Machine Vision and Applications,
Journal Year:
2023,
Volume and Issue:
34(5)
Published: Aug. 10, 2023
Abstract
We
consider
a
variational
approach
to
the
problem
of
structure
+
texture
decomposition
(also
known
as
cartoon
decomposition).
As
usual
for
many
problems
in
image
analysis
and
processing,
energy
we
minimize
consists
two
terms:
data-fitting
term
regularization
term.
The
main
feature
our
choosing
parameters
adaptively.
Namely,
is
given
by
weighted
$$p(\cdot
)$$
p(·)
-Dirichlet-based
$$\int
\!a({\varvec{x}}){\,\!|\,\!}\nabla
u
{\,\!|\,\!}^{p({\varvec{x}})}$$
xmlns:mml="http://www.w3.org/1998/Math/MathML">∫a(x)|∇u|p(x)
,
where
weight
exponent
functions
are
determined
from
an
spectral
content
curvature.
Our
numerical
experiments,
both
qualitative
quantitative,
suggest
that
proposed
delivers
better
results
than
state-of-the-art
methods
extracting
textured
mosaic
images,
well
competitive
on
enhancement
problems.
IEEE Transactions on Visualization and Computer Graphics,
Journal Year:
2017,
Volume and Issue:
24(7), P. 2129 - 2139
Published: June 2, 2017
Retrieving
salient
structure
from
textured
images
is
an
important
but
difficult
problem
in
computer
vision
because
texture,
which
can
be
irregular,
anisotropic,
non-uniform
and
complex,
shares
many
of
the
same
properties
as
structure.
Observing
that
a
image
should
piece-wise
smooth,
we
present
method
to
retrieve
such
structures
using
minimization
modified
form
relative
total
variation
metric.
Thanks
characteristics
shared
by
texture
small
structures,
our
effective
at
retrieving
based
on
scale
well.
Our
outperforms
state-of-art
methods
removal
well
scale-space
filtering.
We
also
demonstrate
method's
ability
other
applications
edge
detection,
clip
art
compression
artifact
removal,
inverse
half-toning.
SIAM Journal on Imaging Sciences,
Journal Year:
2018,
Volume and Issue:
11(3), P. 2021 - 2063
Published: Jan. 1, 2018
We
propose
a
new
type
of
regularization
functional
for
images
called
oscillation
total
generalized
variation
(TGV)
which
can
represent
structured
textures
with
oscillatory
character
in
specified
direction
and
scale.
The
infimal
convolution
TGV
respect
to
several
directions
scales
is
then
used
model
texture.
Such
functionals
constitute
regularizer
good
texture
preservation
properties
flexibly
be
incorporated
into
many
imaging
problems.
give
detailed
theoretical
analysis
the
infimal-convolution-type
function
spaces.
Furthermore,
we
consider
appropriate
discretizations
these
introduce
first-order
primal-dual
algorithm
solving
general
variational
problems
associated
this
regularizer.
Finally,
numerical
experiments
are
presented
show
that
our
proposed
models
recover
well
competitive
comparison
existing
state-of-the-art
methods.
IEEE Transactions on Image Processing,
Journal Year:
2020,
Volume and Issue:
30, P. 1542 - 1555
Published: Dec. 15, 2020
Morphology
component
analysis
provides
an
effective
framework
for
structure-texture
image
decomposition,
which
characterizes
the
structure
and
texture
components
by
sparsifying
them
with
certain
transforms
respectively.
Due
to
complexity
randomness
of
texture,
it
is
challenging
design
components.
This
paper
aims
at
exploiting
recurrence
patterns,
one
important
property
develop
a
nonlocal
transform
sparsification.
Since
plain
patch
holds
both
cartoon
contours
regions,
constructed
based
on
such
sparsifies
well.
As
result,
could
be
wrongly
assigned
component,
yielding
ambiguity
in
decomposition.
To
address
this
issue,
we
introduce
discriminative
prior
recurrence,
that
spatial
arrangement
recurrent
patches
regions
exhibits
isotropic
differs
from
contours.
Based
prior,
only
Incorporating
into
morphology
analysis,
propose
approach
Extensive
experiments
have
demonstrated
superior
performance
our
over
existing
ones.
Image Processing On Line,
Journal Year:
2016,
Volume and Issue:
6, P. 75 - 88
Published: May 5, 2016
We
present
in
this
article
a
detailed
analysis
and
implementation
of
the
cartoon+texture
decomposition
algorithm
proposed
[A.
Buades,
J.L.
Lisani,
'Directional
filters
for
color
cartoon
+
texture
image
video
decomposition',
Journal
Mathematical
Imaging
Vision,
2015].
This
method
follows
approach
by
T.
Le,
J-M.
Morel,
L.
Vese,
'Cartoon+Texture
Image
Decomposition',
IPOL
2011],
based
on
low/high-pass
filtering,
but
replaces
isotropic
bank
low-pass
directional
filters.
The
is
obtained
filtering
direction
that
leads
to
largest
local
total
variation
rate
reduction.
permits
improve
performance
near
discontinuities,
where
an
halo
effect
was
produced
previous
method.
IEEE Transactions on Image Processing,
Journal Year:
2018,
Volume and Issue:
28(4), P. 1882 - 1894
Published: Nov. 19, 2018
This
paper
addresses
the
problem
of
cartoon
and
texture
decomposition.
Microtextures
are
characterized
by
their
power
spectrum,
we
propose
to
extract
components
from
information
provided
spectrum
image
patches.
The
contribution
a
patch
is
detected
as
statistically
significant
spectral
with
respect
null
hypothesis
modeling
non-textured
patch.
null-hypothesis
model
built
upon
coarse
representation
obtained
basic
yet
fast
filtering
algorithm
literature.
Hence,
term
"dual
domain":
decomposition
in
spatial
domain
an
input
proposed
approach.
statistical
also
patches
similar
textures
across
image.
approach,
therefore,
falls
within
family
non-local
methods.
Experimental
results
shown
various
application
areas,
including
canvas
pattern
removal
fine
arts
painting,
or
periodic
noise
remote
sensing
imaging.
