Infrared and visible video fusion method based on local differential correlation features and PID control
Abstract
The
purpose
of
infrared
and
visible
video
fusion
is
to
combine
the
complementary
features
videos
from
different
modalities.
Most
algorithms
ignore
feature
associations
adjacent
frames
guidance
source
process.
Therefore,
this
paper
proposes
a
new
method.
First,
spatial-domain
multi-attribute
detail
extraction
model
proposed,
which
capable
extracting
base
layer,
bright
layer
dark
separately.
Then,
we
propose
an
associated
for
frames,
improves
spatial
continuity
fused
video.
Furthermore,
exponential
homomorphic
filter
simultaneously
increases
dynamic
range
contrast
obtain
general
salient
target
model.
In
stage,
weighted
rule
based
on
edge
intensity
used
in
layer.
Then
design
controller,
transfer
function
measurement
separately,
so
as
construct
closed-loop
proportional-integral-derivative
(PID)
control
system
fuse
ensures
that
maintains
more
information
Experiments
public
datasets
demonstrate
our
method
outperforms
some
state-of-the-art
algorithms.
Research Square (Research Square), Год журнала: 2024, Номер unknown
Опубликована: Май 29, 2024
Язык: Английский