clipped to the road. The kernel density estimation is rescaled to [0, 1] and recoded in three categories: 1 = 0.25 to 0.50, 2 = 0.50 to 0.75, 3 = 0.75 to 1

A SpatialPolygonsDataFrame object with the kernel density estimation

Bruno Silva

Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth edition. Springer.

1
2
3

roads_path <- system.file("extdata/roads.shp", package = "roadHotspots")
count_path <- system.file("extdata/count.csv", package = "roadHotspots")
output <- road_kernel(count_path, roads_path)