Urban Science,
Год журнала:
2025,
Номер
9(5), С. 151 - 151
Опубликована: Май 6, 2025
A
2D
raster
data
representing
building
volumes
of
each
grids
are
derived
from
3D
vector-format
urban
for
use
in
machine
learning
applications.
Since
the
task
is
to
explore
patterns,
i.e.,
heat
islands,
Gaussian
blurring
implemented
on
these
generated
before
training
process.
This
strengthens
visual
capturing
spatial
relationships,
and
as
a
result
correlation
rate
between
air
temperature
volume
also
increased.
After
model
training,
prediction
results
not
simply
evaluated
with
most
widely
used
shallow
metrics
like
Mean
Square
Error
(MSE),
but
thanks
format
input
output
results,
some
image
similarity
such
Structural
Similarity
Index
Measure
(SSIM)
Learned
Perceptual
Image
Patch
(LPIPS)
that
able
detect
consider
relations
during
evaluation
interpretation
process,
because
their
higher
usefulness
mimicking
human
judgements.
The
trained
models
Random
Forest
XGBoost
methods
which
capable
predicting
distribution
by
using
information
compared.
By
doing
so,
this
research
aims
assist
planners
incorporating
environmental
parameters
into
planning
strategies,
thereby
facilitating
more
sustainable
inhabitable
environments.
Geo-spatial Information Science,
Год журнала:
2024,
Номер
unknown, С. 1 - 14
Опубликована: Июнь 6, 2024
An
accurate
and
timely
spatialization
of
electricity
consumption
is
fundamental
for
energy
management
sustainable
urban
development.
While
previous
research
has
relied
heavily
on
statistical
data
or
moderate-resolution
nighttime
light
data,
this
study
presented
a
new
method
by
combining
high-resolution
Luojia
1–01
functional
zoning
information.
The
total
volume
can
be
allocated
to
each
land
use
pixel
based
its
strong
linear
relationship
with
light.
Specifically,
were
connected
the
corresponding
economic
sectors
differentiate
complex
relationships
between
within
different
zones.
digital
number
value
every
multiplied
coefficient.
A
comparisons
have
indicated
that
proposed
more
accurately
characterize
spatial
distribution
consumption.
Our
results
exhibit
much
clearer
outlines
detailed
internal
characteristics.
More
importantly,
information
used
distinguish
various
at
fine
scales.
expected
capture
characteristics
in
manner.
findings
help
local
authorities
formulate
utilization
emission
reduction
strategies.
International Journal of Applied Earth Observation and Geoinformation,
Год журнала:
2024,
Номер
132, С. 104067 - 104067
Опубликована: Авг. 1, 2024
It
is
crucial
to
clarify
the
nonlinear
effects
of
urban
multidimensional
characteristics
on
land
surface
temperature
(LST).
However,
combined
consideration
green
space
(UGS),
water
bodies,
buildings,
and
socio-economic
factors
limited.
And
diurnal
differences
in
their
thermal
have
been
less
considered.
In
this
study,
central
Beijing
was
taken
as
study
area.
Local
climate
zones
(LCZ)
were
firstly
applied
reveal
spatiotemporal
heterogeneity
LST.
Then,
interpretable
machine
learning
methods
utilized
quantitatively
characteristics,
i.e.,
UGS,
building
landscape
features,
features.
The
results
indicated
that
built
type
LCZs
a
higher
average
LST
compared
natural
LCZs.
simultaneously
influenced
by
buildings'
density
height
characteristics.
Daytime
mainly
affected
proportions
trees,
while
nighttime
more
key
exhibit
Whether
during
day
or
night,
impact
coverage
greater
than
height,
consistently
exhibiting
warming
effect.
While,
body
edge
both
exhibited
reversal
trend
between
night.
Our
also
emphasized
importance
trees
UGS
provided
recommendations
for
planning
based
sensitivity
contribution
considerations.
These
findings
can
help
regulate
promote
sustainable
development.
International Journal of Geographical Information Science,
Год журнала:
2024,
Номер
38(11), С. 2183 - 2215
Опубликована: Июль 14, 2024
The
emergence
of
crowdsourced
geographic
information
(CGI)
has
markedly
accelerated
the
evolution
land-use
and
land-cover
(LULC)
mapping.
This
approach
taps
into
collective
power
public
to
share
spatial
information,
providing
a
relevant
data
source
for
producing
LULC
maps.
Through
analysis
262
papers
published
from
2012
2023,
this
work
provides
comprehensive
overview
field,
including
prominent
researchers,
key
areas
study,
major
CGI
sources,
mapping
methods,
scope
research.
Additionally,
it
evaluates
pros
cons
various
sources
methods.
findings
reveal
that
while
applying
with
labels
is
common
way
by
using
analysis,
limited
incomplete
coverage
other
quality
issues.
In
contrast,
extracting
semantic
features
interpretation
often
requires
integrating
multiple
datasets
remote
sensing
imagery,
alongside
advanced
methods
such
as
ensemble
deep
learning.
paper
also
delves
challenges
posed
in
explores
promising
potential
introducing
large
language
models
overcome
these
hurdles.
