infiltration and runoff with hydrological multitemporal clas-
sification. Part of the data were collected around solar noon
with a commercial visual
RGB
camera onboard an
UAV
.
Aquatic weed surveillance was described in Göktogan
et
al.
(2010). The
UAV
is a rotary-wing vehicle with length of 2
m, maximum
TOW
of 15 kg, and a fuel engine equipped with
a 3-CCD video camera. The images were captured and stored
for subsequent treatment and analysis.
Geipel
et al.
(2013) proposed a software-based framework
for connecting sensors and processor(s) onboard during the
flight; this is intended with the aim of exploiting this poten-
tial against platforms that only store information for follow-up
ground processing. They deployed a prototype for microcli-
mate monitoring equipped with low-cost sensors, includ-
ing temperature, humidity, and imaging sensors specifically
adapted or designed for agriculture applications.
Disaster Monitoring
Regarding disaster monitoring it is worth to consider two
relevant aspects, the first related to prevention and the second
to response after the event has occurred. According to this
category, pre- and post-emergency topics are to be considered.
Several years ago, Bendea
et al.
(2008) identified the need
and usefulness of
UAVs
for operations in disaster areas conve-
niently equipped with advanced sensors and technologies. Its
use in National Parks for surveillance was also considered as
convenient in prevention tasks (Restas, 2006).
Hurricanes, typhoons, tornados, earthquakes, fires, nuclear
incidents, spills in the ocean, floods, and avalanches are clear
causes of disaster where
UAVs
can play an important role.
These are manmade or natural events where
UAVs
have been
applied, but they are not exclusive and can serve as founda-
tions for future applications in disasters. Deployment of
UAVs
to build a network of sensors has been considered useful in
disaster management applications (Quaritsch
et al.
, 2010).
Hurricanes, Typhoons, and Tornados
UAVs
in the Hurricane and Severe Storm Sentinel (HS3) pro-
gram launched by
NASA
(2015b), were equipped with different
high-tech instruments to monitor hurricane formation and its
evolution. A multisensory technology was used including a
radar scanner and wind lidar, both based on the Doppler Effect,
multi-frequency radiometer based on interferometry, and a
microwave sounder. One
UAV
“catches” data inside the storm
(winds and precipitation) and a second one “explores” the en-
vironment. A hurricane post-disaster assessment was conduct-
ed in Adams
et al.
(2010) based on imagery analysis captured
by a
RGB
camera with a weight of 250 g onboard a helicopter.
Chou
et al.
(2010) applied imagery technologies with a com-
mercial camera with weight of 1.2 kg onboard a helicopter with
8.5 kg of weight and payload of 5.5 kg. This system was used
to analyze changes take for disaster monitoring following the
MORAKOT typhoon. Rescue operations were studied in DeBusk
(2010) for follow-up analysis caused by the tornado Alley.
Earthquakes
Seismic hazards were evaluated in the old city center of
L’Aquila, Italy after an earthquake with an octo-copter (Domi-
nici
et al.
, 2012) capturing high quality images on roofs and
facades of structures with a reflex digital camera. Baiocchi
et
al.
(2013) used stereoscopic techniques for 3
D
reconstruction
of buildings for determining possible damage in such build-
ings in the same city for post seismic analysis. The platform
is equipped with two cameras operating in the visible and
infrared spectrum, respectively, and a
GPS
.
After the Tohoku, Japan earthquake in 2011, important
efforts were made for the
NEC
(Wada
et al.
, 2015), where
UAVs
,
equipped with optical sensors, provided a rapid interven-
tion with abundant information during several imaging-based
surveillance tasks. After subsequent developments and
improvements, the different technologies were integrated in
the
UAV
, including: communication, control, sensing, image
processing, and networking. Thanks to all these technologies,
the
UAVs
can provide communication links where terrestrial
areas are damaged.
Li
et al.
(2011) proposed a method for image rectification
and mosaicking without control points in earthquake disaster
areas with the aim of early intervention. A
CCD
imaging sensor
was installed onboard a
UAV
with payload of 4.5 kg.
Fire
The use of
UAVs
, equipped with sensory technologies, was
early identified for its potential use in fire detection (Am-
brosia
et al.
, 2003, 2009, and 2011; Ambrosia and Wegener,
2009; Ollero
et al.
2006; Rufino and Moccia, 2005; Wu and
Zhou, 2006a and 2006b; Wu
et al.
, 2007).
UAVs
progressed to
more sophisticated and precise technologies and methods
with high performance, including the design of an effective
architecture (Pastor
et al.
, 2011). During crisis management in
fires, it is essential the coordination between fire brigades is
critical for extinguishing the fire. Persie
et al.
(2011) proposed
an integrated
GIS
where all relevant geospatial information is
automatically distributed to all levels of the organization.
During years 2006 to 2010,
NASA
and US Forest Service
conducted several missions over different forest where the
NASA
Ikhana
UAS
capabilities were demonstrated and verified
for processing multi-spectral data onboard the
UAV
(Ambrosia
et al.
, 2011), including fire monitoring. The great possibilities
for using thermal imaging cameras in firefighting were defined
in Hinkley and Zajkowski (2011) thanks to the collabora-
tion of the two institutions mentioned above. Forest wildfire
monitoring was also addressed in Zhou and Cheng (2005).
Martínez-de-Dios
et al.
(2007) used a fleet of three hetero-
geneous
UAVs
(Helivision-
GRVC
, Marvin and Karma) for fire
detection. The fire is detected by means of histogram analysis
using learning-based strategies. These
UAVs
explore different
areas in cooperative surveillance; if a fire is detected, they
provide positioning of this incident using to the system’s
GPS
. Maza
et al.
(2011) proposed a distributed architecture of
multi-
UAVs
with this identical purpose, where fire detection
tasks have been carried out and tested under the proposed
approach. Figure 8 displays a fire monitoring sequence at
different levels of detail, where the polygons surrounding the
active flames identifies its position and extension. In the se-
quence of the four images in the bottom partion of this Figure,
one can see its expansion and evolution at different times.
Martínez-de Dios
et al.
(2011) and Merino
et al.
(2010 and
2012) designed and tested three
UAVs
(two helicopters and
one blimp) equipped with infrared (non-thermal, operating in
the far-infrared band) and visual cameras in combination with
cameras distributed on ground stations. Fire measurements
and remotely sensed locations are supplied for the interces-
sion of the response brigades for resource planning against
the fire. Figure 9 displays the Interface Human Machine (
IHM
)
in the
GCS
, where visible and infrared images are visualized.
Positioning coordinates and other dynamic parameters are
also displayed, including the
UAV
trajectory and a tele-opera-
tion connection.
The system used in Krüll
et al.
(2012) for early fire de-
tection, described in the Chemical Sensors subsection, is
designed for detecting component of gases that identify the
existence of fire.
Nuclear Leaks
Nuclear leaks represent a very high-risk event, with a great
hazard at all levels, often at a great distance distances from
the event, where immediate action in post-disaster effort is an
immediate priority.
In this regard, Han
et al.
(2013) simulated a nuclear di-
saster for rapid intervention with the aim of delimiting the
298
April 2015
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
