Ground Classification
Definition
The algorithmic process (e.g., using LAStools or PDAL) that separates ground-surface returns from non-ground returns in a LiDAR point cloud, enabling bare-earth DTM generation.
- Input
- LiDAR point cloud
- Output
- Separated ground vs non-ground returns
- Common software
- LAStools, PDAL
- Downstream product
- Bare-earth digital terrain model (DTM)
Common questions
Why is ground classification necessary before producing a usable terrain model?+
A raw LiDAR point cloud mixes returns from the actual ground with returns bouncing off vegetation, vehicles, and structures, so without separating those out first, the resulting elevation surface would show canopy and building heights instead of the true bare-earth topography needed for scene reconstruction or search prediction.
What can go wrong if ground classification is applied with poor settings in a forensic context?+
Overly aggressive filtering can misclassify a low mound, a shallow grave depression, or a small terrain anomaly as noise and remove it, while under-filtering can leave vegetation returns in the ground surface, either error can distort the bare-earth model an investigator relies on to spot subtle surface anomalies.
Related terms
- Bare-Earth DTM
- A digital terrain model produced by filtering LiDAR returns to ground-level pulses only, removing vegetation and structure returns. The resulting surface represents...
- LiDAR
- Light Detection and Ranging: an active remote-sensing method that emits laser pulses and records the time-of-flight of returning echoes to produce a...
- Micro-Topography
- Small-scale (centimetre to metre) variations in ground surface height. Grave mounding, subsidence hollows, and soil-scrape scars are micro-topographic features typically invisible to...
- Point Cloud
- The raw data product of a LiDAR survey: a set of 3D coordinates (X, Y, Z) for each laser return, classified by...
- SfM Point Cloud
- A 3D point cloud derived from overlapping photographs using Structure from Motion photogrammetry. Achieves similar resolution to terrestrial LiDAR at close range...