Artificial intelligence in civil engineering
Elsevier eBooks,
Год журнала:
2024,
Номер
unknown, С. 1 - 74
Опубликована: Янв. 1, 2024
Groundwater potential mapping in arid and semi-arid regions of Kurdistan region of Iraq: A geoinformatics-based machine learning approach
Groundwater for Sustainable Development,
Год журнала:
2024,
Номер
unknown, С. 101337 - 101337
Опубликована: Сен. 1, 2024
Язык: Английский
Predicting construction delay risks in Saudi Arabian projects: A comparative analysis of CatBoost, XGBoost, and LGBM
Expert Systems with Applications,
Год журнала:
2024,
Номер
unknown, С. 126268 - 126268
Опубликована: Дек. 1, 2024
Язык: Английский
Wind farm sites selection using a machine learning approach and geographical information systems in Türkiye
Deleted Journal,
Год журнала:
2025,
Номер
28(1)
Опубликована: Март 27, 2025
Язык: Английский
Using UAVs and Machine Learning for Nothofagus alessandrii Species Identification in Mediterranean Forests
Antonio Cabrera,
Miguel Peralta-Aguilera,
Paula V. Henríquez-Hernández
и другие.
Drones,
Год журнала:
2023,
Номер
7(11), С. 668 - 668
Опубликована: Ноя. 9, 2023
This
study
explores
the
use
of
unmanned
aerial
vehicles
(UAVs)
and
machine
learning
algorithms
for
identification
Nothofagus
alessandrii
(ruil)
species
in
Mediterranean
forests
Chile.
The
endangered
nature
this
species,
coupled
with
habitat
loss
environmental
stressors,
necessitates
efficient
monitoring
conservation
efforts.
UAVs
equipped
high-resolution
sensors
capture
orthophotos,
enabling
development
classification
models
using
supervised
techniques.
Three
algorithms—Random
Forest
(RF),
Support
Vector
Machine
(SVM),
Maximum
Likelihood
(ML)—are
evaluated,
both
at
Pixel-
Object-Based
levels,
across
three
areas.
results
reveal
that
RF
consistently
demonstrates
strong
performance,
followed
by
SVM
ML.
choice
algorithm
training
approach
significantly
impacts
outcomes,
highlighting
importance
tailored
selection
based
on
project
requirements.
These
findings
contribute
to
enhancing
accuracy
remote
sensing
applications,
supporting
biodiversity
ecological
research
Язык: Английский
Development of an Environmental Monitoring System Based on Spatial Marking and Machine Vision Technologies
Journal of the Russian Universities Radioelectronics,
Год журнала:
2023,
Номер
26(4), С. 56 - 69
Опубликована: Сен. 29, 2023
Introduction.
The
use
of
available
satellite
images
and
aerial
photography
by
unmanned
vehicles
(UAVs)
in
the
tasks
environmental
monitoring
is
challenged
imperfection
existing
tools.
Geographic
information
systems
are
characterized
insufficient
flexibility
to
automatically
work
with
heterogeneous
sources.
latest
models
based
on
artificial
intelligence
ecology
require
preliminary
data
preparation.
article
presents
results
designing
a
software
system
for
machine
vision
sensor
data,
which
provides
unification
while
being
flexible
both
terms
sources
methods
their
analysis.
Aim
.
Creation
generalized
coordinated
spatial
marking
tasks.
Materials
Software
engineering
methods,
database
theory
markup
image
processing
methods.
Results
A
method
unifying
was
developed.
analysis
open
from
remote
sensing
Earth,
as
well
UAV
approaches
monitoring.
To
implement
method,
architecture
designed,
model
document-oriented
DBMS
developed,
allows
storing
scaling
procedure.
Conclusion
tools
were
analyzed.
an
architecture,
created.
successfully
implemented
web
interface
Язык: Английский
Landslide Movement of Bendungan District Trenggalek Using an Artificial Neural Network
Environmental Research Engineering and Management,
Год журнала:
2023,
Номер
79(3), С. 95 - 107
Опубликована: Окт. 13, 2023
Landslide
is
one
of
the
disasters
that
often
occurs
in
Indonesia
East
Java
Province,
especially
Bendungan
District,
Trenggalek
Regency.
Analysis
landslide
susceptibility
District
needed
to
spatially
locate
occurrences.
The
purpose
this
study
was
predict
events
using
an
artificial
neural
network.
Rainfall,
topography,
physical
soil
properties,
and
land-use
were
used
as
explanatory
variables.
An
analytic
hierarchy
process
approach
applied
determine
weight
model
satisfactorily
classified
hazards
with
area
under
curve
0.96.
northwest
found
be
a
region
at
high
risk
rainfall
texture
most
influential
parts
triggering
landslides.
Язык: Английский