A Novel Model for Soil Organic Matter and Total Nitrogen Detection Based on Visible/Shortwave Near-Infrared Spectroscopy
Land,
Год журнала:
2025,
Номер
14(2), С. 329 - 329
Опубликована: Фев. 6, 2025
Soil
organic
matter
(SOM)
and
total
nitrogen
(TN)
are
critical
indicators
for
assessing
soil
fertility.
Although
laboratory
chemical
analysis
methods
can
accurately
measure
their
contents,
these
techniques
time-consuming
labor-intensive.
Spectral
technology,
characterized
by
its
high
sensitivity
convenience,
has
been
increasingly
integrated
with
machine
learning
algorithms
nutrient
monitoring.
However,
the
process
of
spectral
data
remains
complex
requires
further
optimization
simplicity
efficiency
to
improve
prediction
accuracy.
This
study
proposes
a
novel
model
enhance
accuracy
SOM
TN
predictions
in
northeast
China’s
black
soil.
Visible/Shortwave
Near-Infrared
Spectroscopy
(Vis/SW-NIRS)
within
350–1070
nm
range
were
collected,
preprocessed,
dimensionality-reduced.
The
scores
first
nine
principal
components
after
partial
least
squares
(PLS)
dimensionality
reduction
selected
as
inputs,
measured
contents
used
outputs
build
back-propagation
neural
network
(BPNN)
model.
results
show
that
processed
combination
standard
normal
variate
(SNV)
multiple
scattering
correction
(MSC)
have
best
modeling
performance.
To
stability
this
model,
three
named
random
search
(RS),
grid
(GS),
Bayesian
(BO)
introduced.
demonstrate
Vis/SW-NIRS
provides
reliable
PLS-RS-BPNN
achieving
performance
(R2
=
0.980
0.972,
RMSE
1.004
0.006
TN,
respectively).
Compared
traditional
models
such
forests
(RF),
one-dimensional
convolutional
networks
(1D-CNNs),
extreme
gradient
boosting
(XGBoost),
proposed
improves
R2
0.164–0.344
predicting
0.257–0.314
respectively.
These
findings
confirm
potential
technology
effective
tools
prediction,
offering
valuable
insights
application
sensing
information.
Язык: Английский
Estimation of Soil Organic Matter Based on Spectral Indices Combined with Water Removal Algorithm
Remote Sensing,
Год журнала:
2024,
Номер
16(12), С. 2065 - 2065
Опубликована: Июнь 7, 2024
Soil
moisture
strongly
interferes
with
the
spectra
of
soil
organic
matter
(SOM)
in
near-infrared
region,
which
reduces
correlation
between
and
decreases
accuracy
prediction
SOM.
In
this
study,
we
explored
feasibility
two
types
spectral
indices,
two-
three-band
mixed
(SI)
indices
(SI3),
water
removal
algorithms,
direct
standardization
(DS)
external
parameter
orthogonalization
(EPO),
to
estimate
SOM
wet
soils
using
a
total
192
samples
at
six
content
gradients.
The
estimation
accuracies
combined
algorithms
were
better
than
those
full
data
algorithms:
SI-EPO
(R2
=
0.735,
RMSEp
3.4102
g/kg)
higher
EPO
0.63,
4.1021
g/kg),
SI-DS
0.70,
3.7085
DS
0.61,
4.2806
g/kg);
SI3-EPO
0.752,
3.1344
was
SI-EPO;
both
effectively
mitigated
influence
moisture,
demonstrating
superior
performance
small-sample
scenarios.
This
study
introduces
novel
approach
counteract
impact
on
estimation.
Язык: Английский
Detecting Changes in Soil Fertility Properties Using Multispectral UAV Images and Machine Learning in Central Peru
AgriEngineering,
Год журнала:
2025,
Номер
7(3), С. 70 - 70
Опубликована: Март 6, 2025
Remote
sensing
is
essential
in
precision
agriculture
as
this
approach
provides
high-resolution
information
on
the
soil’s
physical
and
chemical
parameters
for
detailed
decision
making.
Globally,
technologies
such
remote
machine
learning
are
increasingly
being
used
to
infer
these
parameters.
This
study
evaluates
soil
fertility
changes
compares
them
with
previous
fertilization
inputs
using
multispectral
imagery
situ
measurements.
A
UAV-captured
image
was
predict
spatial
distribution
of
parameters,
generating
fourteen
spectral
indices
a
digital
surface
model
(DSM)
from
103
plots
across
49.83
hectares.
Machine
algorithms,
including
classification
regression
trees
(CART)
random
forest
(RF),
modeled
(N-ppm,
P-ppm,
K-ppm,
OM%,
EC-mS/m).
The
RF
outperformed
others,
R2
values
72%
N,
83%
P,
87%
K,
85%
OM,
70%
EC
2023.
Significant
spatiotemporal
variations
were
observed
between
2022
2023,
an
increase
P
(14.87
ppm)
reduction
(−0.954
mS/m).
High-resolution
UAV
combined
proved
highly
effective
monitoring
fertility.
approach,
tailored
Peruvian
Andes,
integrates
field-collected
data,
offering
innovative
tools
optimize
practices,
address
management
challenges,
merge
modern
technology
traditional
methods
sustainable
agricultural
practices.
Язык: Английский
Improving the accuracy of soil organic matter mapping in typical Planosol areas based on prior knowledge and probability hybrid model
Soil and Tillage Research,
Год журнала:
2024,
Номер
246, С. 106358 - 106358
Опубликована: Ноя. 14, 2024
Язык: Английский
Application Model of Hyperspectral Technology Based on Novel Spectral Indices for Salinity Assessment in Soil Heritage Sites
The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences,
Год журнала:
2024,
Номер
XLVIII-2-2024, С. 477 - 484
Опубликована: Июнь 25, 2024
Abstract.
The
Dunhuang
murals
are
a
precious
treasure
of
China's
cultural
heritage,
yet
they
have
long
been
affected
by
salt
damage.
Traditional
methods
for
detecting
content
costly,
inefficient,
and
may
cause
physical
harm
to
the
murals.
Among
current
techniques
measuring
in
murals,
hyperspectral
remote
sensing
technology
offers
non-invasive
,
circumventing
issues
high
costs,
low
efficiency.
Building
on
this,
our
study
developed
high-spectral
feature
inversion
model
mural
phosphate
using
Fractional
Order
Differentiation
(FOD),
novel
three-band
spectral
index,
Partial
Least
Squares
Regression
(PLSR)
algorithm.
specific
research
contents
include:
1)
Exploring
absorption
mechanism
phosphates
their
characteristic
bands,
combined
with
optimal
index
construct
univariate
linear
regression
model,
providing
basis
rapid
quantitative
measurement
content.
2)
By
comparing
accuracy
PSR
PNDI
indices
based
first
six
orders
highest
were
selected
as
combination,
used
explanatory
variables,
plaster
electrical
conductivity
response
variable,
employing
PLSR
method
model.
study's
findings
outcomes
different
fractional
differentiation,
it
was
found
that
performance
reached
its
optimum
at
0.3
order
differentiation
both
data,
determination
coefficient
(R2)
0.728.
Utilizing
PLSR,
this
employed
previously
determined
six-order
combination
successfully
constructing
0.815.
This
provides
an
effective
technical
means
monitoring
damage
conditions
heritage
such
Язык: Английский
The spectral inversion model for electrical conductivity in mural plaster following phosphate erosion based on fractional order differentiation and novel spectral indices
Heritage Science,
Год журнала:
2024,
Номер
12(1)
Опубликована: Авг. 6, 2024
Abstract
The
Dunhuang
murals
are
a
precious
treasure
of
China’s
cultural
heritage,
yet
they
have
long
been
affected
by
salt
damage.
Traditional
methods
for
detecting
content
costly,
inefficient,
and
may
cause
physical
harm
to
the
murals.
Among
current
techniques
measuring
in
murals,
hyperspectral
remote
sensing
technology
offers
non-invasive,
circumventing
issues
high
costs,
low
efficiency.
Building
on
this,
study
constructs
an
inversion
model
Electrical
Conductivity
(EC)
values
mural
plaster
subjected
phosphate
erosion,
through
integration
Fractional
Order
Differentiation
(FOD),
novel
three-band
spectral
index,
Partial
Least
Squares
Regression
algorithm.
specific
research
contents
include:
(1)
Initially,
preparation
experiments,
materials
used
create
samples
underwent
rigorous
desalting
process,
solutions
were
prepared
using
deionized
water
ensure
uniform
experimental
conditions
accuracy
results.
These
meticulous
preprocessing
steps
guaranteed
that
measured
EC
exhibited
clear
correlation
with
content.
Subsequently,
employing
qualitative
analysis
techniques,
this
was
able
more
accurately
simulate
real-world
scenarios
damage,
enabling
deeper
investigation
into
mechanisms
which
salts
inflict
microscopic
damage
(2)
Explores
absorption
characteristic
bands
after
erosion
plaster.
By
integrating
optimal
indices,
univariate
linear
regression
is
constructed,
providing
basis
rapid
quantitative
measurement
electrical
conductivity
(3)
comparing
Phosphate
Simple
Ratio
(PSR)
Normalized
Difference
Index
(PNDI)
indices
based
model,
first
six
orders
highest
index
selected
as
combination,
explanatory
variables,
response
variable,
PLSR
method
construct
high-spectral
feature
model.
study’s
findings
Surfaces
deteriorated
formed
numerous
irregularly
shaped
crystal
clusters,
exhibiting
uneven
characteristics.
outcomes
different
fractional
differentiation,
it
found
performance
reached
its
optimum
at
0.3
order
differentiation
both
PSR
PNDI
data,
determination
coefficient
(Q
2
)
0.728.
Utilizing
PLSR,
employed
previously
determined
six-order
combination
successfully
constructing
0.815.
This
provides
effective
technical
means
monitoring
heritage
such
Язык: Английский
Enhancing Leaf Area Index Estimation in Southern Xinjiang Fruit Trees: A Competitive Adaptive Reweighted Sampling-Successive Projections Algorithm and Three-Band Index Approach with Fractional-Order Differentiation
Forests,
Год журнала:
2024,
Номер
15(12), С. 2126 - 2126
Опубликована: Дек. 1, 2024
The
Leaf
Area
Index
(LAI)
is
a
key
indicator
for
assessing
fruit
tree
growth
and
productivity,
accurate
estimation
using
hyperspectral
technology
essential
monitoring
plant
health.
This
study
aimed
to
improve
LAI
accuracy
in
apricot,
jujube,
walnut
trees
Xinjiang,
China.
Canopy
data
were
processed
fractional-order
differentiation
(FOD)
from
0
2.0
orders
extract
spectral
features.
Three
feature
selection
methods—Competitive
Adaptive
Reweighted
Sampling
(CARS),
Successive
Projections
Algorithm
(SPA),
their
combination
(CARS-SPA)—were
applied
identify
sensitive
bands.
Various
band
combinations
used
construct
three-band
indices
(TBIs)
optimal
estimation.
Random
forest
(RF)
models
developed
validated
prediction.
results
showed
that
(1)
the
reflectance
spectra
of
jujube
similar,
while
apricot
differed.
(2)
correlation
between
differential
was
highest
at
1.4
1.7,
outperforming
integer-order
spectra.
(3)
CARS-SPA
selected
smaller
set
bands
1100~2500
nm,
reducing
collinearity
improving
index
construction.
(4)
RF
model
TBI4
demonstrated
high
R²,
low
RMSE,
an
RPD
value
>
2,
indicating
prediction
accuracy.
approach
holds
promise
trees.
Язык: Английский