Improving Rock Type Identification Through Advanced Deep Learning-Based Segmentation Models: A Comparative Study
Applied Sciences,
Год журнала:
2025,
Номер
15(3), С. 1630 - 1630
Опубликована: Фев. 6, 2025
The
accurate
identification
of
rock
types
is
crucial
for
understanding
geological
structures
and
planning
mining
activities.
Therefore,
the
precise
labeling
a
fundamental
requirement
researchers
industry
experts
in
these
fields.
This
study
aims
to
identify
by
segmenting
thin-section
images
using
advanced
deep
learning
models.
Commonly
used
models
such
as
DeepLabv3+,
SegFormer,
ConvNext,
Mask2Former
were
evaluated
compare
performance
different
segmentation
Additionally,
an
improved
version
was
analyzed
enhance
its
performance.
Post-segmentation
enhancements
with
SLIC
super
pixel
morphological
operations
further
boundary
delineation.
An
achieved
validation
accuracy
91.26%
mean
Intersection
over
Union
(mIoU)
82.59%,
representing
improvement
more
than
5%
base
Mask2Former.
Another
significant
aspect
detailed
analysis
other
relevant
dataset.
Quartz
Feldspar
Plagioclase
(QAP)
diagram
utilized
mineral
identification,
achieving
recognition
85.7%.
These
results
indicate
robustness
proposed
approach,
which
exceeds
mIoU
comparable
methods
reported
literature.
significantly
enhances
effectiveness
image
processing
techniques
type
identification.
Furthermore,
comparison
provides
valuable
guidance
selecting
most
suitable
model.
Язык: Английский
Fracture Facies Estimation Utilizing Machine Learning Algorithm and Formation Micro-Imager (FMI) Log
Petroleum Research,
Год журнала:
2025,
Номер
unknown
Опубликована: Май 1, 2025
Язык: Английский
Deformation of the void space of pores and fractures of carbonates: comprehensive analysis of core and field data
Energy Geoscience,
Год журнала:
2024,
Номер
unknown, С. 100364 - 100364
Опубликована: Дек. 1, 2024
Язык: Английский
Determination of the fracture closure pressure in fractural-cavity carbonate reservoirs using a failure criterion based on asperity behavior
Frontiers in Earth Science,
Год журнала:
2024,
Номер
12
Опубликована: Дек. 24, 2024
As
fluid
flow
paths
in
fractural-cavity
carbonate
reservoirs,
fractures
have
a
significant
impact
on
the
production
performance
of
reservoirs.
In
particular,
well
depends
apertures
fractures,
which
vary
with
effective
stress
acting
fractures.
Thus,
predicting
fracture
closure
pressure
is
crucial
for
reservoir
development.
our
research,
pressures
are
derived
using
Zienkiewicz–Pande
failure
criterion,
defines
at
most
asperities
come
into
contact.
The
results
reveal
that
influenced
by
geo-stress
field,
rock
mechanics,
and
spatial
location
fracture.
Ultimately,
typical
wells
located
different
tectonic
zones
Shunbei
Oilfield
calculated,
indicate
significantly
affected
dip
angle
between
strike
maximum
principal
stress.
To
demonstrate
accuracy
estimated
pressure,
corresponding
to
was
evaluated.
It
reveals
flowing
bottom
decreases
rapidly
recoverable
oil
reserves
reduce
when
approaches
pressure.
This
observation
verifies
determined
formula
feasible
predictor
Язык: Английский