Plants,
Journal Year:
2024,
Volume and Issue:
13(1), P. 140 - 140
Published: Jan. 4, 2024
Nitrogen
is
a
fundamental
component
for
building
amino
acids
and
proteins,
playing
crucial
role
in
the
growth
development
of
plants.
Leaf
nitrogen
concentration
(LNC)
serves
as
key
indicator
assessing
plant
development.
Monitoring
LNC
provides
insights
into
absorption
utilization
from
soil,
offering
valuable
information
rational
nutrient
management.
This,
turn,
contributes
to
optimizing
supply,
enhancing
crop
yields,
minimizing
adverse
environmental
impacts.
Efficient
non-destructive
estimation
paramount
importance
on-field
Spectral
technology,
with
its
advantages
repeatability
high-throughput
observations,
feasible
method
obtaining
data.
This
study
explores
responsiveness
spectral
parameters
soybean
at
different
vertical
scales,
aiming
refine
management
soybeans.
research
collected
hyperspectral
reflectance
data
leaf
layers
Three
types
parameters,
nitrogen-sensitive
empirical
indices,
randomly
combined
dual-band
“three-edge”
were
calculated.
Four
optimal
index
selection
strategies
constructed
based
on
correlation
coefficients
between
each
layer.
These
included
combinations
(Combination
1),
2),
parameter
3),
mixed
combination
4).
Subsequently,
these
four
used
input
variables
build
models
soybeans
scales
using
partial
least
squares
regression
(PLSR),
random
forest
(RF),
backpropagation
neural
network
(BPNN).
The
results
demonstrated
that
reached
highest
values
upper
leaves,
most
showing
significant
correlations
(p
<
0.05).
Notably,
reciprocal
difference
(VI6)
exhibited
upper-layer
0.732,
wavelength
841
nm
842
nm.
In
constructing
layers,
accuracy
gradually
improved
increasing
height
layer
best
performance,
validation
set
coefficient
determination
(R2)
was
higher
by
9.9%
16.0%
compared
other
layers.
RF
estimating
LNC,
R2
6.2%
8.8%
models.
RMSE
lower
2.1%
7.0%,
MRE
4.7%
5.6%
Among
combinations,
Combination
4
achieved
accuracy,
2.3%
13.7%.
conclusion,
employing
input,
model
0.856,
0.551,
10.405%.
findings
this
provide
technical
support
remote
sensing
monitoring
LNCs
spatial
scales.
Plants,
Journal Year:
2024,
Volume and Issue:
13(6), P. 773 - 773
Published: March 8, 2024
Smooth
bromegrass
(Bromus
inermis)
is
a
perennial,
high-quality
forage
grass.
However,
its
seed
yield
influenced
by
agronomic
practices,
climatic
conditions,
and
the
growing
year.
The
rapid
effective
prediction
of
can
assist
growers
in
making
informed
production
decisions
reducing
agricultural
risks.
Our
field
trial
design
followed
completely
randomized
block
with
four
blocks
three
nitrogen
levels
(0,
100,
200
kg·N·ha−1)
during
2022
2023.
Data
on
remote
vegetation
index
(RVI),
normalized
difference
(NDVI),
leaf
content
(LNC),
area
(LAI)
were
collected
at
heading,
anthesis,
milk
stages.
Multiple
linear
regression
(MLR),
support
vector
machine
(SVM),
random
forest
(RF)
models
utilized
to
predict
yield.
In
2022,
results
indicated
that
application
provided
sufficiently
large
range
variation
(ranging
from
45.79
379.45
kg
ha⁻¹).
Correlation
analysis
showed
indices
RVI,
NDVI,
LNC,
LAI
presented
significant
positive
correlation
yield,
highest
coefficient
was
observed
heading
stage.
data
formulate
predictive
model
for
suggested
utilizing
stage
produced
best
performance.
SVM
RF
outperformed
MLR
prediction,
demonstrating
performance
(R2
=
0.75,
RMSE
51.93
ha−1,
MAE
29.43
MAPE
0.17).
Notably,
accuracy
predicting
year
2023
using
this
had
decreased.
Feature
importance
revealed
LNC
crucial
indicator
smooth
Further
studies
an
expanded
dataset
integration
weather
are
needed
improve
generalizability
adaptability
Agronomy,
Journal Year:
2024,
Volume and Issue:
14(6), P. 1248 - 1248
Published: June 9, 2024
Efficiently
obtaining
leaf
nitrogen
content
(LNC)
in
rice
to
monitor
the
nutritional
health
status
is
crucial
achieving
precision
fertilization
on
demand.
Unmanned
aerial
vehicle
(UAV)-based
hyperspectral
technology
an
important
tool
for
determining
LNC.
However,
intricate
coupling
between
spectral
information
and
remains
elusive.
To
address
this,
this
study
proposed
estimation
method
LNC
that
integrates
hybrid
preferred
features
with
deep
learning
modeling
algorithms
based
UAV
imagery.
The
approach
leverages
XGBoost,
Pearson
correlation
coefficient
(PCC),
a
synergistic
combination
of
both
identify
characteristic
variables
estimation.
We
then
construct
models
using
statistical
regression
methods
(partial
least-squares
(PLSR))
machine
(random
forest
(RF);
neural
networks
(DNN)).
optimal
model
utilized
map
spatial
distribution
at
field
scale.
was
conducted
National
Agricultural
Science
Technology
Park,
Guangzhou,
located
Baiyun
District
Guangdong,
China.
results
reveal
combined
PCC-XGBoost
algorithm
significantly
enhances
accuracy
inversion
compared
standalone
screening
approach.
Notably,
built
DNN
exhibits
highest
predictive
performance
demonstrates
great
potential
mapping
This
indicates
role
enhancement
utilization
efficiency
cultivation.
outcomes
offer
valuable
reference
enhancing
agricultural
practices
sustainable
crop
management.
Agronomy,
Journal Year:
2025,
Volume and Issue:
15(1), P. 175 - 175
Published: Jan. 13, 2025
In
agricultural
production,
the
nitrogen
content
of
sugarcane
is
assessed
with
precision
and
economy,
which
crucial
for
balancing
fertilizer
application,
reducing
resource
waste,
minimizing
environmental
pollution.
As
an
important
economic
crop,
productivity
significantly
influenced
by
various
factors,
especially
supply.
Traditional
methods
based
on
manually
extracted
image
features
are
not
only
costly
but
also
limited
in
accuracy
generalization
ability.
To
address
these
issues,
a
novel
regression
prediction
model
estimating
sugarcane,
named
SC-ResNeXt
(Enhanced
Self-Attention,
Spatial
Attention,
Channel
Attention
ResNeXt),
has
been
proposed
this
study.
The
Self-Attention
(SA)
mechanism
Convolutional
Block
Module
(CBAM)
have
incorporated
into
ResNeXt101
to
enhance
model’s
focus
key
its
information
extraction
capability.
It
was
demonstrated
that
achieved
test
R2
value
93.49%
predicting
leaves.
After
introducing
SA
CBAM
attention
mechanisms,
improved
4.02%.
Compared
four
classical
deep
learning
algorithms,
exhibited
superior
performance.
This
study
utilized
images
captured
smartphones
combined
automatic
feature
technologies,
achieving
precise
economical
predictions
compared
traditional
laboratory
chemical
analysis
methods.
approach
offers
affordable
technical
solution
small
farmers
optimize
management
plants,
potentially
leading
yield
improvements.
Additionally,
it
supports
development
more
intelligent
farming
practices
providing
predictions.
