14
MULTISCALE APPROACH FOR
GROUND FILTERING FROM LIDAR
ALTIMETRY MEASUREMENTS
JOSÉ L. SILVAN-C
ARDENAS AND LE WANG
14.1 INTRODUCTION
The increase in spatial resolution of sensors has enabled human interpreters to see
more details of the landscape and, at the same time, has challenged remote sensing
scientists to develop fully automated or semiautomated algorithms for efficiently
extracting desired features from such massive data sets. Examples of desired features
are digital terrain models (DTMs), which constitute a basic source of information to
many applications, such as watershed analysis, hydrological modeling, river dynamics studies, coastal erosion estimation, environmental resource evaluation, and urban
growth projections, to list just a few.
The small-footprint, discrete-return light detection and ranging (Lidar) system has
become one of the most important means to produce high-resolution DTM data. This
has been due in part to a number of advantages over competing aerial photogrammetric techniques such as independence of sunlight, higher vertical accuracy, less
missing data by occlusion, low redundancy and because it does not rely on the
existence of textured surfaces and discontinuities for a successful point matching
(Pfeifer and Mandlburger, 2009). The small-footprint, discrete-return Lidar system is
based on the accurate measurement of the elapsed time between emitted and backscattered laser pulses. The emitted pulse is typically short time and unimodal, whereas
the backscatter may spread over longer times exhibiting multiple modes called
returns. Returns are associated with distinct layers with which the laser interacted.
267
Scale Issues in Remote Sensing, First Edition. Edited by Qihao Weng.
2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.
MULTISCALE APPROACH FOR
GROUND FILTERING FROM LIDAR
ALTIMETRY MEASUREMENTS
JOSÉ L. SILVAN-C
ARDENAS AND LE WANG
14.1 INTRODUCTION
The increase in spatial resolution of sensors has enabled human interpreters to see
more details of the landscape and, at the same time, has challenged remote sensing
scientists to develop fully automated or semiautomated algorithms for efficiently
extracting desired features from such massive data sets. Examples of desired features
are digital terrain models (DTMs), which constitute a basic source of information to
many applications, such as watershed analysis, hydrological modeling, river dynamics studies, coastal erosion estimation, environmental resource evaluation, and urban
growth projections, to list just a few.
The small-footprint, discrete-return light detection and ranging (Lidar) system has
become one of the most important means to produce high-resolution DTM data. This
has been due in part to a number of advantages over competing aerial photogrammetric techniques such as independence of sunlight, higher vertical accuracy, less
missing data by occlusion, low redundancy and because it does not rely on the
existence of textured surfaces and discontinuities for a successful point matching
(Pfeifer and Mandlburger, 2009). The small-footprint, discrete-return Lidar system is
based on the accurate measurement of the elapsed time between emitted and backscattered laser pulses. The emitted pulse is typically short time and unimodal, whereas
the backscatter may spread over longer times exhibiting multiple modes called
returns. Returns are associated with distinct layers with which the laser interacted.
267
Scale Issues in Remote Sensing, First Edition. Edited by Qihao Weng.
2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.