Ultra-lightweight tomatoes disease recognition method based on efficient attention mechanism in complex environment
Frontiers in Plant Science,
Год журнала:
2025,
Номер
15
Опубликована: Фев. 13, 2025
A
variety
of
diseased
leaves
and
background
noise
types
are
present
in
images
tomatoes
captured
real-world
environments.
However,
existing
tomato
leaf
disease
recognition
models
limited
to
recognizing
only
a
single
leaf,
rendering
them
unsuitable
for
practical
applications
scenarios.
Additionally,
these
consume
significant
hardware
resources,
making
their
implementation
challenging
agricultural
production
promotion.
To
address
issues,
this
study
proposes
framework
that
integrates
detection
with
recognition.
This
includes
model
designed
diverse
complex
environments,
along
an
ultra-lightweight
diseases.
minimize
resource
consumption,
we
developed
five
inverted
residual
modules
coupled
efficient
attention
mechanism,
resulting
effectively
balances
complexity
accuracy.
The
proposed
network
was
trained
on
dataset
collected
from
real
14
contrasting
experiments
were
conducted
under
varying
conditions.
results
indicate
the
accuracy
model,
which
utilizes
is
97.84%,
0.418
million
parameters.
Compared
traditional
image
models,
presented
not
achieves
enhanced
across
noisy
environments
but
also
significantly
reduces
number
required
parameters,
thereby
overcoming
limitation
can
recognize
images.
Язык: Английский
Privacy-Preserving Neural Network Cloud Service System Based on CKKS Homomorphic Encryption
Lecture notes in electrical engineering,
Год журнала:
2025,
Номер
unknown, С. 585 - 592
Опубликована: Янв. 1, 2025
Язык: Английский
ECVNet: A Fusion Network of Efficient Convolutional Neural Networks and Visual Transformers for Tomato Leaf Disease Identification
Agronomy,
Год журнала:
2024,
Номер
14(12), С. 2985 - 2985
Опубликована: Дек. 15, 2024
Tomato
leaf
diseases
pose
a
significant
threat
to
plant
growth
and
productivity,
necessitating
the
accurate
identification
timely
management
of
these
issues.
Existing
models
for
tomato
disease
recognition
can
primarily
be
categorized
into
Convolutional
Neural
Networks
(CNNs)
Visual
Transformers
(VTs).
While
CNNs
excel
in
local
feature
extraction,
they
struggle
with
global
recognition;
conversely,
VTs
are
advantageous
extraction
but
less
effective
at
capturing
features.
This
discrepancy
hampers
performance
improvement
both
model
types
task
identification.
Currently,
fusion
that
combine
still
relatively
scarce.
We
developed
an
efficient
network
named
ECVNet
recognition.
Specifically,
we
first
designed
Channel
Attention
Residual
module
(CAR
module)
focus
on
channel
features
enhance
model’s
sensitivity
importance
channels.
Next,
created
Fusion
(CAF
effectively
extract
integrate
features,
thereby
improving
spatial
capabilities.
conducted
extensive
experiments
using
Plant
Village
dataset
AI
Challenger
2018
dataset,
achieving
state-of-the-art
cases.
Under
condition
100
epochs,
achieved
accuracy
98.88%
86.04%
dataset.
The
introduction
provides
solution
diseases.
Язык: Английский