IEEE Geoscience and Remote Sensing Magazine - March 2022 - 6

Characteristics of the principal MTInSAR algorithms are reviewed
and then demonstrated using data collected around
Hong Kong International Airport. This demonstration
validates the performance of MTInSAR in terms of point
density, height accuracy, and deformation accuracy. The
authors provide an illustration of the typical dynamic infrastructure
behavior that can be observed with MTInSAR
and address some of the key issues, including the availability
of high-quality images, connecting InSAR point clouds
to actual structures, and the retrieval of deformation from
the 3D point clouds. Future directions include developing
operational MTInSAR, the use of artificial intelligence, and
multidisciplinary modeling coupling structural models
and geophysics with MTInSAR.
Our next applications-themed article is by Ho et al.,
covering radiometers for soil moisture measurement.
Active and passive microwave techniques are less sensitive
to solar illumination, vegetation canopy, or weather
than optical methods are. Therefore, they are generally
preferred for the remote sensing of soil moisture. Vegetation
growth is strongly dependent on soil moisture,
so the accurate measurement of this variable is essential
for monitoring water shortages in agriculture, for
example. While the 10km resolution of satellite-based
passive techniques (radiometers) cannot approach the
100-m resolution of active techniques (radar), they do
generally provide a more accurate measurement of the
soil moisture. The brightness temperature of naturally
emitted signals is easier to interpret and has a more
direct connection to the dielectric constant and, thus,
soil moisture.
The last decade saw the launch of microwave radiometers
on several
large satellite missions to provide
global soil moisture retrieval. Looking ahead, there is a
strong need to develop smaller and lower-cost radiometers
for the next generation of more cost-constrained
satellite missions.
This article reviews the basic topologies of microwave
radiometers (total power, Dicke, noise injection, and correlation)
with an eye toward future trends to reduce instrument
size and cost. Antennas play a vital role in radiometer
performance, and various approaches to antenna design
and electronic beam scanning are described. The authors
conclude that future research on low-cost, printable planar
antennas and antenna arrays with electronic beam steering
coupled with a digital radiometer are promising future
research directions to realize a significant reduction in system
cost and size.
Catapano et al. review contactless ground-penetrating
radar (GPR), which utilizes the capability of an electromagnetic
wave to penetrate nonmetallic media, such as
soil or construction materials. The backscattered field is
then processed to detect any underground or embedded
target reflecting the wave. GPR is used in many applications,
ranging from nondestructive inspection, soil-stratigraphy
characterization, mining, archeology, structural in6
spection,
and forensic surveys. GPR can be used on ground
and aerial vehicles. It has even been demonstrated on satellites
orbiting Mars. New challenges are presented, however,
when the radar is not directly coupled to the ground
(i.e., " contactless " ).
Some of the techniques used to address these challenges
include microwave tomography. In this article, standard
processing methods and imaging focusing algorithms are
described, along with full-waveform inversion methods
and machine learning techniques. Specific recent examples
are presented in the latter case. Practical issues, such as clutter,
multipath, and the need for large-scale data processing,
are discussed. The article concludes with an outline of future
research needs in contactless GPR, including the extension
to full 3D imaging and nonredundant data collections
strategies. Fully automated, real-time onboard data processing
remains a highly desirable goal, motivating some of
these research efforts.
The range of practical applications of remote sensing is
quite vast, and the article by Li et al. shows how it can be
used to monitor the power grid. They review the use of the
airborne transient electromagnetic techniques for measuring
ground return currents in high-voltage dc transmission
systems and demonstrate this process using a simulation.
Our last two articles cover applications to the biosphere.
The desert locust (Schistocerca gregaria) is the world's most
destructive migratory pest, and satellite remote sensing is
an important tool for monitoring its potential habitat, as
explained in the article by Rhodes and Sagan. Recognizing
the lack of collaboration between remote sensing experts
and ecological modelers as well as the need for both communities
to understand the assumptions and limitations of
each other's models and techniques, the authors prepared
a review covering the combined use of remote sensing with
machine learning in this application.
They examine current satellite remote sensing approaches
for collecting environmental information indicating a suitable
habitat and explore the sensitivity of this relationship to
the scale and quality of the environmental information. DL
methods promise to better represent this relationship, and
the authors outline a blueprint for their application to general
habitat prediction, not limited to locusts. Six algorithms
are evaluated in two study regions (Niger and Sudan), with
resulting accuracy scores of 88% and higher.
Pests and diseases also harm the forestry sector. Eugenio
et al. explore the use of remotely piloted aircraft systems for
monitoring forest health. Using bibliometric methods, they
identify the main topics and trends in this research, looking
at the forest species, pests, and diseases being studied
as well as the platforms, sensors, and flight parameters utilized.
Early recognition of patterns indicating forest damage
is a challenge, and machine learning is expected to become
more widespread in this application. Another challenge is
the dissemination of new techniques to rural communities.
(continued on p. 9)
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE MARCH 2022
IEEE Geoscience and Remote Sensing Magazine - March 2022

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - March 2022
