GIScience & Remote Sensing,
Journal Year:
2022,
Volume and Issue:
59(1), P. 2247 - 2265
Published: Dec. 12, 2022
Accurate
crop
mapping
is
critical
for
agricultural
applications.
Although
studies
have
combined
deep
learning
methods
and
time-series
satellite
images
to
classification
with
satisfactory
results,
most
of
them
focused
on
supervised
methods,
which
are
usually
applicable
a
specific
domain
lose
their
validity
in
new
domains.
Unsupervised
adaptation
(UDA)
was
proposed
solve
this
limitation
by
transferring
knowledge
from
source
domains
labeled
samples
target
unlabeled
samples.
Particularly,
multi-source
UDA
(MUDA)
powerful
extension
that
leverages
multiple
can
achieve
better
results
the
than
single-source
(SUDA).
However,
few
explored
potential
MUDA
mapping.
This
study
model
(MUCCM)
unsupervised
Specifically,
11
states
U.S.
were
selected
as
domains,
three
provinces
Northeast
China
individual
Ten
spectral
bands
five
vegetation
indexes
collected
at
10-day
interval
Sentinel-2
build
MUCCM.
Subsequently,
SUDA
Domain
Adversarial
Neural
Network
(DANN)
two
direct
transfer
namely,
neural
network
random
forest,
constructed
compared
The
indicated
models
outperformed
significantly,
MUCCM
superior
DANN,
achieving
highest
accuracy
(OA>85%)
each
domain.
In
addition,
also
performed
best
in-season
forecasting
first
apply
demonstrate
novel,
effective
solution
high-performance
regions
without
Scientific Data,
Journal Year:
2021,
Volume and Issue:
8(1)
Published: Feb. 2, 2021
Abstract
Northeast
China
is
the
leading
grain
production
region
in
where
one-fifth
of
national
produced;
however,
consistent
and
reliable
crop
maps
are
still
unavailable,
impeding
management
decisions
for
regional
food
security.
Here,
we
produced
annual
10-m
major
crops
(maize,
soybean,
rice)
from
2017
to
2019,
by
using
(1)
a
hierarchical
mapping
strategy
(cropland
followed
classification),
(2)
agro-climate
zone-specific
random
forest
classifiers,
(3)
interpolated
smoothed
10-day
Sentinel-2
time
series
data,
(4)
optimized
features
spectral,
temporal,
texture
characteristics
land
surface.
The
resultant
have
high
overall
accuracies
(OA)
spanning
0.81
0.86
based
on
abundant
ground
truth
data.
satellite
estimates
agreed
well
with
statistical
data
most
municipalities
(R
2
≥
0.83,
p
<
0.01).
This
first
effort
at
resolution,
which
permits
assessing
performance
soybean
rejuvenation
plan
rotation
practice
China.
Remote Sensing of Environment,
Journal Year:
2021,
Volume and Issue:
266, P. 112708 - 112708
Published: Oct. 1, 2021
Detailed
parcel-level
crop
type
mapping
for
the
whole
European
Union
(EU)
is
necessary
evaluation
of
agricultural
policies.
The
Copernicus
program,
and
Sentinel-1
(S1)
in
particular,
offers
opportunity
to
monitor
land
at
a
continental
scale
timely
manner.
However,
so
far
potential
S1
has
not
been
explored
such
scale.
Capitalizing
on
unique
LUCAS
2018
in-situ
survey,
we
present
first
map
10-m
spatial
resolution
EU
based
S1A
S1B
Synthetic
Aperture
Radar
observations
year
2018.
Random
forest
classification
algorithms
are
tuned
detect
19
different
types.
We
assess
accuracy
this
with
three
approaches.
First,
assessed
independent
core
over
continent.
Second,
an
assessment
done
specifically
main
types
from
farmers
declarations
6
member
countries
or
regions
totaling
>3M
parcels
8.21
Mha.
Finally,
areas
derived
by
compared
subnational
(NUTS
2)
area
statistics
reported
Eurostat.
overall
as
80.3%
when
grouping
classes
76%
considering
all
separately.
Highest
accuracies
obtained
rape
turnip
user
produced
higher
than
96%.
correlation
between
remotely
sensed
estimated
Eurostat
ranges
0.93
(potatoes)
0.99
(rape
rape).
discuss
how
framework
presented
here
can
underpin
operational
delivery
in-season
high-resolution
mapping.
Scientific Data,
Journal Year:
2022,
Volume and Issue:
9(1)
Published: Aug. 5, 2022
Multiple
cropping
is
a
widespread
approach
for
intensifying
crop
production
through
rotations
of
diverse
crops.
Maps
intensity
with
descriptions
are
important
supporting
sustainable
agricultural
management.
As
the
most
populated
country,
China
ranked
first
in
global
cereal
and
percentages
multiple-cropped
land
twice
average.
However,
there
no
reliable
updated
national-scale
maps
patterns
China.
Here
we
present
recent
annual
500-m
MODIS-based
national
multiple
systems
using
phenology-based
mapping
algorithms
pixel
purity-based
thresholds,
which
provide
information
on
three
staple
crops
(maize,
paddy
rice,
wheat).
The
produced
achieved
an
overall
accuracy
89%
based
ground
truth
data,
good
agreement
statistical
data
(R
International Journal of Applied Earth Observation and Geoinformation,
Journal Year:
2023,
Volume and Issue:
117, P. 103178 - 103178
Published: Jan. 9, 2023
Northeast
China
is
one
of
the
most
major
grain
banks
in
and
has
an
overwhelming
influence
on
food
security.
To
mitigate
challenges
caused
by
increasing
demands
soil
protection,
crop
rotation
fallowing
policies
have
been
introduced
China.
These
protection
change
annual
planting
area
distribution.
monitor
type
its
changes
a
regional
scale
time
series,
we
explore
automatic
sampling
approach
hexagon
strategy
tile-based
classification
random
forest
(RF)
algorithm
using
time-series
Landsat-8
Operational
Land
Imager
(OLI)
images
during
2013–2021.
The
maps
high
credibility
with
overall
accuracies
(OA)
wall-to-wall
ranging
from
0.89
to
0.97,
also
close
agreement
statistical
data
city
city.
This
study
provides
highly
reliable
long-term
dataset,
which
can
be
helpful
for
security
agricultural
production
management.
Scientific Data,
Journal Year:
2022,
Volume and Issue:
9(1)
Published: July 15, 2022
As
a
routine
agricultural
practice,
irrigation
is
fundamental
to
protect
crops
from
water
scarcity
and
ensure
food
security
in
China.
However,
consistent
reliable
maps
about
the
spatial
distribution
extent
of
irrigated
croplands
are
still
unavailable,
impeding
resource
management
planning.
Here,
we
produced
annual
500-m
cropland
across
China
for
2000-2019,
using
two-step
strategy
that
integrated
statistics,
remote
sensing,
existing
products
into
hybrid
dataset.
First,
generated
intermediate
(MIrAD-GI)
by
fusing
MODIS-derived
greenness
index
statistical
data.
Second,
collected
all
available
over
them
with
MIrAD-GI
an
improved
series
maps,
constrained
statistics
synergy
mapping
method.
The
resultant
had
moderate
overall
accuracies
(0.732~0.819)
based
on
nationwide
reference
ground
samples
outperformed
inter-comparison.
first
this
kind
China,
delineated
spatiotemporal
pattern
could
contribute
sustainable
use
development.
International Journal of Applied Earth Observation and Geoinformation,
Journal Year:
2022,
Volume and Issue:
107, P. 102692 - 102692
Published: Jan. 29, 2022
The
30
m
resolution
U.S.
Department
of
Agriculture
(USDA)
crop
data
layer
(CDL)
is
a
widely
used
type
map
for
agricultural
management
and
assessment,
environmental
impact
food
security.
A
finer
can
potentially
reduce
errors
related
to
area
estimation,
field
size
characterization,
precision
agriculture
activities
that
requires
growth
information
at
scales
than
field.
This
study
develop
method
mapping
using
Sentinel-2
10
bands
(i.e.,
red,
green,
blue,
near-infrared)
examine
the
benefit
derived
map.
was
conducted
two
areas
with
significantly
different
sizes
types
in
South
Dakota
California,
respectively.
surface
reflectance
normalized
difference
vegetation
index
(NDVI)
acquired
2019
growing
season
were
generate
monthly
median
composites
as
classification
input.
training
evaluation
samples
from
CDL
by
(i)
finding
good
quality
pixels
(ii)
identifying
single
representative
pixel
time
series
each
pixel.
random
forest
algorithm
trained
80%
evaluated
20%
remaining
samples,
results
showed
high
overall
accuracies
94%
83%
California
areas,
major
crops
both
obtained
user's
producer's
(>87%).
There
agreement
between
class
proportions
R2
≥
0.94
root
mean
square
error
(RMSE)
≤
3%.
More
importantly,
compared
CDL,
has
much
less
salt-pepper
boundary-aliasing
effects
defines
better
small
features
(e.g.,
fields,
roads,
rivers).
potential
large
discussed.
