Frontiers in Plant Science,
Год журнала:
2025,
Номер
16
Опубликована: Март 13, 2025
Automated
detection
of
apple
leaf
diseases
is
crucial
for
predicting
and
preventing
losses
enhancing
yields.
However,
in
complex
natural
environments,
factors
such
as
light
variations,
shading
from
branches
leaves,
overlapping
disease
spots
often
result
reduced
accuracy
detecting
diseases.
To
address
the
challenges
small-target
on
leaves
backgrounds
difficulty
mobile
deployment,
we
propose
an
enhanced
lightweight
model,
ELM-YOLOv8n.To
mitigate
high
consumption
computational
resources
real-time
deployment
existing
models,
integrate
Fasternet
Block
into
C2f
backbone
network
neck
network,
effectively
reducing
parameter
count
load
model.
In
order
to
enhance
network’s
anti-interference
ability
its
capacity
differentiate
between
similar
diseases,
incorporate
Efficient
Multi-Scale
Attention
(EMA)
within
deep
structure
in-depth
feature
extraction.
Additionally,
design
a
detail-enhanced
shared
convolutional
scaling
head
(DESCS-DH)
enable
model
capture
edge
information
issues
poor
performance
object
across
different
scales.
Finally,
employ
NWD
loss
function
replace
CIoU
function,
allowing
locate
identify
small
targets
more
accurately
further
robustness,
thereby
facilitating
rapid
precise
identification
Experimental
results
demonstrate
ELM-YOLOv8n’s
effectiveness,
achieving
94.0%
F1
value
96.7%
mAP50
value—a
significant
improvement
over
YOLOv8n.
Furthermore,
are
by
44.8%
39.5%,
respectively.
The
ELM-YOLOv8n
better
suited
devices
while
maintaining
accuracy.
Artificial Intelligence in Agriculture,
Год журнала:
2024,
Номер
12, С. 1 - 18
Опубликована: Март 11, 2024
Leaf
blight
spot
disease,
caused
by
bacteria
and
fungi,
poses
a
threat
to
plant
health,
leading
leaf
discoloration
diminished
agricultural
yield.
In
response,
we
present
MobileNetV3-based
classifier
designed
for
the
Jasmine
plant,
leveraging
lightweight
Convolutional
Neural
Networks
(CNNs)
accurately
identify
disease
stages.
The
model
integrates
depth-wise
convolution
layers
max
pool
enhanced
feature
extraction,
focusing
on
crucial
low-level
features
indicative
of
disease.
Through
preprocessing
techniques,
including
data
augmentation
with
Conditional
GAN
Particle
Swarm
Optimization
selection,
achieves
robust
performance.
Evaluation
curated
datasets
demonstrates
an
outstanding
97%
training
accuracy,
highlighting
its
efficacy.
Real-world
testing
diverse
conditions,
such
as
extreme
camera
angles
varied
lighting,
attests
model's
resilience,
yielding
test
accuracies
between
94%
96%.
dataset's
tailored
design
CNN-based
classification
ensures
result
reliability.
Importantly,
classification,
marked
fast
computation
time
reduced
size,
positions
it
efficient
solution
real-time
applications.
This
comprehensive
approach
underscores
proposed
classifier's
significance
in
addressing
challenges
commercial
crops.