Cornea
is
the
most
widely
transplanted
human
tissue.
To
ensure
good
surgical
results
and
effective
visual
recovery,
thorough
screening
of
donors
donor
tissues
necessary
throughout
process
[1].
This
minimizes
risk
transmitting
infectious
diseases
environmental
contaminants
to
recipient,
while
also
guaranteeing
transplant’s
high
optical
functional
quality.
Therefore,
quality
assessment
cornea
at
various
stages
-
from
selection
transplantation
procedure
crucial
its
longevity
effectiveness.
Foods,
Год журнала:
2025,
Номер
14(3), С. 386 - 386
Опубликована: Янв. 24, 2025
Citrus
fruits,
classified
under
the
Rutaceae
family
and
genus,
are
valued
for
their
high
nutritional
content,
attributed
to
rich
array
of
natural
bioactive
compounds.
To
ensure
both
quality
value,
precise
non-destructive
testing
methods
crucial.
Among
these,
computer
vision
spectroscopy
technologies
have
emerged
as
key
tools.
This
review
examines
principles
applications
technologies—including
traditional
vision,
hyperspectral,
multispectral
imaging—as
well
various
techniques,
such
infrared,
Raman,
fluorescence,
terahertz,
nuclear
magnetic
resonance
spectroscopy.
Additionally,
data
fusion
that
integrate
these
discussed.
The
explores
innovative
uses
approaches
in
inspection
grading,
damage
detection,
adulteration
identification,
traceability
assessment.
Each
technology
offers
distinct
characteristics
advantages
tailored
specific
requirements
production.
Through
fusion,
can
be
synergistically
combined,
enhancing
accuracy
depth
assessments.
Future
advancements
this
field
will
likely
focus
on
optimizing
algorithms,
selecting
effective
preprocessing
feature
extraction
developing
portable,
on-site
detection
devices.
These
innovations
drive
industry
toward
increased
intelligence
precision
control.
Abstract
Spectral
imaging
is
a
technique
that
captures
and
analyzes
the
spectral
information
of
an
object,
such
as
its
reflectance,
transmittance,
or
fluorescence.
It
has
been
widely
used
in
various
fields,
remote
sensing,
food
quality
assessment.
In
recent
years,
also
emerged
promising
tool
for
crop
disease
diagnosis,
it
can
provide
rapid,
non-destructive,
accurate
detection
plant
pathogens
symptoms.
This
review
aims
to
concise
overview
principles,
methods,
applications,
challenges
diagnosis.
First,
we
introduce
basic
sensing
concepts
types
imaging,
hyperspectral,
multispectral
imaging.
Second,
discuss
main
steps
techniques
involved
analysis,
image
acquisition,
processing,
feature
extraction,
classification.
Third,
present
some
representative
examples
applications
fungal,
bacterial,
viral,
nematode
infections.
Finally,
highlight
importance
artificial
intelligence
integration
alongside
current
limitations
future
directions
Journal of Modern Optics,
Год журнала:
2024,
Номер
unknown, С. 1 - 13
Опубликована: Ноя. 19, 2024
Image
spatial
resolution
reflects
the
ability
of
an
optical
system
to
capture
detailed
information
about
object.
Compared
Low-resolution
(LR)
images,
High-resolution
(HR)
images
contain
greater
pixel
density
and
textural
detail.
In
experiments,
it
is
difficult
obtain
ideal
HR
due
effects
acquisition
equipment
image
degradation.
To
address
problems,
this
paper
proposes
a
reconstruction
method
based
on
adaptive
sparse
dictionary.
The
experimental
show
that
area
modulation
transfer
function
(MTF)
curve
reconstructed
improved
by
16.41%,
which
represents
increase
in
frequency
components
contains.
Meanwhile
cut-off
from
0.2692cy/pix
0.4018cy/pix.
achieves
good
results
both
efficiency
accuracy.
peak
signal-to-noise
ratio
(PSNR)
20.1650
28.0192.
feature
similarity
(FSIM)
detail
retention
are
above
90%.
