Journal of Civil Engineering Education,
Journal Year:
2023,
Volume and Issue:
150(2)
Published: Nov. 17, 2023
This
research
describes
a
case
study
of
an
integrated
human–computer
interaction
(HCI)
course
for
construction
engineering
students
using
project-based
learning
and
experiential
cycle
methods.
To
help
keep
pace
with
the
increasing
use
information
technology
(IT)
in
industry,
educational
institutions
have
started
to
add
IT-related
courses
their
civil
curriculum.
However,
these
usually
focus
more
on
fundamental
knowledge
technical
skills
such
as
programming
system
development,
design
HCI,
which
plays
one
most
critical
roles
IT
field,
has
been
rarely
discussed
current
training
programs.
Therefore,
this
developed
HCI
that
focuses
helping
cultivate
nontechnical
skills,
communication
teamwork.
The
was
implemented
graduate-level
at
National
Taiwan
University
Science
Technology.
Based
feedback
collected
from
students,
did
them
identify
real
problems
through
interacting
potential
users
developing
corresponding
solutions
IT.
Students
also
felt
package
gain
tangible
experience
identifying
by
users,
brainstorming
possible
peers,
new
technologies.
Resources Conservation and Recycling,
Journal Year:
2023,
Volume and Issue:
202, P. 107375 - 107375
Published: Dec. 22, 2023
The
growing
environmental
concerns
have
emerged
the
necessity
of
sustainable
waste
management
construction
and
demolition
(C&D)
wastes.
This
review
explores
advancements
in
artificial
intelligence
(AI)
robotics
to
automate
C&D
sorting.
A
comprehensive
examination
this
domain
is
conducted
by
structuring
paper
around
six
research
questions.
Current
trends
potential
future
directions
are
revealed
performing
methodology
data
analysis
involving
bibliometric
scientometric
studies.
Notably,
recent
emphasises
circular
economy,
AI,
robotics,
underscoring
importance
enhance
AI
for
precise
categorisation.
scarcity
publicly
available
datasets
a
central
challenge
domain,
that
hinders
effective
applications.
However,
augmentation,
synthesis,
generative
transfer
learning
been
identified
as
crucial
techniques
dataset
quality
categorization
accuracy.
While
draws
significant
attention
shows
lack
AI-enabled
systems
due
complex
nature
sorting
collection.
In
summary,
study's
findings
highlight
need
new
methods
integrating
multisensory
fusion,
unsupervised
machine
continuously
learn
adapt
streams
materials,
making
them
highly
efficient
management.
Automation in Construction,
Journal Year:
2023,
Volume and Issue:
157, P. 105158 - 105158
Published: Oct. 31, 2023
Low
situational
awareness
contributes
to
safety
incidents
in
construction.
Existing
Deep
Learning
(DL)-based
applications
lack
the
capability
provide
context-specific
and
interactive
feedback
that
is
essential
for
workers
fully
understand
their
surrounding
environments.
This
paper
proposes
Visual
Construction
Safety
Query
(VCSQ)
system.
The
system
encompasses
real-time
Image
Captioning
(IC),
safety-centric
Question
Answering
(VQA),
keyword-based
Image-Text
Retrieval
(ITR),
integrated
with
head-mounted
Augmented
Reality
(AR)
devices.
System
validation
includes
benchmarks
real-world
images.
ITR
module
posted
high
recall
rates
of
0.801
0.835
Recall@5
@10.
VQA
achieved
an
89.7%
accuracy
rate,
IC
had
a
SPICE
score
0.449.
Feasibility
tests
surveys
confirmed
system's
practical
advantages
different
construction
scenarios.
study
establishes
integration
roadmap
adaptable
future
advancements
DL
immersive
AR.
Smart Construction and Sustainable Cities,
Journal Year:
2024,
Volume and Issue:
2(1)
Published: July 2, 2024
Abstract
Digital
visual
data,
such
as
images
and
videos,
are
valuable
sources
of
information
for
various
construction
engineering
management
purposes.
Advances
in
low-cost
image-capturing
storing
technologies,
along
with
the
emergence
artificial
intelligence
methods
have
resulted
a
considerable
increase
using
digital
imaging
sites.
Despite
these
advances,
rich
data
not
typically
used
to
their
full
potential
because
they
processed
documented
subjectively,
several
contents
could
be
overlooked.
Semantic
content
analysis
annotation
enhance
retrieval
application
relevant
instances
large
databases.
This
research
proposes
an
ensemble
approach
use
deep
learning-based
object
recognition,
pixel-level
segmentation,
text
classification
medium-level
(ongoing
activities)
high-level
(project
type)
still
from
outdoor
scenes.
The
proposed
method
can
annotate
without
actors,
i.e.
equipment
workers.
experimental
results
shown
this
annotating
activities
82%
overall
recall
rate.
Alexandria Engineering Journal,
Journal Year:
2024,
Volume and Issue:
99, P. 242 - 256
Published: May 13, 2024
The
widespread
implementation
of
surveillance
systems
on
construction
sites
has
led
to
the
accumulation
vast
amounts
visual
data,
highlighting
need
for
an
effective
semantic
analysis
methodology.
Natural
language,
as
most
intuitive
mode
expression,
can
significantly
enhance
interpretability
such
data.
adoption
multi-modality
models
promotes
interaction
between
video
and
textual
thereby
enabling
managers
swiftly
comprehend
on-site
dynamics.
This
study
introduces
a
Visual
Question
Answering
(VQA)
approach
industry
presents
specialized
dataset
address
unique
requirements
management.
Utilizing
Vision
Transformer
(ViT)
architecture,
proposed
model
conducts
feature
extraction,
fusion
features.
An
additional
projection
layer
is
added
establish
transfer
learning
strategy
that
optimized
site
novel
facilitates
rapid
alignment
language
features
in
validated
through
ablation
studies.
achieves
testing
accuracy
83.8%,
effectively
converting
image
data
from
into
natural
descriptions
processes.
Compared
existing
methods,
this
does
not
rely
object
detection
allows
direct
extraction
deep-level
information
images.
further
discusses
feasibility
applying
VQA
within
engineering
(AEC)
industry,
examines
its
limitations,
offers
suggestions
viable
future
directions
development.
