2018 Winter Simulation Conference (WSC),
Journal Year:
2023,
Volume and Issue:
unknown, P. 2710 - 2721
Published: Dec. 10, 2023
This
study
introduces
a
deep
learning-based
method
for
indoor
3D
object
detection
and
localization
in
healthcare
facilities.
incorporates
spatial
channel
attention
mechanisms
into
the
YOLOv5
architecture,
ensuring
balance
between
accuracy
computational
efficiency.
The
network
achieves
an
AP50
of
67.6%,
mAP
46.7%,
real-time
rate
with
FPS
67.
Moreover,
proposes
novel
mechanism
estimating
coordinates
detected
objects
projecting
them
onto
maps,
average
error
0.24
m
0.28
x
y
directions,
respectively.
After
being
tested
validated
real-world
data
from
university
campus,
proposed
shows
promise
improving
disinfection
efficiency
facilities
by
enabling
robot
navigation.
Buildings,
Journal Year:
2024,
Volume and Issue:
14(4), P. 898 - 898
Published: March 26, 2024
After
a
disaster,
ascertaining
the
operational
state
of
extensive
infrastructures
and
building
clusters
on
regional
scale
is
critical
for
rapid
decision-making
initial
response.
In
this
context,
use
remote
sensing
imagery
has
been
acknowledged
as
valuable
adjunct
to
simulation
model-based
prediction
methods.
However,
key
question
arises:
how
link
these
images
dependable
assessment
results,
given
their
inherent
limitations
in
incompleteness,
suboptimal
quality,
low
resolution?
This
article
comprehensively
reviews
methods
post-disaster
damage
recognition
through
sensing,
with
particular
emphasis
thorough
discussion
challenges
encountered
detection
various
approaches
attempted
based
resultant
findings.
We
delineate
process
literature
review,
research
workflow,
areas
present
study.
The
analysis
result
highlights
merits
image-based
methods,
such
cost,
high
efficiency,
coverage.
As
result,
evolution
using
categorized
into
three
stages:
visual
inspection
stage,
pure
algorithm
data-driven
stage.
Crucial
advances
algorithms
pertinent
topic
are
reviewed,
details
motivation,
innovation,
quantified
effectiveness
assessed
test
data.
Finally,
case
study
performed,
involving
seven
state-of-the-art
AI
models,
which
applied
sample
sets
obtained
from
2024
Noto
Peninsula
earthquake
Japan
2023
Turkey
earthquake.
To
facilitate
cohesive
grasp
implementation
practical
application,
we
have
deliberated
analytical
outcomes
accentuated
characteristics
each
method
practitioner’s
lens.
Additionally,
propose
recommendations
improvements
be
considered
advancement
advanced
algorithms.
Buildings,
Journal Year:
2024,
Volume and Issue:
14(8), P. 2344 - 2344
Published: July 29, 2024
Natural
disasters
pose
significant
threats
to
human
life
and
property,
exacerbated
by
their
sudden
onset
increasing
frequency.
This
paper
conducts
a
comprehensive
bibliometric
review
explore
robust
methodologies
for
post-disaster
building
damage
assessment
reconnaissance,
focusing
on
the
integration
of
advanced
data
collection
technologies
computational
techniques.
The
objectives
this
study
were
assess
current
landscape
methodologies,
highlight
technological
advancements,
identify
trends
gaps
in
literature.
Using
structured
approach
collection,
analyzed
370
journal
articles
from
Scopus
database
2014
2024,
emphasizing
recent
developments
remote
sensing,
including
satellite
UAV
technologies,
application
machine
learning
deep
detection
analysis.
Our
findings
reveal
substantial
advancements
analysis
techniques,
underscoring
critical
role
sensing
enhancing
disaster
assessments.
results
are
as
they
areas
requiring
further
research
development,
particularly
fusion
real-time
processing
capabilities,
model
generalization,
technology
enhancements,
training
rescue
team.
These
crucial
improving
management
practices
community
resilience.
our
is
relevant
developing
more
effective
emergency
response
strategies
informing
policy-making
disaster-prepared
social
infrastructure
planning.
Future
should
focus
closing
identified
leveraging
cutting-edge
advance
field
management.
Concrete
cracking
in
bridges
significantly
endangers
their
safety
and
integrity.
Traditional
crack
detection
methods,
reliant
on
human
visual
inspection,
are
labor-intensive
prone
to
errors.
This
paper
introduces
a
unique
framework
for
bridge
integration
with
building
information
models
(BIM),
trialed
423-ft
Atlanta,
Georgia.
The
comprises
two
main
stages:
(1)
creating
BIM
model
using
drone-captured
images
structure
from
motion
(SFM)
photogrammetry,
(2)
utilizing
deep
learning-based
encoder-decoder
network
segment
cracks
orthomosaic
superimpose
these
segmented
onto
the
model.
suggested
method
showed
robust
performance,
achieving
mean
intersection
over
union
(mIoU)
of
0.787,
precision
0.751,
recall
0.742.
These
results
underline
potential
proposed
improve
efficiency
inspection
processes.
Construction Research Congress 2022,
Journal Year:
2024,
Volume and Issue:
unknown, P. 1029 - 1038
Published: March 18, 2024
Natural
disasters
such
as
wildfires,
landslides,
and
earthquakes
result
in
obstructions
on
roads
due
to
fallen
trees,
rocks.
Such
can
cause
significant
mobility
problems
for
both
evacuees
first
responders,
especially
the
immediate
aftermath
of
disasters.
Unmanned
Aerial
Vehicles
(UAVs)
provide
an
opportunity
perform
rapid
remote
reconnaissance
planned
routes
thus
decision-makers
with
information
relating
a
route's
feasibility.
However,
detecting
obstacles
manually
is
laborious
error-prone
task,
when
attention
diverted
needs
that
are
more
urgent
during
disaster
scenarios.
This
paper
proposes
computer
vision
machine-learning
framework
detect
road
automatically
ensure
its
possibility
The
implements
YOLO
algorithm
segment
images
from
UAVs
reference
publicly
available
datasets.
retrieved
segmented
counted
pixels
roadway
comparison
difference
identify
obstruction
road.
In
addition,
method
proposed
found
region
interest
(ROI)
only
videos
UAVs.
Preliminary
results
test
runs
presented
along
future
steps
implementing
real-time
UAV-based
system.
