Skip to main content

Crate sylva_rs

Crate sylva_rs 

Source
Expand description

sylva-rs: LiDAR processing for forest ecology and remote sensing research.

Copyright (C) 2026 Tim Devereux, The University of Queensland. Free software under the GNU General Public License v3.0 or later; see the LICENSE file. There is no warranty, to the extent permitted by law.

The crate is organised around two data models:

  • PointCloud — an (N, 3) set of coordinates plus named per-point attributes, the unit of exchange with LAS/LAZ/PLY/text files.
  • Shots — pulse-centric data (per-pulse origin and direction, with a CSR list of echoes), which is what ray-based canopy metrics need.

Algorithms are grouped by topic: filters, cluster, ground, canopy, voxel, trees, registration and qsm.

Re-exports§

pub use error::Error;
pub use error::Result;
pub use pointcloud::Attr;
pub use pointcloud::PointCloud;
pub use raster::Raster;
pub use shots::Shots;
pub use transform::Transform;

Modules§

als
Large-area airborne lidar: a catalogue of LAS/LAZ tiles and processing in buffered chunks.
canopy
Canopy structure: voxels, plant area density, gap fraction, LAI.
change
Change detection between two epochs of one plot.
cluster
Neighbourhood graphs, connected components and shortest paths.
coreg
Coregistration of scans: stem-map matching, reflectors, ICP, pose graphs and the survey pipeline.
error
What can go wrong, and the message it carries.
filters
Subsampling, cropping and noise filtering.
fusion
Fusion of terrestrial (TLS) and airborne (ALS) lidar over the same forest.
geo
Geospatial support: coordinate systems and transforms, interpolation and masking.
ground
Ground classification, DTM, height normalisation and CHM.
io
Point cloud readers and writers, dispatched on file extension.
leaves
Foliage for a structure model: leaf angle distribution, leaf area from leaf points, and leaf meshes placed on a QSM.
pointcloud
Core point cloud container.
qsm
Quantitative structure models: skeletonisation and cylinder fitting.
quality
Point-cloud quality from stems.
raster
Georeferenced 2-D grid used for DTMs and CHMs.
registration
Rigid registration of scan positions.
riscan
RiSCAN PRO projects and RiSCAN PRO’s filters.
shots
Pulse-centric data: per-shot origin and direction with a CSR echo list.
synthetic
Small synthetic scenes for examples, tutorials and tests.
transform
Rigid / affine 4x4 transforms.
trees
Tree-level processing: stem detection, DBH, segmentation, heights, crowns.
util
Internal plumbing shared across domains: numerics, numpy-compatible RNG, JSON, progress and limits.
voxel
Ray-traced voxel grid with Beer–Lambert statistics (Vincent et al. 2017).
waveform
Full-waveform lidar: a container of digitised waveforms, readers and writers for LAS 1.3/1.4 waveform data packets (las) and PulseWaves (pulsewaves), Gaussian decomposition (decompose) and a synthetic waveform generator (simulate).

Type Aliases§

Point
A 3-D point.