Geoscience Reference
In-Depth Information
provided a sufficient amount of pulse energy can penetrate through the canopy. In this
case, the first return can be an indication of tree height, while the last return typically
indicates the ground surface. Intermediate pulse returns can then be obtained from
the understorey or inner canopy. In Fig.
15.1
only two discrete returns are recorded.
There is a growing tendency for LiDAR instruments to record an increasing
amount of information about the laser pulses. This characteristic of point cloud
data makes them different and more substantial than simple vector GIS points. In
addition to the traditional point attributes such as location and return number, instru-
ments often record the intensity and scan angle of the laser return pulse. Intensity
values relate to the energy of the return pulse, and the scan angle relates the angle at
which the laser fired its pulse relative to nadir. Modern LiDAR sensors can record
the entire waveform for each returning laser pulse, hence their name
full waveform
LiDAR. Instead of computing discrete returns, they record the variation in returned
pulse intensity over time. Notice that the analog waveform is still quantized before it
is stored in digital format (see Fig.
15.1
). This latest technology introduces interest-
ing opportunities for vegetation applications, including the characterization of tree
structure and biomass estimation. However, full waveformdata greatly increases data
storage and processing time. Moreover, little commercial software currently support
full waveform LiDAR data processing. Consequently, full waveform data are mostly
used in a research context today.
For visualizing LiDAR point clouds you can rasterize the point cloud first to a
typical gridded format (see Sect.
15.4
) and use a conventional viewer (e.g. QGIS).
There are also a number of viewers available that can visualize the point cloud
directly in its native format. For instance, LAStools (see Sect.
15.3.3
) provides a
viewer
lasview
for Windows platforms, but it is not open source. An interesting
new open source initiative from Howard Butler and Uday Verma is Plasio.
2
It is
implemented in a modern web browser using HTML5 technologies such as WebGL
and it is shown in Fig.
15.2
.
15.2 LiDAR Data Formats and APIs
There are several data formats in which LiDAR point clouds can be stored. ASCII
(text) format is commonly used, but the binary LAS format,
3
issued by the American
Society for Photogrammetry and Remote Sensing (ASPRS), is widely used in the
industry today. Binary formats are more efficient for computers to access, which is
important due to the large volume of data that is typically associated with LiDAR
datasets. The latest specification of LAS is version 1.4 (Graham 2012) and has been
approved by ASPRS on November 14, 2011.
4
Search WWH ::
Custom Search