AIoT-powered building digital twin for smart firefighting and super real-time fire forecast
Advanced Engineering Informatics,
Journal Year:
2025,
Volume and Issue:
65, P. 103117 - 103117
Published: Jan. 21, 2025
Language: Английский
Digital-Twin Enabled Evacuation to Improve Individual and Community Resilience of Building Occupants against Indoor Fires: Framework and Route-Finding Algorithm
Developments in the Built Environment,
Journal Year:
2025,
Volume and Issue:
22, P. 100672 - 100672
Published: April 1, 2025
Language: Английский
Digital twin of buildings and occupants for emergency evacuation: Framework, technologies, applications and trends
Jia‐Rui Lin,
No information about this author
Ke-Yin Chen,
No information about this author
Sheng-Yu Song
No information about this author
et al.
Advanced Engineering Informatics,
Journal Year:
2025,
Volume and Issue:
66, P. 103419 - 103419
Published: May 13, 2025
Language: Английский
Deep Learning Models for Vision-Based Occupancy Detection in High Occupancy Buildings
Journal of Building Engineering,
Journal Year:
2024,
Volume and Issue:
unknown, P. 111355 - 111355
Published: Nov. 1, 2024
Language: Английский
Complex Indoor Human Detection with You Only Look Once: An Improved Network Designed for Human Detection in Complex Indoor Scenes
Yang Xu,
No information about this author
Yan Fu
No information about this author
Applied Sciences,
Journal Year:
2024,
Volume and Issue:
14(22), P. 10713 - 10713
Published: Nov. 19, 2024
Indoor
human
detection
based
on
artificial
intelligence
helps
to
monitor
the
safety
status
and
abnormal
activities
of
body
at
any
time.
However,
complex
indoor
environment
background
pose
challenges
task.
The
YOLOv8
algorithm
is
a
cutting-edge
technology
in
field
object
detection,
but
it
still
affected
by
low-light
environments
large
changes
scale.
To
address
these
issues,
this
article
proposes
novel
method
called
CIHD-YOLO,
which
specifically
designed
for
detection.
proposed
combines
spatial
pyramid
pooling
backbone
with
an
efficient
partial
self-attention,
enabling
network
effectively
capture
long-range
dependencies
establish
global
correlations
between
features,
obtaining
feature
information
different
scales.
At
same
time,
GSEAM
module
GSCConv
were
introduced
into
neck
compensate
loss
caused
differences
lighting
levels
combining
depth-wise
separable
convolution
residual
connections,
extract
effective
features
from
visual
data
poor
illumination
levels.
A
dataset
HCIE
dataset,
was
constructed
used
evaluate
model
paper.
research
results
show
that
compared
original
YOLOv8s
framework,
accuracy
has
been
improved
2.67%,
required
floating-point
operations
have
reduced.
comprehensive
case
analysis
comparative
evaluation
highlight
superiority
effectiveness
tasks.
Language: Английский
Displacement-based dynamic stability determination and assessment of single-layer reticulated dome structures under seismic actions
Tianhong Zheng,
No information about this author
Deng Mao,
No information about this author
Guangliang Zhang
No information about this author
et al.
Journal of Building Engineering,
Journal Year:
2024,
Volume and Issue:
unknown, P. 111401 - 111401
Published: Nov. 1, 2024
Language: Английский
Attendance Tracking System using Many Battery-free Photovoltaic Bluetooth Beacon Badges
Chenyang Wu,
No information about this author
Lei Che,
No information about this author
Siyu Jin
No information about this author
et al.
Published: Nov. 4, 2024
The
concept
of
ambient
IoT
was
introduced
by
3GPP
to
describe
low-cost,
self-powered,
or
battery-free
sensor
nodes,
which
may
reach
up
10
trillion
units
in
the
future.
By
replacing
chemical
batteries
many
standalone
devices,
we
can
achieve
ubiquitous
connectivity
through
environmentally
friendly
and
maintenance-free
solutions.
This
paper
presents
a
systematic
design
an
attendance
tracking
system
utilizing
photovoltaic
(PV)
Bluetooth
beacon
badges,
along
with
several
stationary
gateways.
In
particular,
experiences
real-world
massive
deployments
these
badges
are
emphasized.
Thanks
customized
power
management
that
only
nano-watt
leakage
before
activation,
operate
low-light
environments
gradually
accumulating
energy
from
indoor
lighting,
functioning
conditions
as
low
17
Lux.
prioritizes
cost-effectiveness,
making
suitable
for
potential
commercialization,
each
compact
badge
measuring
40
×
30
4
mm3
bill
materials
(BOM)
costing
less
than
$1.
field
tests,
every
attendee
at
academic
conference
wore
souvenir
PV
badge,
broadcasted
packet
once
sufficient
accumulated.
Stationary
gateways
installed
various
rooms
collected
information,
while
cloud-based
program
developed
visualize
results.
Language: Английский