Forest and canopy
Everything a forest inventory asks of a point cloud sits on top of one answer: where the ground is. Get that wrong on a slope and every height above it is wrong by the same amount.
The ground comes first
Under closed canopy, few returns reach the ground, and on a slope the returns that do are easily mistaken for low vegetation. A filter tuned for a flat corridor shaves the hillside; a filter tuned for a hillside leaves undergrowth in the terrain. The run measures the site's slope and canopy cover and chooses, and the report states the choice rather than hiding it.
Measured on a five-point-eight-million-point parcel on a twenty-four per cent forested slope: sixteen seconds from raw, unlabelled points.
Then the canopy
- Canopy height, as a raster of the height between the ground and the surface.
- Cover and vertical strata on a grid, at the cell size you ask for.
- Structure indices, for stands you are comparing to each other or to a previous flight.
- Per-tree segmentation where the crowns separate, with a height and a position each.
Between two flights
Growth, harvest and storm damage are all the same question asked of two epochs: what moved, by how much, and where. That is a run of its own, and it writes the difference as a surface and as geometry rather than as a picture somebody has to interpret.
What you get
Deliverables go out as LAZ, GeoTIFF, GeoJSON, DXF, shapefile with a .prj, E57, PLY, CSV, OBJ, glTF, LandXML, COPC and 3D Tiles.
Questions we are actually asked
Does it find the ground under closed canopy?
That is the whole difficulty, and it is why the ground filter is chosen from the site rather than fixed. On a steep forested parcel a cloth simulation follows the terrain where a morphological filter cuts the hillside off; the run picks from the slope and the canopy cover it measures, and the report says which it picked and why.
How did that do on a real forested slope?
Measured on a five-point-eight-million-point parcel on a twenty-four per cent forested slope: sixteen seconds from raw, unlabelled points.
What canopy metrics does it produce?
Height percentiles, canopy cover, vertical strata, a rumple index and a leaf-area proxy, on a grid at the cell size you ask for, plus per-tree segmentation where the canopy separates.
Can I compare two flights of the same stand?
Yes. Change detection between two epochs is a run of its own: it aligns the pair, reports where the surface moved and by how much, and writes the difference as a raster and as geometry.
What if my cloud has no vertical datum?
Horizontal and vertical EPSG codes are read, stamped and reprojected as a first-class record rather than guessed at.
Does it need a licence for a desktop package?
The engine is pure Rust and needs no PDAL or PROJ installation to run its native commands.