Logic Journal of IGPL,
Год журнала:
2024,
Номер
unknown
Опубликована: Май 2, 2024
Abstract
The
use
of
technologies
like
artificial
intelligence
can
drive
productivity
growth,
efficiency
and
innovation.
goal
this
study
is
to
develop
an
anomaly
detection
method
for
locating
flaws
on
the
surface
sandwich
panels
using
YOLOv5.
proposed
algorithm
extracts
information
locally
from
image
through
a
prediction
system
that
creates
bounding
boxes
determines
whether
panel
contains
flaws.
It
attempts
reject
or
accept
product
based
quality
levels
specified
in
standard.
To
evaluate
method,
comparison
was
made
with
damage
convolutional
neural
network
methods
thresholding.
findings
show
which
object
detector,
more
accurate
than
alternatives.
characteristics
model,
according
standard
limit
allowable
manufacturing
obtain
product,
also
enable
improve
industrial
standards
producing
while
increasing
speed.
Algorithms,
Год журнала:
2023,
Номер
16(7), С. 343 - 343
Опубликована: Июль 17, 2023
The
verticillium
fungus
has
become
a
widespread
threat
to
olive
fields
around
the
world
in
recent
years.
accurate
and
early
detection
of
disease
at
scale
could
support
solving
problem.
In
this
paper,
we
use
YOLO
version
5
model
detect
trees
using
aerial
RGB
imagery
captured
by
unmanned
vehicles.
aim
our
paper
is
compare
different
architectures
evaluate
their
performance
on
task.
are
evaluated
two
input
sizes
each
through
most
widely
used
metrics
for
object
classification
tasks
(precision,
recall,
[email protected][email protected]:0.95).
Our
results
show
that
YOLOv5
algorithm
able
deliver
good
detecting
predicting
status,
with
having
strengths
weaknesses.
International Journal of Applied Earth Observation and Geoinformation,
Год журнала:
2023,
Номер
117, С. 103185 - 103185
Опубликована: Янв. 12, 2023
Sandy
beaches
are
subject
to
changes
due
multiple
factors,
that
both
natural
(e.g.
storms)
and
anthropogenic.
Great
efforts
being
made
monitor
these
ecosystems
understand
their
dynamics
in
order
assure
conservation.
The
identification
of
anthropogenic
its
differentiation
from
ones
is
an
important
task
for
coastal
monitoring.
In
this
study,
we
present
a
methodology
the
detection
ecosystem
by
automatically
detecting
active
bulldozers
continuous
beach
video
data.
PCA
used
highlight
consecutive
images
moving
objects.
Next,
YOLO
object
algorithm
identify
change
images.
was
specifically
trained
task,
obtaining
precision
0.94
recall
0.81.
An
automatic
tool
developed,
process
carried
out
on
two
months
data,
consisting
approximately
19
000
resulting
information
compared
with
derived
3D
data
obtained
permanent
laser
scanner.
correlation
among
results
methodologies
computed.
For
validation
area
daily
time
frame
0.88
between
number
detected
affected
height
larger
than
0.3
m.
Agronomy,
Год журнала:
2024,
Номер
14(5), С. 1042 - 1042
Опубликована: Май 14, 2024
Phenotyping
of
genetic
resources
is
an
important
prerequisite
for
the
selection
resistant
varieties
in
breeding
programs
and
research.
Computer
vision
techniques
have
proven
to
be
a
useful
tool
digital
phenotyping
diseases
interest.
One
pathogen
that
increasingly
observed
Europe
Diplocarpon
coronariae,
which
causes
apple
blotch
disease.
In
this
study,
high-throughput
method
was
established
evaluate
susceptibility
D.
coronariae.
For
purpose,
inoculation
trials
with
coronariae
were
performed
laboratory
images
infested
leaves
taken
7,
9
13
days
post
inoculation.
A
pre-trained
YOLOv5s
model
chosen
establish
model,
trained
image
dataset
927
RGB
images.
The
had
size
768
×
pixels
divided
into
738
annotated
training
images,
78
validation
111
background
without
symptoms.
accuracy
symptom
prediction
95%.
These
results
indicate
our
can
accurately
efficiently
detect
spots
acervuli
on
detached
leaves.
Object
detection
therefore
used
leaf
assays
assess
laboratory.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,
Год журнала:
2024,
Номер
17, С. 11915 - 11930
Опубликована: Янв. 1, 2024
Detection
of
individual
trees
in
Moso
bamboo
forests
is
critical
to
forestry
resource
management.
However,
accurate
and
rapid
detection
remains
a
significant
challenge
due
the
high
density
complex
canopy
structure.
This
study
proposed
new
counting
method
based
on
multiband
images.
First,
this
used
dynamic
thresholding
extract
forests'
unique
hook
tip
features
coupled
original
unmanned
aerial
vehicle
(UAV)
visible
light
images
construct
Then,
utilized
three
object
networks
(faster
R-CNN,
YOLOv5,
YOLOv7)
detect
count
number
sample
plots
using
NMS
method.
assessed
method's
accuracy
compared
UAV
with
84
forest
plots.
The
results
showed
that
detecting
improved
all
networks.
On
test
dataset,
YOLOv7
network
multi-band
had
highest
AP
(89.15%)
R
2
(93.17%),
respectively,
which
were
3.18%
15.5%
higher
than
when
Faster
R-CNN
YOLOv5
also
by
7.3%
7.2%,
respectively.
In
addition,
largest
RMSE
reduction
after
images,
37.93%
reduction.
Water Resources Research,
Год журнала:
2025,
Номер
61(2)
Опубликована: Янв. 31, 2025
Abstract
With
flooding
events
expected
to
increase
in
both
intensity
and
frequency
the
future
due
climate
change,
ensuring
safety
of
river
embankments
is
vital
withstand
flood
disasters.
Piping
one
most
harmful
embankment
hazards
season,
recent
advances
unmanned
aerial
vehicles
(UAVs)
deep
learning‐based
object
detection
have
enabled
efficient
automated
hazard
detection.
In
this
study,
a
novel
approach
that
integrates
UAV
with
edge
computing
was
proposed
for
rapid
automatic
piping
First,
total
104
field
simulation
experiments
were
conducted
across
12
different
sites
flood‐prone
areas
fill
gaps
high‐quality
data
set,
thermal
infrared
visible
sets
produced,
including
various
times
(forenoon,
afternoon,
night),
weather
conditions
(clear‐sky,
cloudy,
rainy),
locations
(bare
land,
paddy,
grassland,
pond)
flight
altitudes
(10,
20,
30
m).
Second,
model
selected
trained
on
sets.
The
well‐trained
models
precisions
92.7%
70.4%,
respectively,
recalls
84.9%
69.7%.
Furthermore,
exhibited
great
resistance
interference
from
several
types
aquatic
vegetation
could
effectively
detect
rainy
days.
integration
real‐time
piping.
method
enhances
efficiency,
contributing
intelligent
emergency
management.
