Optics Express,
Год журнала:
2024,
Номер
33(2), С. 2940 - 2940
Опубликована: Дек. 3, 2024
Mueller
matrix
polarization
measurement
technique,
as
a
non-invasive
and
label-free,
provides
comprehensive
optical
information
on
polarization-related
structural
characteristics
of
the
measured
target.
It
has
been
widely
applied
in
biomedical,
agricultural,
industrial
fields.
However,
traditional
time-division
modulation
method
requires
multiple
measurements,
which
suffers
from
long
time
susceptibility
to
cumulative
errors
moving
parts.
The
snapshot
spatial
can
capture
target's
interferograms
full
element
images
single
exposure,
but
it
lower
resolution.
To
address
strengths
limitations
both
temporal
modulation,
this
paper
proposes
tempo-spatially
modulated
imaging
polarimeter
(TSM-MMIP).
This
approach
is
based
Stokes
with
modified
Savart
plates
core
device,
allowing
acquisition
16
elements
target
only
four
measurements.
Through
computer
simulation
experimental
platforms,
we
validate
that
similarity
between
input
output
exceeds
0.85,
demonstrates
reliability
feasibility
proposed
method.
In
addition,
use
bee
wing
reveal
potential
technique
analyze
targets
by
extracting
analyzing
key
parameters
matrix.
Transillumination
imaging
is
commonly
used
in
medicine
and
biometrics
to
provide
non-invasive
insights
into
internal
structures.
However,
the
prevalent
image
blurring
resulting
from
scattering
effects
poses
a
significant
challenge
effective
application
of
transillumination.
Artificial
intelligence
algorithms
have
gained
prominence
for
enhancing
transillumination
images
addressing
challenges
such
as
suppression,
depth
estimation,
three-dimensional
structure
reconstruction.
These
advances
require
high-quality
input
optimize
model
performance.
Acquiring
large-scale
dataset
practical
AI
applications
difficult
due
subjective
conditions.
This
study
aimed
overcome
this
obstacle
by
introducing
comprehensive
images.
Methods
generating
depth-dependent
point-spread
function
were
presented.
The
comprised
12,000
pairs
clear
scattered
media,
each
associated
with
corresponding
information.
valuable
advancing
AI-based
solutions
provides
foundation
further
research
on
deblurring,
perception,
3D
Optics Express,
Год журнала:
2024,
Номер
32(25), С. 44252 - 44252
Опубликована: Ноя. 4, 2024
Fourier
ptychography
(FP)
imaging,
drawing
on
the
idea
of
synthetic
aperture,
has
been
demonstrated
as
a
potential
approach
for
remote
sub-diffraction-limited
imaging.
Nevertheless,
farthest
imaging
distance
is
still
limited
to
around
10
m,
even
though
there
significant
improvement
in
macroscopic
FP.
The
most
severe
issue
increasing
field
view
(FoV)
limitation
caused
by
far-field
conditions
diffraction.
Here,
we
propose
modify
condition
rough
reflective
objects,
aiming
overcome
small
FoV
using
divergent
beam
illuminate
objects.
A
joint
optimization
pupil
function
and
target
image
utilized
attain
aberration-free
while
estimating
simultaneously.
Benefiting
from
optimized
reconstruction
algorithm,
which
effectively
expands
camera’s
effective
experimentally
implement
several
FP
systems
suited
distances
12
65
120
m
with
maximum
aperture
200
mm.
thus
improved
more
than
one
order
magnitude
state-of-the-art
works
furthest
distance,
an
over
fourfold
resolution
compared
single
aperture.
Our
findings
demonstrate
advancing
FP,
propelling
it
into
new
stage
development.
Optics Express,
Год журнала:
2024,
Номер
32(20), С. 34678 - 34678
Опубликована: Июнь 17, 2024
Amidst
the
global
health
crisis
sparked
by
coronavirus
pandemic,
proliferation
of
respiratory
illnesses
has
captured
worldwide
attention.
An
increasing
number
individuals
wear
masks
to
mitigate
risk
viral
transmission.
This
trend
posed
a
critical
challenge
for
development
automatic
face
mask
wearing
detection
systems.
In
response,
this
paper
proposed
what
we
believe
is
novel
framework
DOLP-YOLOv5,
which
innovatively
employs
polarization
imaging
enhance
leveraging
unique
characteristics
surfaces.
For
extracting
essential
semantic
details
and
diminish
impact
background
noise,
lightweight
shuffle
attention
(SA)
mechanism
integrated
in
backbone.
Further,
Content-Aware
Bidirectional
Feature
Pyramid
Network
(CA-BiFPN)
applied
feature
fusion,
sufficiently
integrating
information
at
each
stage
improving
ability
presentation.
Moreover,
Focal-EIoU
loss
utilized
bounding
box
regression
improve
accuracy
efficiency
detection.
Benchmark
evaluation
performed
on
self-constructed
(PFM)
dataset
compared
with
five
other
mainstream
algorithms.
The
mAP50-95
DOLP-YOLOv5
reached
63.5%,
3.08%
4.44%
improvements
over
YOLOv8s
YOLOv9s,
achieved
response
speed
384.6f/s.
research
not
only
demonstrates
superiority
detection,
but
also
certain
reference
significance
imaging.
