To
efficiently
control
healthcare
costs,
early
and
precise
diagnosis
of
lung
illnesses
is
essential.
Using
deep
learning
(DL)
transfer
(TL),
this
study
proposes
a
fresh
method
for
categorizing
five
types
illnesses.
Ten
thousand
chest
X-ray
pictures
were
used
as
training
assessment
dataset.
improve
results,
we
TL
model
predicated
on
the
MobileNet
V2
structure.
The
testing
findings
show
that
suggested
works,
successfully
diagnosing
with
an
overall
accuracy
95.05%.
Performance
parameters
showed
evaluated
performance
each
class
was
superior
to
other
state-of-the-art
models.
model's
utility
in
disorders
exemplifies
its
applicability
imaging
diagnostics.
Furthermore,
strategy
shown
be
more
accurate
computationally
efficient
when
compared
preexisting
Testing
variety
data
sets
attested
sturdiness
generalizability.
This
shows
promise
improving
detection
diseases
by
utilizing
DL
TL.
paradigm
makes
it
easier
implement
preventative
measures,
individualize
patient
care,
boost
health
outcomes.
Tomato-spotted
wilt
virus
(TSWV)
is
a
severe
plant
disease
that
causes
significant
economic
losses
in
tomato
production
worldwide.
Early
detection
and
intensity
classification
of
TSWV-infected
plants
critical
for
effective
management.
This
study
proposes
novel
TSWV
approach
based
on
convolutional
neural
network
(CNN)
long
short-term
memory
(LSTM)
ensemble
model.
A
dataset
comprising
30,000
images
infected
with
was
gathered
annotated
six
levels,
ranging
from
0
(indicating
no
symptoms)
to
5
symptoms).
framework
developed,
aiming
enhancing
the
model’s
performance
r
proposed
achieved
an
overall
accuracy
97.37%
test
set,
outperforming
several
state-of-the-art
approaches.
We
also
performed
statistical
analysis
inter-intensity
level
variability
found
increased
level.
Our
results
suggest
has
potential
be
used
early
plants,
which
could
aid
timely
application
preventive
measures
reduce
caused
by
TSWV.
Grape
black
rot
is
a
devastating
disease
that
affects
grape
crops
globally.
Detecting
and
preventing
the
as
early
possible
crucial
for
minimizing
crop
loss
increasing
yield
quality.
In
this
study,
we
propose
CNN-LSTM
hybrid
model
multi-classification
of
severity
based
on
six
distinct
degrees.
The
obtained
an
accuracy
93.06%
after
being
trained
dataset
10,000
leaf
images
collected
from
Indian
vineyard,
outperforming
all
other
deep
learning
(DL)
traditional
machine
models.
capacity
proposed
to
capture
both
spatial
temporal
characteristics
images,
well
application
data
augmentation
techniques
halting,
contributed
its
superior
performance.
can
be
utilized
efficient
instrument
detection
prevention
disease,
thereby
contributing
enhancement
quality
crops.
However,
model's
performance
varied
depending
degree
with
reduced
classification
leaves
severe
degree.
To
enhance
ability
precisely
classify
rot,
additional
research
required.
Overall,
promising
approach
potential
applications
plant
tasks.
Pepper
Leaf
Blight
Disease
(PLBD)
is
a
widespread
plant
ailment
that
has
severe
impact
on
pepper
cultivation
across
the
globe.
The
rapid
detection
and
precise
classification
of
PLBD
severity
levels
are
crucial
for
efficient
disease
control
optimal
agricultural
productivity.
present
study
introduces
novel
model
based
Faster
region-based
convolutional
neural
network
(R-CNN)
multi-classification
in
leaves.
dataset
used
training
testing
consisted
10,000
images.
model's
performance
was
evaluated
its
accuracy
accuracy,
which
were
found
to
be
99.39%
98.38%,
respectively.
computational
efficiency
assessed
determined
sufficient
deployment
real-time
applications.
average
inference
time
0.23
seconds
per
image
renders
it
appropriate
high-throughput
study's
findings
indicate
faster
RCNN
successful
method
detecting
classifying
This
potential
enhance
management
crop
yield
farming.
Coconut
leaf
spot
(CLS)
disease
is
a
major
threat
to
coconut
production
and
can
cause
severe
economic
losses.
In
this
study,
we
propose
deep
learning
(DL)-based
ResNext50
model
for
automated
detection
severity
classification
of
CLS
disease.
Our
leverages
mode;
trained
tested
on
dataset
images
with
six
levels,
ranging
from
healthy
leaves
critical
severity.
The
proposed
approach
achieves
high
accuracy
in
detecting
classifying
the
levels
findings
suggest
that
method
successful
properly
identifying
categorizing
illness
an
rate
91.77%
overall.
strategy
has
been
presented
possibility
significantly
improve
efficiency
monitoring,
ultimately
leading
better
management
strategies
increased
productivity
industry.
Sugarcane
is
a
widely
cultivated
crop
due
to
its
high
demand
and
supply
in
various
industries.
However,
the
increase
production
levels
has
led
an
number
of
diseases
that
affect
crop.
One
most
devastating
sugarcane
red
rot
(SRR)
disease.
A
multi-layer
perceptron
(MP)
based
deep
learning
(DL)
model
was
built
for
identification
classification
SRR
illness
using
dataset
20,000
photos
leaves
order
address
this
problem.
This
trained
on
photographs
leaves.
The
classified
into
5
different
disease
severity
levels.
proposed
achieved
accuracy
rate
97.97%
binary
98.03%
overall
multi-classification.
Furthermore,
comparison
stages
carried
out,
it
demonstrated
effective
tool
accurately
categorizing
images
study
contributes
development
efficient
accurate
early
detection
diagnosis
crops,
which
essential
improving
yield
preventing
economic
losses.
2022 13th International Conference on Computing Communication and Networking Technologies (ICCCNT),
Journal Year:
2023,
Volume and Issue:
unknown
Published: July 6, 2023
The
fungus
known
as
sugarcane
downy
mildew
is
extremely
destructive
and
represents
a
substantial
risk
to
output
all
over
the
world.
To
effectively
manage
disease
safeguard
crops,
early
precise
diagnosis
of
severity
levels
caused
by
essential.
In
this
study,
we
present
novel
method
for
detecting
based
on
using
CNN-LSTM
ensemble
model.
model
combines
spatial
feature
extraction
skills
Convolutional
Neural
Networks
(CNN)
with
temporal
modeling
abilities
Long
Short-Term
Memory
(LSTM)
networks.
For
training
assessment,
dataset
consisting
leaf
pictures
that
have
been
afflicted
labeled
ranging
from
1
5
employed.
results
experiments
show
suggested
successful,
it
achieves
high
accuracy,
precision,
recall,
F1
score
while
attempting
forecast
mildew.
capacity
categorize
throughout
demonstrated
overall
accuracy
94.16%.
was
provided
provides
potential
solution
automated
identification
illness
enables
prompt
interventions
optimizes
management
practices.
proposed
study
contributes
growing
field
agricultural
informatics
helps
promote
environmentally
responsible
methods
sugarcane.
2022 13th International Conference on Computing Communication and Networking Technologies (ICCCNT),
Journal Year:
2023,
Volume and Issue:
unknown
Published: July 6, 2023
Rice
hispa
disease
is
a
severe
risk
to
agricultural
production
and
can
result
in
considerable
reductions
crop
yield.
It
necessary,
put
into
practice
efficient
management
techniques,
correctly
categorize
the
its
various
degrees
of
intensity.
In
this
study,
we
present
hybrid
model
for
multi-classification
rice
illness
that
incorporates
convolutional
neural
networks
(CNN)
Random
Forest
(RF).
A
dataset
consisting
10,000
photos,
each
which
represents
distinct
degree
intensity,
was
gathered
pre-processed.
The
CNN
component
responsible
extracting
distinguishing
characteristics
from
pictures,
while
RF
classifier
charge
incorporating
these
final
classification.
displays
competitive
performance
when
compared
classic
machine
learning
(ML)
techniques
deep
(DL)
models,
with
an
overall
accuracy
97.46%.
For
certain
measures,
such
as
accuracy,
recall,
F1-score,
are
shown.
confusion
matrix
analysis
provides
more
evidence
distinguish
between
different
states.
Our
potential
approach
accurate
reliable
disease,
paves
way
enhanced
control
production.