Analysis Metrics
Analysis Metrics holds the tools that tell you what is in a point cloud and measure things from it, rather than modifying it. Start here with any new file, and come back here for the analyses that extract objects: individual trees, roof planes, locally planar surfaces.
Start here
Section titled “Start here”Lidar Info
Section titled “Lidar Info”The first thing to run on any file. It produces a report listing the number of points, the extent, the range of intensities, which classification classes are present and how the returns are distributed.
Two questions it answers immediately. Is this cloud classified? If class 2, ground, is absent, you will have to classify it before doing anything useful. How dense is it? That decides the resolution of every raster you can sensibly derive.
Lidar Point Density
Section titled “Lidar Point Density”Maps how many points per unit area the survey actually delivered, cell by cell. Very useful before gridding, because it shows the flight lines, the gaps and the overlaps.
The practical rule for gridding: your raster cell should hold several points. A cloud at 4 points per square metre supports a 50 centimetre raster comfortably, and a 10 centimetre raster only by inventing detail.
Individual Tree Detection
Section titled “Individual Tree Detection”Finds tree tops, as local maxima of the canopy with an adaptive search radius, and returns a point layer of probable stem positions. The base of any forest inventory built from LiDAR.
Run it on a normalised cloud, where heights are relative to the ground. Applied to raw elevations, it will find the tops of the hills rather than the tops of the trees. See Normalize Lidar in Filtering & Classification.
Lidar Rooftop Analysis
Section titled “Lidar Rooftop Analysis”Detects planar roof segments inside building footprints you supply, and returns them as polygons with roof attributes such as orientation and slope. This is the analysis behind solar potential studies at city scale.
All the tools
Section titled “All the tools”| Tool | What it does |
|---|---|
| Lidar Info | Metadata report: point count, extent, intensity range, class histogram, return distribution. |
| Lidar Histogram | Frequency distribution of an attribute such as elevation, intensity, scan angle or class, as an HTML chart. |
| Lidar Point Density | Points per unit area, as a raster. |
| Lidar Point Stats | Per cell statistics: point count, pulse count, points per pulse, elevation and intensity range, predominant class. |
| Lidar Point Return Analysis | Audits the consistency of the return sequence and flags anomalies. |
| Lidar Kappa | Agreement report between two classified clouds, with a class agreement raster. |
| Individual Tree Detection | Finds tree tops as local canopy maxima, returning a point layer. |
| Lidar Rooftop Analysis | Finds planar roof segments within building footprints, with roof attributes. |
| Lidar Ransac Planes | Identifies locally planar points by fitting planes in a neighbourhood. |
| Lidar Eigenvalue Features | Derives shape descriptors, such as planarity and linearity, from the local geometry around each point. |
| Normal Vectors | Computes a surface normal per point and stores it, with an RGB rendering. |
| Colourize Based On Class | Colours points by classification, using the standard ASPRS colours. |
| Colourize Based On Point Returns | Colours points by return order, showing the multiple return structure. |
| Find Flightline Edge Points | Extracts the points at the edges of acquisition swaths, for quality control. |
Reports are downloads. Lidar Info, Lidar Histogram and the audit tools produce text or HTML documents, which are downloaded rather than added to the map.
Colourising helps you see. The two Colourize tools write colours into the point records themselves, which makes a cloud far easier to read in any viewer that renders point colours.