International Journal of Innovative Science and Research Technology (IJISRT),
Год журнала:
2024,
Номер
unknown, С. 783 - 789
Опубликована: Июль 25, 2024
The
emerging
field
of
"Smart
Face
Recognition"
utilizes
IoT
and
machine
learning
to
accurately
identify
individuals
based
on
their
facial
characteristics.
Various
industries
such
as
security,
retail,
healthcare
are
leveraging
this
technology
enhance
customer
satisfaction
increase
productivity.
By
combining
learning,
large
amounts
data
can
be
collected
from
multiple
sources,
cameras
sensors,
used
train
algorithms
for
real-time,
precise
identification
individuals.
This
is
gaining
popularity
due
its
accuracy,
speed,
scalability,
making
it
essential
applications
like
security
access
control.
Recognizing
human
emotions
a
key
focus
in
today's
technological
landscape,
with
robotic
across
various
sectors
highlighting
the
importance
emotion
recognition
effective
human-robot
interaction.
project
aims
develop
implement
new
automated
system
detection
using
Artificial
Intelligence
(AI)
Internet
Things
(IoT).
Journal of Machine and Computing,
Год журнала:
2024,
Номер
unknown, С. 308 - 316
Опубликована: Апрель 5, 2024
Even
though
Convolutional
Neural
Networks
(CNNs)
have
greatly
improved
face-related
algorithms,
it
is
still
difficult
to
keep
both
accuracy
and
efficiency
in
real-world
applications.
The
most
cutting-edge
approaches
use
deeper
networks
improve
performance,
but
the
increased
computing
complexity
number
of
parameters
make
them
impractical
for
usage
mobile
To
tackle
these
issues,
this
article
presents
a
model
object
detection
that
combines
Deeplabv3+
with
Swin
transformer,
which
incorporates
GLTB
Swin-Conv-Dspp
(SCD).
start
with,
order
lessen
impact
hole
phenomena
loss
fine-grained
data,
we
employ
SCD
component,
capable
efficiently
extracting
feature
information
from
objects
at
various
sizes.
Secondly,
properly
address
issue
challenging
recognition
due
occlusion,
study
builds
spatial
pyramid
pooling
shuffle
module.
This
module
allows
extraction
important
detail
few
noticeable
pixels
blocked
objects.
Crocodile
search
algorithm
(CSA)
enhances
classification
by
selecting
model's
fine-tuning.
On
benchmark
dataset
known
as
WFLW,
experimentally
validates
suggested
model.
Compared
other
light
models,
experimental
findings
show
delivers
higher
performance
significantly
fewer
reduced
complexity.
Jurnal Online Informatika,
Год журнала:
2024,
Номер
9(1), С. 18 - 28
Опубликована: Апрель 26, 2024
The
bus
public
transportation
system
has
low
reliability
and
ability
to
predict
the
number
of
passengers.
accuracy
predicting
passengers
by
transport
operators
is
still
weak,
which
results
in
failure
implement
solutions
operators.
A
prediction
model
with
LSTM
based
on
deep
learning
proposed
for
4
(Go
Bus,
New
Zealand
Pavlovich,
Ritchies)
are
evaluated
MSLE,
MAPE,
SMAPE
variations
epoch,
batch
size,
neurons.
dataset
a
CSV
performance
report
Auckland
Transport
(AT)
metro
patronage
buses
(01/01/2019-07/31/2023).
best
was
obtained
from
lowest
evaluation
value
relatively
fast
time
at
epoch
60,
size
16,
neurons
32.
training
testing
data
improved
suitability
tuning.
performs
predictions
12
months
later
simultaneously
predicted
fluctuations
occurring
simultaneously.
Strong
negative
correlation
Bus-Pavlovich,
strong
positive
Go
Bus
Ritchies
Pavlovich.
Predictions
that
less
closely
related
dependent
against
Ritchies.
modeling
can
be
used
as
basis
creating
operator
policies
strategies
deal
passenger
development
new
models.
Journal of Intelligent Systems,
Год журнала:
2024,
Номер
33(1)
Опубликована: Янв. 1, 2024
Abstract
Road
toll
tax
contributes
significantly
in
the
economic
development
of
any
nation.
In
developing
countries,
collection
is
carried
out
either
manually
or
electronically.
However,
both
approaches
suffer
from
various
challenges,
including
prolonged
waiting
times,
lack
transparency,
high
operational
costs,
and
concerns
regarding
data
security
privacy.
This
research
aims
to
address
these
challenges
using
a
blockchain-based
system.
The
proposed
system
employs
advanced
image
processing
techniques,
specifically
“You
Only
Look
Once”
version
5
(YOLOv5),
accurately
capture
store
vehicles’
registration
numbers
local
server
situated
at
plazas.
Subsequently,
vehicle
identification,
along
with
driver’s
credentials,
transmitted
an
application
server,
where
Ethereum
smart
contract
verifies
information
automatically
deducts
charges
account.
results
this
study
indicate
that
effectively
reduces
time
facilitates
uninterrupted
vehicular
movement.
Additionally,
ensures
transaction
safeguards
privacy
details,
non-stop
payments,
rendering
unnecessary
cash
payments
radio-frequency
identification
scanning
booths,
incorporates
decentralized
architectural
framework
enhance
mitigate
potential
failures.
Transport Policy,
Год журнала:
2024,
Номер
155, С. 309 - 320
Опубликована: Июль 9, 2024
Airborne
infections
pose
significant
challenges
to
public
transportation
systems
which
can
result
in
decline
ridership
levels
and
financial
stress
for
operators.
This
systematic
review
presents
a
comprehensive
overview
of
measures
strategies
employed
by
ground
agencies
protect
passengers
staff
while
ensuring
the
uninterrupted
operation.
study
also
conducted
bibliometric
analysis
provide
insights
into
key
topics,
publication
patterns,
major
contributors
field
airborne
transmission
research
transportation.
We
have
included
studies
published
from
January
2003
June
2024,
reported
recommendations
managing
reduce
virus
transmission.
Of
2848
initially
identified
studies,
69
met
our
eligibility
criteria.
Our
four
prevent
transportation,
including
air
quality
improvement,
cleaning,
mask-wearing,
social
distancing
vehicles
stations.
While
poses
challenge
integration
crowd
management
techniques
technology-driven
information
dissemination
effective
capacity.
The
adoption
solutions,
such
as
efficient
filtration
systems,
automated
mask
detection
mechanisms,
ultraviolet
disinfection
devices,
real-time
passenger
information,
is
required
implement
these
effectively.
Transportation
utilize
an
infection
risk
calculator
during
pandemics
beyond
assess
mitigate
various
modes
Lessons
Covid-19
pandemic
underscored
need
developing
advanced
technologies
enhance
safety
International Journal of Innovative Science and Research Technology (IJISRT),
Год журнала:
2024,
Номер
unknown, С. 783 - 789
Опубликована: Июль 25, 2024
The
emerging
field
of
"Smart
Face
Recognition"
utilizes
IoT
and
machine
learning
to
accurately
identify
individuals
based
on
their
facial
characteristics.
Various
industries
such
as
security,
retail,
healthcare
are
leveraging
this
technology
enhance
customer
satisfaction
increase
productivity.
By
combining
learning,
large
amounts
data
can
be
collected
from
multiple
sources,
cameras
sensors,
used
train
algorithms
for
real-time,
precise
identification
individuals.
This
is
gaining
popularity
due
its
accuracy,
speed,
scalability,
making
it
essential
applications
like
security
access
control.
Recognizing
human
emotions
a
key
focus
in
today's
technological
landscape,
with
robotic
across
various
sectors
highlighting
the
importance
emotion
recognition
effective
human-robot
interaction.
project
aims
develop
implement
new
automated
system
detection
using
Artificial
Intelligence
(AI)
Internet
Things
(IoT).
