Remote Sensing,
Год журнала:
2024,
Номер
16(11), С. 1987 - 1987
Опубликована: Май 31, 2024
Challenges
in
enhancing
the
multiclass
segmentation
of
remotely
sensed
data
include
expensive
and
scarce
labeled
samples,
complex
geo-surface
scenes,
resulting
biases.
The
intricate
nature
geographical
surfaces,
comprising
varying
elements
features,
introduces
significant
complexity
to
task
segmentation.
limited
label
used
train
models
may
exhibit
biases
due
imbalances
or
inadequate
representation
certain
surface
types
features.
For
applications
like
land
use/cover
monitoring,
assumption
evenly
distributed
simple
random
sampling
be
not
satisfied
spatial
stratified
heterogeneity,
introducing
that
can
adversely
impact
model’s
ability
generalize
effectively
across
diverse
areas.
We
introduced
two
statistical
indicators
encode
geo-features
under
scenes
designed
a
corresponding
optimal
scheme
select
representative
samples
reduce
bias
during
machine
learning
model
training,
especially
deep
models.
results
scores
showed
entropy-based
gray-based
detected
from
scenes:
indicator
was
sensitive
boundaries
different
classes
contours
objects,
while
Moran’s
I
had
better
performance
identifying
structure
information
objects
remote
sensing
images.
According
scores,
methods
appropriately
adapted
distribution
training
geo-context
enhanced
their
representativeness
relative
population.
single-score
method
achieved
highest
improvement
DeepLab-V3
(increasing
pixel
accuracy
by
0.3%
MIoU
5.5%),
multi-score
SegFormer
ACC
0.2%
2.4%).
These
findings
carry
implications
for
quantifying
hence
enhance
semantic
high-resolution
images
with
less
bias.
GIScience & Remote Sensing,
Год журнала:
2025,
Номер
62(1)
Опубликована: Янв. 2, 2025
Revealing
the
factors
associated
with
traffic
accident
risk
across
cities
nationwide,
including
demographic
and
economic
elements,
is
crucial
for
supporting
safety
policy,
urban
planning,
insurance
evaluation.
Spatial
stratified
heterogeneity
models,
such
as
Geographically
Optimized
Zone-based
Heterogeneity
(GOZH)
model,
are
widely
used
analyzing
spatial
association
in
large
scale.
However,
their
discretization
process
heavily
depends
on
a
manually
set
complexity
parameter
(cp),
introducing
significant
uncertainty.
To
address
this,
we
developed
Robust
GOZH
(RGOZH),
by
inter-parameter
relationships
within
Q
function
an
optimization
to
achieve
precise
controlled
geographic
partitioning.
By
selecting
optimal
cp,
RGOZH
produces
most
reliable
partitioning
results.
Testing
Germany's
dataset
revealed
strong
patterns,
achieving
superior
groupings
while
maintaining
over
80%
of
explanatory
power
–
stark
contrast
less
interpretable
results
from
GOZH.
identified
vehicle
ownership,
government
employee
proportion,
income
level
primary
shaping
risk.
This
study
highlights
critical
role
large-scale
pattern
analysis
management
establishes
robust
framework
future
interdisciplinary
geospatial
research.
Furthermore,
provides
replicable
method
that
can
adapt
various
regional
datasets,
enhancing
its
utility
international
studies.
As
methodological
advancement,
demonstrates
value
integrating
optimized
parameters
into
predictive
models.
Water,
Год журнала:
2025,
Номер
17(2), С. 230 - 230
Опубликована: Янв. 16, 2025
Fujian
Province
is
an
important
soil
and
water
conservation
region
in
hilly
South
China.
However,
there
has
been
limited
attention
paid
to
the
assessment
of
production
at
provincial
level,
distribution
patterns
ecosystem
services
under
different
environmental
gradients
regions
have
not
revealed.
This
study
evaluated
spatiotemporal
characteristics
yield
based
on
InVEST
model
2000,
2010,
2020,
explored
their
differences
six
gradients:
elevation,
slope,
terrain
position
index,
geomorphy,
LULC,
NDVI.
The
results
statistics
showed
significant
spatial
differentiation
temporal
change
yield;
changes
both
exhibited
obvious
clustering
cold
hot
spots
(low
high
values);
cities
were
higher
than
those
conservation.
index
Geodetector
that
retention
gradients;
generally
lower
degree
more
sensitive
response
factors
(slope,
TPI,
DEM).
high-value
1000
2160
m
for
DEM,
25°
70.2°
0.81
1.42
medium
mountain
forest
land
0.9
0.92
NDVI,
which
indicates
mountainous
with
altitude,
steep
slopes,
changes,
vegetation
coverage.
exhibit
distributions
across
gradients,
should
be
adapting
local
conditions
ecological
environment
development.
