Land,
Год журнала:
2024,
Номер
13(11), С. 1840 - 1840
Опубликована: Ноя. 5, 2024
Fractional
vegetation
cover
(FVC)
plays
a
key
role
in
ecological
and
environmental
status
assessment
because
it
directly
reflects
the
extent
of
its
status,
yet
is
an
important
component
ecosystems.
FVC
estimation
methods
have
evolved
from
traditional
manual
interpretation
to
advanced
remote
sensing
technologies,
such
as
satellite
data
analysis
unmanned
aerial
vehicle
(UAV)
image
processing.
Extraction
based
on
high-resolution
UAV
are
being
increasingly
studied
fields
ecology
sensing.
However,
research
UAV-based
extraction
against
backdrop
high
soil
reflectance
arid
regions
remains
scarce.
In
this
paper,
12
visible
light
images
differentiated
scenarios
Ebinur
Lake
basin,
Xinjiang,
China,
various
used
for
high-precision
estimation:
Otsu’s
thresholding
method
combined
with
Visible
Vegetation
Indices
(abbreviated
Otsu-VVIs)
(excess
green
index,
excess
red
minus
normalized
green–red
difference
green–blue
red–green
ratio
color
index
extraction,
visible-band-modified
soil-adjusted
modified
red–green–blue
visible-band
index),
space
(red,
green,
blue,
hue,
saturation,
value,
lightness,
‘a’
(Green–Red
component),
‘b’
(Blue–Yellow
component)),
linear
mixing
model
(LMM),
two
machine
learning
algorithms
(a
support
vector
neural
network).
The
results
show
that
following
exhibit
accuracy
across
scenarios:
Otsu–CIVE,
(‘a’:
Green–Red
LMM,
SVM
(Accuracy
>
0.75,
Precision
0.8,
kappa
coefficient
0.6).
Nonetheless,
higher
scene
complexity
entropy
reduce
applicability
precise
methods.
This
study
facilitates
accurate,
efficient
information
within
semiarid
regions,
providing
technical
references
similar
areas.
Frontiers in Plant Science,
Год журнала:
2024,
Номер
15
Опубликована: Ноя. 4, 2024
Vegetation
serves
as
a
crucial
indicator
of
ecological
environment
and
plays
vital
role
in
preserving
ecosystem
stability.
However,
urbanization
escalates
rapidly,
natural
vegetation
landscapes
are
undergoing
continuous
transformation.
Paradoxically,
is
pivotal
mitigating
the
environmental
challenges
posed
by
urban
sprawl.
The
middle
lower
Yangtze
River
Basin
(MLYRB)
China,
particularly
its
economically
thriving
reaches,
has
witnessed
surge
urbanization.
Consequently,
this
study
explored
spatiotemporal
variations
normalized
difference
index
(NDVI)
MLYRB,
with
an
emphasis
on
elucidating
impact
climate
change
dynamics.
results
indicate
that
significant
increasing
trend
NDVI
across
MLYRB
from
2000
to
2020,
pattern
expected
persist.
An
improvement
was
observed
94.12%
prefecture-level
cities
area,
predominantly
western
southern
regions.
Temperature
wind
speed
stand
out
dominant
contributors
improvement.
Nevertheless,
degradation
detected
some
highly
urbanized
central
eastern
parts
mainly
attributed
negative
effects
escalating
Interestingly,
positive
correlation
between
rate
observed,
which
may
be
largely
related
proactive
preservation
policies.
Additionally,
global
climatic
oscillations
were
identified
key
force
driving
periodic
variations.
These
findings
hold
importance
promoting
harmonious
preservation,
thereby
providing
invaluable
insights
for
future
planning
efforts.
Land,
Год журнала:
2024,
Номер
13(11), С. 1840 - 1840
Опубликована: Ноя. 5, 2024
Fractional
vegetation
cover
(FVC)
plays
a
key
role
in
ecological
and
environmental
status
assessment
because
it
directly
reflects
the
extent
of
its
status,
yet
is
an
important
component
ecosystems.
FVC
estimation
methods
have
evolved
from
traditional
manual
interpretation
to
advanced
remote
sensing
technologies,
such
as
satellite
data
analysis
unmanned
aerial
vehicle
(UAV)
image
processing.
Extraction
based
on
high-resolution
UAV
are
being
increasingly
studied
fields
ecology
sensing.
However,
research
UAV-based
extraction
against
backdrop
high
soil
reflectance
arid
regions
remains
scarce.
In
this
paper,
12
visible
light
images
differentiated
scenarios
Ebinur
Lake
basin,
Xinjiang,
China,
various
used
for
high-precision
estimation:
Otsu’s
thresholding
method
combined
with
Visible
Vegetation
Indices
(abbreviated
Otsu-VVIs)
(excess
green
index,
excess
red
minus
normalized
green–red
difference
green–blue
red–green
ratio
color
index
extraction,
visible-band-modified
soil-adjusted
modified
red–green–blue
visible-band
index),
space
(red,
green,
blue,
hue,
saturation,
value,
lightness,
‘a’
(Green–Red
component),
‘b’
(Blue–Yellow
component)),
linear
mixing
model
(LMM),
two
machine
learning
algorithms
(a
support
vector
neural
network).
The
results
show
that
following
exhibit
accuracy
across
scenarios:
Otsu–CIVE,
(‘a’:
Green–Red
LMM,
SVM
(Accuracy
>
0.75,
Precision
0.8,
kappa
coefficient
0.6).
Nonetheless,
higher
scene
complexity
entropy
reduce
applicability
precise
methods.
This
study
facilitates
accurate,
efficient
information
within
semiarid
regions,
providing
technical
references
similar
areas.