Journal of Function Spaces,
Год журнала:
2022,
Номер
2022, С. 1 - 11
Опубликована: Дек. 31, 2022
This
paper
considers
a
multichannel
deconvolution
model
with
Gaussian
white
noises.
The
goal
is
to
estimate
the
d
-th
derivatives
of
an
unknown
function
in
model.
For
super-smooth
case,
we
construct
adaptive
linear
wavelet
estimator
by
projection
method.
regular-smooth
provide
nonlinear
hard-thresholded
In
order
measure
global
performances
our
estimators,
show
upper
bounds
on
convergence
rates
using
Lp
-risk
(
1≤p<∞
).
Applied Sciences,
Год журнала:
2023,
Номер
13(17), С. 9806 - 9806
Опубликована: Авг. 30, 2023
Addressing
the
issue
of
simultaneous
reconstruction
intensity
and
phase
information
in
multiscale
digital
holography,
an
improved
deep-learning
model,
Mimo-Net,
is
proposed.
For
holograms
with
uneven
distribution
useful
information,
local
feature
extraction
performed
to
generate
different
scales,
branch
input
training
used
realize
learning,
receptive
fields
obtained.
The
up-sampling
path
outputs
simultaneously
through
dual
channels.
experimental
results
show
that
compared
Y-Net,
which
a
network
capable
reconstructing
simultaneously,
Mimo-Net
can
perform
on
three
scales
only
one
training,
improving
efficiency.
peak
signal-to-noise
ratio
structural
similarity
for
are
higher
than
those
Y-Net
reconstruction,
performance.
In
this
article,
we
propose
a
generative
adversarial
network
based
fringe
pattern
normalization
method.
We
investigate
the
method's
effectiveness
under
various
noise
levels
by
evaluating
root
mean
square
error
(RMSE)
and
structural
similarity
index
measure
(SSIM).
Digital Holography and 3-D Imaging 2022,
Год журнала:
2022,
Номер
unknown, С. W2A.7 - W2A.7
Опубликована: Янв. 1, 2022
In
this
article,
we
present
deep
learning
approach
to
estimate
displacement
derivatives
in
digital
holographic
interferometry.
The
results
show
the
capability
of
proposed
method
on
noisy
experimental
fringes.
Journal of Function Spaces,
Год журнала:
2022,
Номер
2022, С. 1 - 11
Опубликована: Дек. 31, 2022
This
paper
considers
a
multichannel
deconvolution
model
with
Gaussian
white
noises.
The
goal
is
to
estimate
the
d
-th
derivatives
of
an
unknown
function
in
model.
For
super-smooth
case,
we
construct
adaptive
linear
wavelet
estimator
by
projection
method.
regular-smooth
provide
nonlinear
hard-thresholded
In
order
measure
global
performances
our
estimators,
show
upper
bounds
on
convergence
rates
using
Lp
-risk
(
1≤p<∞
).