基于线激光的轨道门异形曲面点云数据三维重建
刘泽琳 Liu Zelin,
汤嘉兴 Tang Jiaxing,
陈圣怡 Chen Shengyi
и другие.
Laser & Optoelectronics Progress,
Год журнала:
2025,
Номер
62(8), С. 0815004 - 0815004
Опубликована: Янв. 1, 2025
Virtual binocular stereo vision measurement method without phase unwrapping
Optics Communications,
Год журнала:
2025,
Номер
unknown, С. 131869 - 131869
Опубликована: Апрель 1, 2025
Язык: Английский
Binocular stereo vision-based analysis of skid resistance in asphalt mixtures: A study of AC-13, SMA-13, and OGFC-13
Construction and Building Materials,
Год журнала:
2025,
Номер
481, С. 141522 - 141522
Опубликована: Май 2, 2025
Язык: Английский
Unsupervised pavement rutting detection using structured light and area-based deep learning
Automation in Construction,
Год журнала:
2025,
Номер
176, С. 106235 - 106235
Опубликована: Май 12, 2025
Язык: Английский
A Review of Vision-Based Pothole Detection Methods Using Computer Vision and Machine Learning
Sensors,
Год журнала:
2024,
Номер
24(17), С. 5652 - 5652
Опубликована: Авг. 30, 2024
Potholes
and
other
road
surface
damages
pose
significant
risks
to
vehicles
traffic
safety.
The
current
methods
of
in
situ
visual
inspection
for
potholes
or
cracks
are
inefficient,
costly,
hazardous.
Therefore,
there
is
a
pressing
need
develop
automated
systems
assessing
conditions,
aiming
efficiently
accurately
reconstruct,
recognize,
locate
potholes.
In
recent
years,
various
utilizing
(a)
computer
vision,
(b)
three-dimensional
(3D)
point
clouds,
(c)
smartphone
data
have
been
employed
map
quality
conditions.
Machine
learning
deep
techniques
increasingly
enhanced
the
performance
these
methods.
This
review
aims
provide
comprehensive
overview
cutting-edge
vision
machine
algorithms
pothole
detection.
It
covers
topics
such
as
sensing
acquiring
two-dimensional
(2D)
3D
data,
classical
based
on
2D
image
processing,
segmentation-based
using
cloud
modeling,
learning,
algorithms,
hybrid
approaches.
highlights
that
combining
traditional
processing
advanced
offer
highest
accuracy
approaches,
particularly
demonstrate
superior
adaptability
detection
rates,
while
valuable
baseline
techniques.
By
reviewing
evaluating
existing
vision-based
methods,
this
paper
clarifies
landscape
technologies
identifies
opportunities
future
research
development.
Additionally,
insights
provided
by
can
inform
design
implementation
more
robust
effective
condition
assessment,
thereby
contributing
roadway
safety
infrastructure
management.
Язык: Английский
Construction of an Interactive Platform for Piano Teaching in Virtual Reality Environment
Applied Mathematics and Nonlinear Sciences,
Год журнала:
2024,
Номер
9(1)
Опубликована: Янв. 1, 2024
Abstract
Under
the
impact
of
wave
new
technology,
form
piano
teaching
is
no
longer
limited
to
classroom
and
paper
textbooks,
modern
information
such
as
Internet
virtual
reality,
constantly
integrated
into
traditional
teaching.
This
utilizes
Unity3D
engine
create
a
reality
environment
for
interaction,
based
on
VR
construct
an
interactive
platform
Firstly,
binocular
stereo
vision
three-dimensional
reconstruction
scene,
improvement
LeNet5
network
user’s
head
gesture
tracking,
constructed
scene
realizes
real-time
movement
with
human
head.
The
test
set
images
using
CPU-i76700HQ
GTX960
graphics
cards
consume
78ms
25ms
per
frame,
respectively,
which
meets
system
requirements.
In
this
paper,
improved
algorithm
pose
estimation
without
face
detection
applicable
has
high
accuracy
99.31%,
better
than
algorithms
Adaboost+Hough,
LBF+BP,
CNN+Heatmaps,
DCNN,
FSA-Net,
able
satisfy
requirements
scene.
average
rating
more
4
in
attractiveness,
subjectivity,
interactivity,
convenience,
but
running
speed
only
3.46,
so
it
necessary
strengthen
technical
support
improve
software
hardware
level
break
through
development
bottleneck.
Traditional
given
vitality
kinetic
energy
by
environment.
Язык: Английский