lidR (version 4.1.2)

its_watershed: Individual Tree Segmentation Algorithm

Description

This function is made to be used in segment_trees. It implements an algorithm for tree segmentation based on a watershed. It is based on the bioconductor package EBIimage. You need to install this package to run this method (see its github page). Internally, the function EBImage::watershed is called.

Usage

watershed(chm, th_tree = 2, tol = 1, ext = 1)

Arguments

chm

`RasterLayer`, `SpatRaster` or `stars`. Canopy height model. Can be computed with rasterize_canopy or read from an external file.

th_tree

numeric. Threshold below which a pixel cannot be a tree. Default is 2.

tol

numeric. Tolerance see ?EBImage::watershed.

ext

numeric. see ?EBImage::watershed.

Details

Because this algorithm works on a CHM only there is no actual need for a point cloud. Sometimes the user does not even have the point cloud that generated the CHM. lidR is a point cloud-oriented library, which is why this algorithm must be used in segment_trees to merge the result into the point cloud. However, the user can use this as a stand-alone function like this:

 chm <- raster("chm.tif")
 crowns <- watershed(chm)()

See Also

Other individual tree segmentation algorithms: its_dalponte2016, its_li2012, its_silva2016

Other raster based tree segmentation algorithms: its_dalponte2016, its_silva2016

Examples

if (FALSE) {
LASfile <- system.file("extdata", "MixedConifer.laz", package="lidR")
poi <- "-drop_z_below 0 -inside 481280 3812940 481320 3812980"
las <- readLAS(LASfile, select = "xyz", filter = poi)
col <- pastel.colors(250)

# Using raster because focal does not exist in stars
chm <- rasterize_canopy(las, res = 0.5, p2r(0.3), pkg = "raster")
ker <- matrix(1,3,3)
chm <- raster::focal(chm, w = ker, fun = mean, na.rm = TRUE)
las <- segment_trees(las, watershed(chm))

plot(las, color = "treeID", colorPalette = col)
}