Nonlocal Total Variation Based Speckle
Denoising Model
Arundhati Bagchi Misra and Hyeona Lim
Abstract A large range of methods covering various fields of mathematics are
available for denoising an image. The initial denoising models are derived from
energy minimization using nonlinear partial differential equations (PDEs). The
filtering models based on smoothing operators have also been used for denoising.
Among them the recently developed nonlocal means method proposed by Buades,
Coll and Morel in 2005 is quite successful. Though the method is very accurate, it
is very slow and hence quite impractical. In 2008, Gilboa and Osher extended
some known PDE and variational techniques in image processing to the nonlocal
framework and proposed the nonlocal total variation method for Gaussian noise.
We used this idea to develop a nonlocal model for speckle noise. Here we have
extended the speckle model introduced by Krissian et al. in 2005 to the nonlocal
framework. The Split Bregman scheme is used solve this new model.
Keywords Image denoising � Speckle denoising models � Nonlocal means �Nonlocal PDE � Nonlocal TV
A. B. Misra (&)
Department of Mathematical Sciences, Saginaw Valley State University, University Center,
MI 48710, USA
e-mail: [email protected]
H. Lim
Department of Mathematics and Statistics, Mississippi State University, Mississippi,
MS 39762, USA
e-mail: [email protected]
Mohan S. and S. S. Kumar (eds.), Proceedings of the Fourth International Conference
on Signal and Image Processing 2012 (ICSIP 2012), Lecture Notes in Electrical
Engineering 221, DOI: 10.1007/978-81-322-0997-3_46, Ó Springer India 2013
517
1 Introduction
Image restoration is a very important process and is often necessary as a pre-
processing for other imaging techniques such as segmentation and compression. In
general, an observed image f is represented by the equation
f ¼ uþ n; ð1Þ
where u is the original noise free image, f is the observed noisy image and n is the
Gaussian noise. Here u; f : X � I R2 ! I R. For any denoising model, the main
objective is to reconstruct u from an observed image f .
In 1992 Rudin et al. [8] proposed the total variation (TV) denoising model as
the minimization problem:
minu
Z
X
ruj jdx ð2Þ
subject to the constraints,
Z
X
fd x ¼Z
X
u dx ð3Þ
and
Z
X
1
2f ÿ uð Þ2dx ¼ r2; ð4Þ
where r is the standard deviation of the noise n. These constraints ensure that the
resulting image and the observed images are quite close to each other.
Combining the above constraints the TV functional is obtained by
FðuÞ ¼Z
X
ruj jdxþ k
2
Z
X
f ÿ uð Þ2dx: ð5Þ
Here k is a constraint parameter. The equivalent Euler–Lagrange equation gives
the TV denoising model as
ou
otÿr � ru
jruj
� �
¼ kðf ÿ uÞ: ð6Þ
To avoid singularities, it was regularized by using jruj � jreuj ¼ ðu2x þ u2y þ e2Þ1=2.
518 A. B. Misra and H. Lim
2 Preliminaries
Here we present the initial speckle denoising model and the filtering based non-
local means method.
2.1 Speckle Denoising Model
In 2005 Krissian et al. [6] considered the speckle noise model by
f ¼ uþffiffiffi
up
n; ð7Þ
where u is the desired image to find, n is Gaussian noise and f denotes the observed
image. Thus,we have n ¼ f ÿ uffiffiffi
up . Then replacing f ÿ u in (5)with the new expression
for n, the minimization functional for speckle denoising is given by
FðuÞ ¼Z
X
jruj þ k
2
f ÿ uffiffiffi
up
� �2" #
dx: ð8Þ
From energy minimization of this functional, the TV-based speckle denoising
model can be derived as:
ou
otÿ u2
f þ ujreuj r � ru
jreuj
� �
¼ k jreuj ðf ÿ uÞ: ð9Þ
2.2 Nonlocal Means Method
In 2005 Buades et al. proposed the new state of art image denoising algorithm
known as nonlocal means algorithm [2]. The algorithm is given by the formula
NL½u�ðxÞ ¼ 1
CðxÞ
Z
X
eÿðGaIjuðxþ�Þÿuðyþ�Þj2Þð0Þ
h2 uðyÞdy; ð10Þ
where CðxÞ ¼Z
X
eÿðGaIjuðxþ�Þÿuðzþ�Þj2Þð0Þ
h2 dz. Here Ga is the Gaussian kernel with stan-
dard deviation a, h is a filtering parameter and uðxþ �Þ denotes the neighborhoodof the pixel x. Here each pixel value is denoised using the weighted average of all
the pixels in the image. Thus for a given discrete noisy image u ¼ fuðiÞ : i 2 Ig,the estimated value NL½u�ðiÞ, for a pixel i, is computed as
Nonlocal Total Variation Based Speckle Denoising Model 519
NL½u�ðiÞ ¼X
j2Iwði; jÞuðjÞ; ð11Þ
where the weight wði; jÞ depends on the similarity of the i and j pixels and satisfies
the conditions 0�wði; jÞ� 1 andX
jwði; jÞ ¼ 1. In this paper the authors mea-
sured the similarity of a square neighborhoodNi of a fixed size at each pixel i as a
decreasing function of the weighted Euclidean distance, jjuðNiÞ ÿ uðNjÞjj22;a,where j 2 I and a[ 0 is the standard deviation of the Gaussian kernel. The weight
function is defined as
wði; jÞ ¼ 1
CðiÞ eÿ
jjuðNiÞÿuðNjÞjj22;ah2 ; ð12Þ
where CðiÞ is the normalizing constant given by
CðiÞ ¼X
j
eÿ
jjuðNiÞÿuðNjÞjj22;ah2 : ð13Þ
The similar neighborhoods have a very small Euclidean distance which in turn
results in larger weight. Since the nonlocal means uses the similarity neighborhood
for denoising, it works really well for periodic or patterned textured cases. But
though the method is very good in removing noise, it is very slow and hence quite
impractical.
2.3 Nonlocal TV Model
In 2008, Gilboa and Osher extended some known PDE and variational techniques
in image processing to the nonlocal framework [3]. The motivation behind this
was to make any point interact with any other point in the image. Since the
classical derivatives were local operators, they had to redefine the required
operators in [3] following the ideas of Zhou and Scholkopf [10, 11]. Gilboa and
Osher proposed the nonlocal TV (NLTV) model based on the NLTV operator
(R
XjrNLuj), as follows:
ou
ot¼ jNLðuÞ ÿ kðuÿ f Þ
¼Z
X
wðx; yÞðuðyÞ ÿ uðxÞÞðjrNLujÿ1ðxÞ þ jrNLujÿ1ðyÞÞdyÿ kðuÿ f Þ;ð14Þ
where rNLð�Þ ¼ nonlocal gradient, divNLð�Þ ¼ nonlocal divergence and jNLð�Þ¼ divNL
rNL
jrNLj denotes the nonlocal curvature. Since the steepest descent scheme
(14) is very slow, a faster numerical scheme was introduced by Bresson in [1]. The
520 A. B. Misra and H. Lim
scheme is based on the Split Bregman method introduced in [4]. It was proved to
be very fast for regular TV.
3 Nonlocal TV Based Speckle Denoising Model
We now present our new model for speckle denoising. We extend the idea pre-
sented in Sect. 2.3 and develop the following nonlocal TV based speckle denoising
model:
minu
FðuÞ; FðuÞ ¼Z
X
jrNLuj þ kf ÿ uffiffiffi
up
� �2" #
dx ð15Þ
where rNL is a nonlocal gradient defined in [3]. For faster computation, we adopt
Split Bregman scheme [4] for finding a solution to our new model (15). We first
develop in Sect. 3.1 the Split Bregman scheme for Krissian et al. [6].
3.1 Split Bregman Scheme for Krissian et al. Model
As discussed in Sect. 2.1, the minimization functional was given by (8). For Split
Bregman scheme, we introduce d ¼ ru and construct the unconstrained functional
as
minu
FðuÞ; FðuÞ ¼Z
X
jdj þ kðf ÿ uÞ2
udxþ b
2kd ÿruÿ bk22: ð16Þ
Here b is a penalty parameter. We can now split this in two subproblems of u and d
given as:
minu
Z
X
kðf ÿ uÞ2
udxþ b
2kd ÿruÿ bk22; ð17Þ
mind
Z
X
jdjdxþ b
2kd ÿruÿ bk22: ð18Þ
Now,
Nonlocal Total Variation Based Speckle Denoising Model 521
o
oukðf ÿ uÞ2
u
!
¼ kÿ2uðf ÿ uÞ ÿ ðf ÿ uÞ2
u2
" #
¼ kðuÿ f Þð2uÿ uþ f Þ
u2
� �
¼ k
u2ðuÿ f Þðuþ f Þ:
ð19Þ
Therefore the optimality condition for u in (17) gives us
kðuþ f Þu2
ðuÿ f Þ þ b � ðd ÿruÿ bÞ ¼ 0
) kÿ bu2
uþ fD
� �
u ¼ kf ÿ bu2
uþ f� ðd ÿ bÞ:
ð20Þ
The optimality condition can be discretized using the definition of discrete gra-
dient, divergence and Laplacian as discussed in [3]. This gives
kþ 4bu2ij
uij þ fij
!
uij ¼ kfij þ bu2ij
uij þ fij
�
uiþ1;j þ uiÿ1;j þ ui;jþ1 þ ui;jÿ1
þ dx;iÿ1;j ÿ dx;ij þ dy;i;jÿ1 ÿ dy;ij
ÿ bx;iÿ1;j þ bx;ij ÿ by;i;jÿ1 þ by;ij
�
:
ð21Þ
Thus, we get
ukþ1ij ¼ 1
kþ 4bukij
2
ukijþfij
� �
h
kfij þ bukij
2
ukij þ fij
�
ukiþ1;j þ ukiÿ1;j þ uki;jþ1 þ uki;jÿ1
þ dkx;iÿ1;j ÿ dkx;ij þ dky;i;jÿ1 ÿ dky;ij ÿ bkx;iÿ1;j þ bkx;ij ÿ bky;i;jÿ1 þ bky;ij
�i
:
ð22Þ
Since there is no coupling between elements of d, the optimal value of d is
computed using the shrinkage operator described in [7, 9]:
dkþ1ij ¼
rukþ1ij þ bkij
jrukþ1ij þ bkijj
maxfjrukþ1ij þ bkijj ÿ 1=b; 0g: ð23Þ
The variable b is initialized to zero and is updated after each Bregman iteration as:
522 A. B. Misra and H. Lim
bkþ1ij ¼ bkij þrukþ1
ij ÿ dkþ1ij : ð24Þ
The scheme (22)–(24) provides a much faster solution for the TV based speckle
denoising model introduced by Krissian et al. [6].
3.2 Split Bregman scheme for nonlocal TV speckle model
For our new model (15) the Split Bregman functional will be of the form
minu
FðuÞ; FðuÞ ¼Z
X
jdj þ kðf ÿ uÞ2
uþ b
2kd ÿrNLuÿ bk22: ð25Þ
Then the optimality condition for u gives us the equation
kÿ bu2
uþ fDNL
� �
u ¼ kf ÿ bu2
uþ fdivNLðd ÿ bÞ: ð26Þ
Denoting the discretized points x; y 2 X� X by i and j, and using the discrete
definition for divNL and Dbadhbox [1, 3], the discrete minimization scheme is given
as:
kui ÿ bu2i
ui þ fi
X
j
wijðuj ÿ uiÞ !
¼ kfi ÿ bu2i
ui þ fi
X
j
ffiffiffiffiffiffi
wijp ðdij ÿ dji ÿ bij þ bjiÞ
!
:
ð27Þ
Hence, the iteration steps are:
ukþ1i ¼ 1
kþ bu2i
uiþfi
P
j wij
"
bu2i
ui þ fi
X
j
wijukj þ kfi
ÿ bu2i
ui þ fi
X
j
ffiffiffiffiffiffi
wij
p ðdij ÿ dji ÿ bij þ bjiÞ!#
;
ð28Þ
Table 1 Summary of numerical results
Krissian et al. Nonlocal means NLTV
Images Time (s) PSNR Time (s) PSNR Time (s) PSNR
Lenna (PSNR ¼ 24:51) 0:72 26:39 11:46 28:21 4:67 29:10
Gallstones 0:14 ÿ 11:92 ÿ 4:62 ÿ
Nonlocal Total Variation Based Speckle Denoising Model 523
dkþ1ij ¼
ffiffiffiffiffiffi
wijp ðukþ1
j ÿ ukþ1i Þþ bkij
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P
jwijðukþ1j ÿ ukþ1
i Þ2þ bkij2
q max
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
X
j
wijðukþ1j ÿ ukþ1
i Þ2þ bkij2
s
ÿ 1
b;0
0
@
1
A;
ð29Þ
bkþ1ij ¼ bkij þ
ffiffiffiffiffiffi
wij
p ðukþ1j ÿ ukþ1
i Þ ÿ dkþ1ij : ð30Þ
(28–30) provides a fast and accurate nonlocal scheme for TV based speckle
denoising model.
4 Numerical Results
The numerical results are displayed in the next pages. We have considered the
images of size 128� 128 and the summary of results are in Table 1. Other than the
visual results, we have used peak-signal-to-noise ratio (PSNR) for synthetic
images, which was mentioned in [5], to measure the efficiency of the models.A
cleaner image provides a higher PSNR value. We have provided the absolute
residual ðjf ÿ ujÞ and the speckle noise residual�
n ¼ fÿuffiffi
up�
for all images. It is
evident that the new nonlocal speckle model works best among the models
compared here. For this model only 1–2 iterations were enough for all images. The
Krissian et al. is not denoising very well compared to the new one. The residual
shows presence of edges and fine textures due to the blurring effect of TV. But the
Krissian et al. is certainly the fastest one. The nonlocal means is not working well
for fine textures and slower than the new model. It is also the slowest of all.
Nonlocal speckle model maintains texture and removes noise very well. The noise
residuals show that nonlocal speckle model has picked up more noise compared to
the other ones.
524 A. B. Misra and H. Lim
5 Conclusion
In this paper we have incorporated the positive aspects of both PDE and filtering
models. The PDE based TV model is very fast but has a blurring effect. Nonlocal
Ultrasound image result (Gallstones)
526 A. B. Misra and H. Lim
means has very effective denoising properties but it is very slow. Therefore, have
developed an accurate model using both of them. We have extended the existing
Krissian et al. speckle denoising model [6] to a nonlocal framework, and provided
an efficient nonlocal TV based speckle denoising model. We have also developed a
numerical scheme for solving this new model. The numerical scheme is based on
the Split Bregman method introduced by Gilboa and Osher. The results show the
new model is more accurate than both the Krissian et al and nonlocal means
model. It is also much faster than nonlocal means.
References
1. Bresson X (2009) A short note for nonlocal tv minimization. Technical report
2. Buades A, Coll B, Morel JM (2005) A non-local algorithm for image denoising. In: IEEE
computer society conference on computer vision and pattern recognition, CVPR 2005, vol 2,
pp 60–65. doi:10.1109/CVPR.2005.38
3. Gilboa G, Osher S (2008) Nonlocal operators with applications to image processing.
Multiscale Model Simul 7(3):1005–1028. http://dblp.uni-trier.de/db/journals/mmas/
mmas7.html#GilboaO08
4. Goldstein T, Osher S (2009) The split bregman method for l1-regularized problems. SIAM J
Imaging Sci 2:323–343. doi:10.1137/080725891, http://portal.acm.org/
citation.cfm?id=1658384.1658386
5. Kim S, Lim H (2007) A non-convex diffusion model for simultaneous image denoising and
edge enhancement. Electron J Differ Equ Conference 15:175–192
6. Krissian K, Kikinis R, Westin CF, Vosburgh K (2005) Speckle-constrained filtering of
ultrasound images. In: CVPR ’05: Proceedings of the 2005 IEEE computer society
conference on computer vision and pattern recognition (CVPR’05), vol 2. IEEE Computer
Society, Washington, DC, pp 547–552. doi:10.1109/CVPR.2005.331
7. Li C (2003) An efficient algorithm for total variation regularization with applications to the
single pixel camera and compressive sensing. Master’s thesis, Rice University, Houston,
Texas
8. Rudin LI, Osher S, Fatemi E (1992) Nonlinear total variation based noise removal
algorithms. Physica D: Nonlinear Phenom 60(1–4):259–268. doi:10.1016/0167-
2789(92)90242-F
9. Wang Y, Yin W, Zhang Y (2007) A fast algorithm for image deblurring with total variation
regularization. Technical report, Department of Computational and Applied Mathematics,
Rice University, Houston, Texas
10. Zhou D, Schlkopf B (2004) A regularization framework for learning from graph data. In:
ICML workshop on statistical relational learning, pp 132–137
11. Zhou D, Schlkopf B (2005) Regularization on discrete spaces. In: Pattern recognition.
Springer, Heidelberg, pp 361–368
Nonlocal Total Variation Based Speckle Denoising Model 527