Due
to
the
wide
application
of
welding
in
modern
industry,
effective
detection
weld
surface
defects
is
an
important
measure
ensure
quality
components,
monitor
Service
life
structure,
and
safety
users.
However,
there
are
wrinkles
stains
on
surface,
which
makes
difficult.
Based
dynamic
pulsed
eddy
current
thermography,
a
multi-feature
fusion
algorithm
infrared
features
visible
information
proposed
this
paper.
In
detection,
relative
position
cracks
field
view
constantly
changing,
therefore,
thermal
image
sequences
spatially
aligned
obtain
transient
response
curve
static
mode.
Feature
extraction
dimensionality
reduction
carried
out
time
domain.
The
processed
data
fused
with
features,
classified
pixel-level
applying
pattern
recognition
network.
experimental
results
show
that
can
effectively
suppress
noise
caused
by
texture
stains,
more
clear
accurate
defect
information.
All
21
be
detected,
ability
greatly
improved.
IEEE Transactions on Pattern Analysis and Machine Intelligence,
Год журнала:
2023,
Номер
45(8), С. 10535 - 10554
Опубликована: Март 30, 2023
Visible
and
infrared
image
fusion
(VIF)
has
attracted
a
lot
of
interest
in
recent
years
due
to
its
application
many
tasks,
such
as
object
detection,
tracking,
scene
segmentation,
crowd
counting.
In
addition
conventional
VIF
methods,
an
increasing
number
deep
learning-based
methods
have
been
proposed
the
last
five
years.
Different
types
CNN-based,
autoencoder-based,
GAN-based,
transformer-based
proposed.
Deep
undoubtedly
become
dominant
for
task.
However,
while
much
progress
made,
field
will
benefit
from
systematic
review
these
methods.
this
paper
we
present
comprehensive
We
discuss
motivation,
taxonomy,
development
characteristics,
datasets,
performance
evaluation
detail.
also
future
prospects
field.
This
can
serve
reference
researchers
those
interested
entering
fast-developing
