PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
March 2014
201
ASPRS Research on Quantifying the
Geometric Quality of Lidar Data
Aparajithan Sampath
1
, Hans K. Heidemann
2
, Gregory L. Stensaas
3
, Jon B. Christopherson
4
Introduction and Background
Lidar data are well on their way to becoming as important
as photogrammetric imagery to geospatial analysis. How-
ever, the standards of Quality Assurance and Control (QA
and QC) that transform photogrammetric imagery from a
mere photograph to a metric tool are not as developed for
lidar data. The current lidar data quality assessment meth-
ods are not adequate in the reporting of a) the quality of
calibration of lidar system, which is an essential indicator
of the overall quality of data, and b) the horizontal accuracy
of the data. Recognizing this, the USGS has partnered with
the ASPRS Lidar Division and the Airborne Lidar Com-
mittee to form a Calibration/Validation (Cal/Val) Working
Group with a goal to promote industry-accepted guidelines
and tools to help assess the quality of lidar data. The Cal/
Val Working Group consists of representatives from the
Government (e.g. USGS, National Geodetic Survey, Nation-
al Geospatial Intelligence Agency, etc.), the industry (lidar
instrument manufacturers, data providers/vendors, software
developers) and academia. This paper discusses the Cal/Val
Working Group’s research efforts and their current status.
1
Calibration/Validation Engineer, Stinger Ghaffarian
Technologies (SGT), Contractor to US Geological Survey
2
Physical Scientist, Lidar Science, US Geological Survey
3
Project Manager, Remote Sensing Technologies and National
Land Imaging Requirements, US Geological Survey
4
Contract Task Manager Remote Sensing Technologies
Project, Stinger Ghaffarian Technologies (SGT), Contractor
to U.S. Geological Survey
U.S. Geological Survey, Earth Resources Observation and
Science (EROS) Center, Mundt Federal Building, 47914 252
nd
Street, Sioux Falls, SD 57198
The Working Group is synthesizing these efforts into a draft
best practices/guidelines document (titled Guidelines on
Geometric Accuracy and Quality of Lidar Data) for QA/QC
processes. These efforts broadly fall under the following steps:
• Defining procedures for measuring the inter-swath good-
ness of fit. These procedures include defining three Data
Quality Measures (DQMs)
• Suggesting the use of targets and Ground Control Points
(GCPs) on natural surfaces of all slopes to measure the
absolute accuracy
• Suggesting the use of sensor model based rigorous lidar
system calibration methods.
These steps are encapsulated in the framework shown
in Figure 1. The framework is designed such that the pro-
cesses for measuring the accuracy (both inter-swath and
absolute) of lidar data are independent of the data acqui-
sition process and the sensor model of the instrument.
Among the three steps, most of the efforts of the Cal/
Val Working Group have focused on defining methodologies
and algorithms to measure the inter-swath goodness of fit.
Existing lidar data specifications of many organizations
specify requirements for inter-swath accuracy (Heidemann
2012), or ask for calibration reports or generic calibration
parameters (NGS 2009; NGA 2012). However, these spec-
ifications do not detail how the testing should be done,
what measurements are acceptable, etc. Therefore, there
is no widely accepted standardized process of testing the
inter-swath goodness of fit or accuracy of lidar data. It is
expected that the ASPRS approved document will create a
standard methodology of testing and quantifying the qual-
ity of lidar data, which will improve the interoperability
of data from multiple sources, generating increased confi-
dence in the data and increasing its scientific applications.
The American Society for Photogrammetry and Remote Sensing’s
Lidar Cal/Val (calibration/validation) Working Group led
by the US Geological Survey (USGS) to establish “Guidelines
on Geometric Accuracy and Quality of Lidar Data” has made
excellent progress via regular teleconferences and meetings.
The group is focused on identifying data quality metrics and
establishing a set of guidelines for quantifying the quality of lidar
data. The working group has defined and agreed on lidar Data
Quality Measures (DQMs) to be used for this purpose. The DQMs
are envisaged as the first ever consistent way of checking lidar
data. It is expected that these metrics will be used as standard
methods for quantifying the geometric quality of lidar data.
The goal of this article is to communicate these developments to
readers and the larger geospatial community and invite them to
participate in the process.
