Geoscience Reference
In-Depth Information
Chapter 15
3D Point Cloud Data Processing
In this chapter we will deal with three dimensional (3D) point cloud data. They are a
subset of point data, where each point contains an attribute with height information.
As for all point data, data can, but must not be stored in a regular grid. Vector point data
come in various formats, including those discussed in Chap.
2
and spatially enabled
relational database management systems (PostGIS, Spatialite). We concentrate on a
specific type of 3D point data, obtained with laser scanning instruments. The example
data are obtained from an airborne topographic instrument. A thorough discussion
on point cloud data is beyond the scope of this topic. The open source community
for point cloud data is not as well established and consolidated as it is for vector
and raster data. Standards are still under development, in particular with respect to
the relatively new full waveform (FWF) data. We provide a brief introduction on
LiDAR data and how to process them. This ranges from extracting basic information
to filtering techniques to derive different digital elevation models (DEM). A selection
of current open source utilities is also presented here, without entering into as much
detail as for vector and raster data. In addition, the toolbox offered by SPDLib is
15.1 Introduction to LiDAR Data
A light detection and ranging (LiDAR) instrument emits and receives laser pulses
while sweeping in different directions. The majority of topographic LiDAR sensors
operate in the near-infrared region of the electromagnetic spectrum, typically at
1.064nm. For bathymetric applications, lower wavelengths are also used (e.g. in
green) in order to better penetrate—relatively shallow and clear—water.
Instruments with pulse rates more than 150kHz are common nowadays. This
can result in a cloud of tens to hundreds of millions of points, depending on the
spatial extent of the acquisition. A laser pulse can be characterized as a beam of
Search WWH ::
Custom Search