2022 7th International Conference on Communication, Image and Signal Processing (CCISP),
Год журнала:
2023,
Номер
unknown, С. 367 - 371
Опубликована: Ноя. 17, 2023
This
research
mainly
involves
the
multidimensional
imaging
technology
of
tumor
lesions
removal,
using
SALAM
3D
algorithm,
set
graphics
multi-view
geometric
convolutional
neural
image
edge
classification
pattern
recognition
interaction
stereo
algorithm
and
vision
algorithm.
It
achieves
2D
to
through
point
cloud,
rendering
quality
update,
Generative
discriminative
observation
model,
Transformer
network
architecture,
Unreal
other
methods,
then
uses
point-based
multimodal
fusion
method.
In
this
study,
multi-arm
mechanical
principle
is
adopted
form
a
virtual
lesion
area
effect
cancer
realized
after
adopted.
Applied Optics,
Год журнала:
2024,
Номер
63(12), С. 3219 - 3219
Опубликована: Март 25, 2024
This
paper
presents
an
adaptive
focus
stacking
method
for
large
depth-of-field
(DOF)
3D
microscopic
structured-light
imaging
systems.
Conventional
methods
typically
capture
images
under
a
series
of
pre-defined
settings
without
considering
the
attributes
measured
object.
Therefore,
it
is
inefficient
since
some
might
be
redundant.
To
address
this
problem,
we
first
employ
focal
sweep
technique
to
reconstruct
initial
rough
shape
objects.
Then,
leverage
data
determine
effective
that
camera
on
valid
areas
Finally,
high-quality
point
cloud
using
fringe
obtained
from
these
by
stacking.
Experimental
results
demonstrate
success
proposed
method.
Imaging Neuroscience,
Год журнала:
2024,
Номер
2, С. 1 - 22
Опубликована: Июнь 1, 2024
Abstract
Despite
the
interest
in
studying
and
quantifying
structural
integrity
of
myelin
postmortem
brain
tissue,
current
methods
for
high-resolution
imaging
with
optical
microscopy
are
not
sufficient.
While
must
have
adequate
resolution
sensitivity
to
detect
microstructural
alterations
that
relevant
aging
neurodegenerative
disease,
an
equally
critical
aspect
is
minimize
damage
induced
during
tissue
processing
steps.
Birefringence
(BRM)
a
powerful
technique
leverages
anisotropy
provide
detailed,
label-free
images
at
any
diffraction-limited
resolution,
while
maintaining
simple
low-cost
setup.
Building
on
our
previous
work,
we
developed
new
BRM
system
image
pipeline
enable
efficient,
high-throughput
structure
multiple
scales.
Here,
utilize
this
systematically
assess
by
several
common
steps
sections
from
rhesus
monkey.
Images
taken
same
myelinated
axons,
before
after
each
step,
direct
evidence
mishandling
sample
preparation
can
cause
significant
myelin.
First,
report
key
advancements
system,
procedure,
pipeline,
which
increases
speed
efficiency
BRM.
These
include
integrating
fast
piezoelectric
rotational
stages,
minimizing
number
required
(to
three
images)
determining
birefringence
parameter
maps,
implementing
analytical
solution
directly
maps.
Second,
using
demonstrate
effective
requires
(1)
avoidance
prolonged
drying
or
dehydration
(2)
selection
optimal
mounting
medium
(85%
glycerol),
(3)
permeabilization
detergents
(i.e.,
Triton
X-100
Saponin),
(4)
suitable
tissue-section
thickness
(15,
30
60
µm)
based
region
interest.
In
addition
serving
as
guide
users
interested
myelin,
these
basic
experiments
highlight
very
sensitive
changes
underlying
lipid
suggest
optimized
studies
breakdown
disease.
show
leading
method
detailed
characterization
highly
inadequate
imaging,
has
previously
been
properly
characterized
For
most
effective,
structure,
should
be
kept
minimum,
prevented
mounted
85%
glycerol.
With
proper
preservation
provides
exquisitely
facilitate
assessment
pathology
associated
injury
Optics Letters,
Год журнала:
2023,
Номер
48(16), С. 4348 - 4348
Опубликована: Июль 24, 2023
This
Letter
presents
a
novel,
to
the
best
of
our
knowledge,
method
calibrate
multi-focus
microscopic
structured-light
three-dimensional
(3D)
imaging
systems
with
an
electrically
adjustable
camera
focal
length.
We
first
leverage
conventional
system
reference
length
f0.
Then
we
other
discrete
lengths
fi
by
determining
virtual
features
on
reconstructed
white
plane
using
Finally,
fit
polynomial
function
model
calibration
results
for
fi.
Experimental
demonstrate
that
proposed
can
consistently
and
accurately.
Applied Physics Letters,
Год журнала:
2024,
Номер
124(2)
Опубликована: Янв. 8, 2024
Capturing
details
of
objects
beyond
the
focal
plane
is
challenging
due
to
limited
depth-of-field
(DoF)
optical
systems.
Here,
we
report
a
computational
refocusing
ghost
Imaging
(CRGI)
method
extend
DoF
imaging
(CGI)
An
ultra-fast
and
in
situ
point
spread
function
(PSF)
estimation
put
forward
utilizing
characterization
system
compressive
sensing
modulation.
The
PSF
distribution
measured
with
algorithm
according
reciprocity
property
using
same
CGI
system.
convolution
PSFs
various
depths
modulation
patterns
reshaped
into
measurement
matrices
computationally
refocus
at
different
depths.
From
one
measurement,
CRGI
can
rebuild
distinct
well-focused
images
multiple
According
experiments,
nearly
quadruple
typical
methods.
represents
significant
advancement
domain
by
surpassing
limitations.
This
discovery
enables
recording
object
features
focus
extended
depth-of-field.
Applied Optics,
Год журнала:
2023,
Номер
62(27), С. 7299 - 7299
Опубликована: Сен. 5, 2023
Microscopic
fringe
projection
profilometry
(MFPP)
technology
is
widely
used
in
3D
measurement.
The
measurement
precision
performed
by
the
MFPP
system
closely
related
to
calibration
accuracy.
However,
owing
shallow
depth
of
field,
frequently
influenced
low-quality
target
images,
which
would
generate
inaccurate
features
and
parameter
estimates.
To
alleviate
problem,
this
paper
proposes
an
unsupervised-learning-based
robust
defocus
noise,
could
effectively
enhance
image
quality
increase
In
method,
first,
unsupervised
deblurring
network
(UIDNet)
developed
recover
a
sharp
from
deteriorated
one.
Free
capturing
strictly
paired
images
specific
vision
or
generating
dataset
simulation,
deep
learning
framework
can
learn
more
accurate
multi-quality
convenient
acquisition.
Second,
multi-perceptual
loss
Fourier
frequency
are
introduced
into
UIDNet
improve
training
performance.
Third,
compensation
strategy
based
on
2D
discrete
transform
also
evaluate
detection
accuracy
reference
feature
centers
for
fine
calibration.
relevant
experiments
demonstrate
that
proposed
method
achieve
superior
performance
terms
precision.
Non-contact
high-precision
three-dimensional(3D)
perception
of
tiny
targets
is
widely
required
in
precision
machining,
micro-assembly,
quality
inspection
and
other
fields.
This
paper
studies
the
depth-of-field
(DOF)
extension
method
microscopic
structured
light
system
based
on
image
fusion.
Firstly,
a
hardware
device
with
electrically
tunable
lens
(ETL)
introduced,
overall
framework
using
this
to
capture
sequential
fringe
images
three-dimensional
imaging
given.
In
order
realize
DOF
expansion,
key
principles
commonly
used
spatial
domain
frequency
fusion
technologies
are
analyzed,
effect
tested
respectively.
Finally,
according
actual
performance,
Laplacian
pyramid
high
accuracy
selected
for
3D
imaging.
Experimental
results
show
that
can
better
reconstruct
two
objects
at
different
height
positions
scene
compared
single-focus
case,
measured
greater
than
10mm
under
measurement
field
view
11mm
(W)
×7mm
(H).
