HHS Public Access

Author manuscript
Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
Author Manuscript

Published in final edited form as:

Inf Sci (Ny). 2017 December ; 418-419: 61–73. doi:10.1016/j.ins.2017.06.042.

An active contour model based on local fitted images for image

segmentation
Lei Wanga,b, Yan Changa, Hui Wanga, Zhenzhou Wua, Jiantao Pub,*, and Xiaodong Yanga,*
aMedicalImaging Department, Suzhou Institute of Biomedical Engineering and Technology,
Chinese Academy of Sciences, Suzhou, 215163, China
bDepartments of Radiology and Bioengineering, University of Pittsburgh, Pittsburgh, 15213, USA
Author Manuscript

Abstract
Active contour models are popular and widely used for a variety of image segmentation
applications with promising accuracy, but they may suffer from limited segmentation
performances due to the presence of intensity inhomogeneity. To overcome this drawback, a novel
region-based active contour model based on two different local fitted images is proposed by
constructing a novel local hybrid image fitting energy, which is minimized in a variational level set
framework to guide the evolving of contour curves toward the desired boundaries. The proposed
model is evaluated and compared with several typical active contour models to segment synthetic
and real images with different intensity characteristics. Experimental results demonstrate that the
proposed model outperforms these models in terms of accuracy in image segmentation.
Author Manuscript

Keywords
Image segmentation; Active contour models; Local fitted image; Level set

1. Introduction
Since the introduction by Kass et al. [12], active contour models [8] have become very
popular in the field of image segmentation in the past years, and can yield closed and
smooth contours of the desired objects with promising accuracy [4]. These models are in
general characterized by a predefined energy functional with intensity and/or gradient
information based on the intrinsic properties of images. This energy functional can be
minimized by using the level set method [17], where contour curves are represented
Author Manuscript

implicitly as the zero level set of higher dimensional function. This energy minimization
procedure is effective [6] to drive the motion of contour curves towards the boundaries of
target objects and achieve satisfactory segmentation results for images with different
intensity characteristics.

In the past, a variety of active contour models have been proposed for image segmentation
purpose [1,2,7]. They can be coarsely categorized into the edge- [18,45] and region-based

*
Corresponding authors. [email protected] (J. Pu), [email protected] (X. Yang).
 Wang et al. Page 2

models [13–15,19,20,25,29] according to image information used in the energy functional.

Author Manuscript

Edge-based active contour models usually utilize intensity gradients to approximate the
target boundaries and attract contour curves toward desirable boundaries [11]. This type of
models is efficient but sensitive to image noise and the initial curve placement due to the
local characteristic of image gradients. Different from this type of models, region-based
models employ a great deal of image information, such as pixel intensities, shape and texture
features in global or local regions, to make contour curves accurately converge to desirable
target boundaries. The region-based models are in general more robust to image noise, less
sensitive to the initial contour curve placement, and more capable of dealing with weak
boundaries and intensity inhomogeneity owing to the use of more image information, which
in turn causes more segmentation time.

Among region-based models, the Chan–Vese (CV) model [2] has been widely used for many
image segmentation tasks. It assumes that images to be processed have homogeneous
Author Manuscript

intensities and can be partitioned into several disjoint subregions according to intensity
characteristics. However, this model may not work well for images with intensity
inhomogeneity due to the mere use of global intensity averages. To accurately segment
inhomogeneous images, a number of region-based models [14,15] have been developed by
constructing a proper energy functional to drive the evolution of contour curves. Li et al.
[14] and Zhang et al. [44] improved the CV model by using the local binary fitting (LBF)
energy and local image fitting (LIF) energy, respectively. Ali et al. [1] developed a
variational model from multiplicative and difference images to handle intensity
inhomogeneity, which turned out to outperform the CV model and its many variants [36].
Mabood et al. [21] proposed an absolute median deviation based model for noisy images,
which was accurate and efficient as compared to the local Chan–Vese (LCV) model [36].
More models can be found in [9,16,30,32] by defining energy functional with other
Author Manuscript

information, such as texture features [22,46], prior shape information [3,23,40] and local
patch [35]. With proper initialization, these models are able to successfully extract the
desirable objects depicted on images, but typically need complicated estimation strategies
[10,31].

Another strategy to accurately segment the objects in an image is to incorporate feature

information into the energy functional to reduce the adverse influence of intensity
inhomogeneity. There are a variety of feature information [5,7,21] used in segmentation, but
most of them are not applicable for inhomogeneous images and cannot effectively highlight
desirable objects located in inhomogeneous regions. Hence, it is desirable to design novel
information for accurate segmentation.
Author Manuscript

In this study, we proposed to construct a local hybrid image fitting (LHIF) energy functional
for accurate image segmentation by leveraging the strength of both LBF and LIF models.
Specifically, a novel local fitted image was derived from the LBF and LIF models, aiming to
approximate the square of local image and called square fitted image (SFI). Using this SFI
image, a novel local intensity fitting energy was constructed, together with a regularization
term of level set function. Minimizing this energy functional in a variational level set
formulation will drive contour curves toward the boundaries of desirable objects in an
optimal way. The developed model was validated using a number of synthetic and real

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 3

images as well as medical images. A comparison with several typical active contour models
Author Manuscript

was also performed.

2. Related work
2.1. CV model
The CV model can be viewed as a variant of the Mumford–Shah model [16] based on the
assumption of image homogeneity. It partitions a given image into two disjoint subregions
with an initial contour curve C (i.e., inside and outside regions of C, expressed as Cin and
Cout, respectively). The initial curve evolves to the desired boundaries according to the
following energy functional:

(1)
Author Manuscript

where constants c1 and c2 are the average intensities for the regions Cin and Cout,
respectively; I(x) denotes the pixel intensity in the position of x. λ1 and λ2 are weighting
parameters for intensity fitting terms, which play a major role in driving the motion of the
contour curve; while υ is the weighting parameter for length penalty term |C|. In calculus of
variations, the level set formulation of the CV model is obtained by minimizing this energy
functional with respect to the level set function ϕ and given by as follows:

(2)
Author Manuscript

where δ(·) is the Dirac function; div(·) and ∇ are the divergence and gradient operator,
respectively. c1 and c2 are computed according to the following formulations:

(3)

where H1(ϕ) = H(ϕ) and H2(ϕ) = 1 − H(ϕ) are used to identify the internal and external
regions of ϕ; is the smooth Heaviside function [39] with a small
positive constant ε, and its derivative is the Dirac function .

In the CV model, the differences between c1 and c2 should be large enough to attract contour
Author Manuscript

curves to the desired boundaries according to the assumption of image homogeneity. When
image intensities are not homogeneous, c1 may be approximately equal to c2 in certain
regions, which means that the model loses the capability of identifying the desired objects
and image background, together with boundary leakages and inadequate segmentation.

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 4

2.2. LBF model

Author Manuscript

To overcome the drawbacks of the CV model, the LBF model [14] was proposed by
introducing a binary fitting energy in a local region specified by a Gaussian kernel, and this
LBF energy can be given by:

(4)

where is a Gaussian kernel with standard deviation σ and used

to define a local region centered at x; y is a pixel coordinate in this local region; f1(x) and
f2(x) are spatially varying intensity averages for the internal and external regions of contour
curve C, which are quite different from constants c1 and c2 of the CV model due to the
Author Manuscript

introduction of Kσ(x, y). The other symbols have the same meanings as those mentioned
above in the CV model.

The presence of Gaussian kernel Kσ(x, y) enables the LBF model to identify small intensity
differences in local regions and have the capability of handling intensity inhomogeneity.
Meanwhile, this model can avoid the time-consuming reinitialization step by introducing a
new regularization term to penalize the deviation from a signed distance function [18].
However, this model relies on the initial curve placement to avoid the local minimums of the
energy functional in the presence of severe intensity inhomogeneity due to the localization
property of Gaussian kernel and simple intensity statistical estimation.

2.3. LIF model

Author Manuscript

The LIF model [44] was proposed by introducing the local image fitting energy with the
local intensity averages mi, i=1,2 based on the piecewise-smooth assumption that image I
can be simply reconstructed by the internal region of m1 and the external region of m2 in a
local region. The LIF energy can be regarded as a constraint of the differences between the
local fitted image and the original image, and expressed as:

(5)

(6)
Author Manuscript

(7)

where Ω is the whole image region; ILFI denotes the local fitted image (LFI) for the
approximation of the image I in a local region; m1 and m2 are intensity averages in this

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 5

region defined by Gaussian window Wk with the stand deviation of σ and the radius of k.
Author Manuscript

This model is able to achieve similar segmentation accuracy as the LBF model, and needs
less computational time due to the presence of Gaussian smoothing for the level set function
ϕ. However, the LIF model tends to ignore some details in the segmentation of small objects,
because of the usage of the Gaussian smoothing filter, and thereby may suffer from
inadequate image segmentation in the presence of intensity inhomogeneity.

3. Proposed method
To obtain reasonable segmentation performances for inhomogeneous images, we proposed
to construct a novel energy functional by incorporating other valuable information. We
noticed that a local region can not only be divided into two disjoint parts with fi(x) in the
LBF model, but also reconstructed approximately with mi in the LIF model. This means that
the LBF and LIF models can be combined together for image segmentation purpose. The
Author Manuscript

combination will result in a novel local fitted image, called square fitted image (SFI) and
quite different from the LFI image, and a novel active contour model, constructed by
defining a local hybrid image fitting (LHIF) energy based on images ISFI, ILFI and I. The SFI
image and LHIF energy are explained in detail in the following.

3.1. The SFI image

In the LIF model, a local region can be approximated by a local fitted image with the
formulation of ILFI(x) = m1H1(ϕ) + m2H2(ϕ), where m1 and m2 are intensity averages in the
local neighborhood specified by Gaussian window. The average mi equals to fi(x) in the LBF
model (i.e., m1 = f1(x) and m2 = f2(x)) when setting the same Gaussian parameters, by
which image ILFI can be incorporated into the LBF model. To this end, we set the same
Gaussian parameters for mi and fi(x), with λ1 = λ2 = 1, and then the LBF fitting energy can
Author Manuscript

be expanded as:

(8)

The minimization of the LBF fitting energy is equivalent to approximately minimizing the
formulation of in a local neighborhood centered
Author Manuscript

at x, largely ignoring the effect of Gaussian kernel Kσ(x, y), which is mainly used to specify
a local neighborhood. Due to I ≈ ILFI in a local region, this approximation formulation can
be further expressed as , leading to another fitted image
, called square fitted image ISFI. Hence, the minimization of the LBF
fitting energy can be obtained when local intensity average mi satisfying the following two
conditions, i.e., I(x) ≈ ILFI(x) and I2(x) ≈ ISFI(x).

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 6

3.2. The LHIF fitting energy

Author Manuscript

To take advantage of two local fitted images ISFI and ILFI, a local hybrid image fitting
(LHIF) energy was proposed by quantifying the image differences between original image I
and its two approximated versions, i.e., ISFI and ILFI in terms of Kullback–Leibler
divergence, which can be expressed as:

(9)
Author Manuscript

(10)

where λ1 and λ2 are the weighting parameters for the first and second integral terms; ILFI(x)
and ISFI(x) are used to approximate images of I and I2, respectively.

Note that images I and I2 can be expressed as

(11)
Author Manuscript

where DLFI(x) and DSFI(x) are the residual errors for the approximations of I and I2 and
correlated according to the formulation

(12)

Here, we do not focus on the relationship between these two errors, but on how to minimize
the LHIF fitting energy because when both DLFI(x) and DSFI(x) are small enough and
approach zero, this energy functional can be successfully minimized in theory.

To obtain smooth contour curves, two commonly used regularization terms

and L(ϕ) = ∫|∇H(ϕ)|dx [18] were introduced into our model,
Author Manuscript

resulting in the final fitting energy functional given by:

(13)

where μ and υ are the weighting parameters for the signed distance function P(ϕ) and the
length of level set function L(ϕ), respectively.

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 7

This energy functional is minimized by standard gradient descent [41,42]: for a fixed ϕ, E(ϕ)
Author Manuscript

is minimized with respect to fi(x) or mi, which is given by:

(14)

For fixed fi(x), E(ϕ) is minimized with respect to ϕ based on the gradient descent flow (see
Appendix A)

(15)
Author Manuscript

(16)

(17)

4. Experiments
Author Manuscript

To demonstrate the performance of the developed mode, we applied it to segment both

synthetic and real images in a variational level set framework. ϕ was initially set to a positive
constant c0 outside a region and −c0 inside; and then computed iteratively by: ϕn+1 = ϕn + Δt
· Δϕn; where Δt is time-step and n is iteration number. Gaussian kernel Kσ was truncated by
a (2k + 1)(2k + 1) window, where k has been recommended for the smallest odd number no
less than 2σ [14,15,34]. Other parameters were set by default as follows: σ = 3, c0 = 2, λ1 =
1.0, λ2 = 0.1, Δt = 0.1, μ = 1, ε = 0.1 and υ = 0.001 × 255 × 255. The tests were conducted
in Matlab R2013a programming environment on a PC with 3.3 GHz Intel Core system and
8GB RAM.

4.1. Segmentation of synthetic and real images

Fig. 1 shows the segmentation results of the proposed model for both synthetic and real
Author Manuscript

images, which are inhomogeneous. Three randomly chosen initial rectangular contour
curves shown in the first row, can evolve successfully to the boundaries of the desired
objects with different sizes and shapes. According to these evolution results, it can be seen
that our proposed model is capable of accurately segmenting the desired objects from noisy
and inhomogeneous images. Also, our model needs relatively few iterations to drive the
contours to desirable boundaries.

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 8

4.2. Segmentation of medical images

Author Manuscript

Fig. 2 shows the segmentation results when applying the developed model to four clinically
medical images with both intensity inhomogeneity and complex image background. As
demonstrated by these examples, our proposed model can accurately identify the boundaries
of all targeted objects based on randomly chosen rectangle initial curves, while successfully
excluding a great deal of undesired background information due to the existence of two
different local information from images ILFI and ISFI. Both of these two fitted images are
capable of effectively highlighting the desirable objects with different texture characteristics,
and significantly reduces the adverse influence of complicated image background in image
segmentation. ISFI demonstrates a unique strength in emphasizing the intensity differences
between the foreground and background of images to be segmented, which can be partly
verified according to the two estimated images of DLFI and DSFI.
Author Manuscript

4.3. Comparative evaluation

We also compared the developed model with region-based active contour models 1 (i.e., the
CV model, LBF model, LIF model, local Gaussian distribution fitting (LGDF) model [33],
local region-based CV model (LRCV) [20] and cross entropy based model (CEM) [27]) to
display their performance differences. The comparison of these models was presented in Fig.
3, where the initial green contour curves and their final evolution results were displayed in
different rows, respectively. These segmentation results can effectively demonstrate their
performances in terms of handling severe intensity inhomogeneity, extracting multiple target
objects, and excluding complex image background. Specifically, the CV and CEM models
cannot handle intensity inhomogeneity and had the worst segmentation performances among
these models; and the LBF, LIF, LGDF, and LRCV models can, to some extent, extract the
desirable objects from those regions with severe intensity inhomogeneity, but had the
Author Manuscript

inadequate or excessive segmentation problems in the neighborhood nearby the target

boundaries. Different from above these models, our proposed model achieved better
segmentation performances in terms of accuracy for all these four images based on the same
initial contour curves.

We also quantitatively assessed the performance of these models using the Dice Similarity
Coefficient (DSC) [31] as defined as:

(18)

where ∩ is the intersection operator; N (·) is the number of pixels enclosed set. A is
Author Manuscript

segmentation results of a given algorithm; B is the ground truths, which in general

approximately replaced by the manual segmentation. The value of DSC ranges from 0 to 1,
and the higher the value of DSC is, the more accurate the result of the algorithm is.

1The Matlab source code of LBF, LIF and LGDF algorithms can be found from http://www.engr.uconn.edu/~cmli/, http://
www4.comp.polyu.edu.hk/~cslzhang/ and http://www.unc.edu/~liwa/, respectively.

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 9

According to this metric, the segmentation results of these models, based on images
Author Manuscript

displayed in Figs. 1 and 2, were quantified and shown in Tables 1 and 2, respectively.

The quantitative assessments demonstrate that our model has the largest values of DSC for
all test images, the CV and CEM models have the lowest values of this measure due to
absence of local image information. Although the LBF, LIF, LGDF and LRCV models can
segment the desired objects in the presence of intensity inhomogeneity, they are less
accurate than our proposed model. In addition, the computational cost of these models in our
experiments were given in Table 3 for images displayed in Fig. 1. Although our model need
more segmentation time than the CV model for these images due to the existence of various
local information, it is more efficient than the LBF, LIF, LGDF and LRCV models. As for
the CEM model, it is based on the same global information as the CV model, but its
segmentation time varies dramatically with images. In summary, our model has better
segmentation performances than the global and local region-based models, considering both
Author Manuscript

computational time and accuracy.

We further compared our developed model with available local region-based models (i.e.,
the LBF, LIF, LGDF, and LRCV models) based on publicly available Berkeley segmentation
data set 500 (BSDS500) [24,28]. This dataset consists of a number of natural images and
their ground truth manually generated by different individuals. The performance was
quantitatively evaluated using the probabilistic rand index (PRI) [26], which was given by:

(19)
Author Manuscript

where B = {B1, B2, ···, BM} is M ground truths, A = {x1, x2, ···, xN} is the result of a
specified algorithm where a subscript indexes one of N pixels. sij is a membership function,
and set to 1 when pixels i and j belong to the same cluster, and otherwise to 0. qij is the
ground truth probability of pixels i and j belonging to the same cluster. The PRI takes a
value between 0 and 1, with a higher value representing a more accurate segmentation. With
the PRI, the comparative results of all models were displayed in Table 4. 2 As demonstrated
by the results, our model has the highest PRI values for 12 images, along with highest mean
of the PRI.

5. Discussion
5.1. Parameters analysis
Although our proposed model demonstrated a relatively high performance in image
Author Manuscript

segmentation, it is not easy to achieve an optimal tradeoff among different weighting

parameters. Most of these parameters have been analyzed and discussed in previous studies
[33] in detail. Here, we only focus on parameters λ1 and λ2, which are mainly used in our
model to keep a tradeoff between the two intensity integral terms derived from ILFI(x) and
ISFI(x), respectively. The first integral term tends to segment the objects with different

2The segmentation results of the LBF, LIF, LGDF and LRCV models have already been obtained by Ji et al. [10].

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 10

intensity characteristics, including certain background regions nearby the desired

Author Manuscript

boundaries; while the second term highlights to some extent the differences of image
intensities in local regions, consequently making it difficult to extract undesired background
regions. To correctly segment desired foreground and avoid undesired background, it is
advisable to set proper values for λ1 and λ2 according to the characteristics of the image
itself, including intensity inhomogeneity, noise level and image contrast. For example, (λ1,
λ2) are set to (0.85, 0.1), (1.0, 0.1) and (1.0, 0.25) respectively for the segmentations of
images shown in Fig. 4. Their segmentation results demonstrate that our proposed model is
relatively stable when solely changing the parameter λ1 in a small range, but unstable when
changing the parameter λ2 under the condition of fixing the other parameters. This means
that the first integral term plays greater important part in image segmentation than the
second term; and is more tolerant for the changes of its corresponding weighting parameter.

5.2. Sensitivity to initialization and noise

Author Manuscript

To estimate our model in terms of sensitivity to initial curve placements and robustness
against noise, it was tested, by keeping all the parameters fixed, for a medical image
displayed in Fig. 5 with different initial curves; and then compared with the LBF, LIF,
LGDF and LRCV models based on the same initial curve for another medical image shown
in Fig. 6, which was normalized and corrupted by Gaussian white noise with zeros mean and
variances of 0.005, 0.01, and 0.015, respectively. The red evolution results presented in Fig.
5 demonstrate that our model has nearly the same results for different initial curves, and
deteriorated performances as noise variances increase according to Fig. 6. These noises
cause some undesirable regions are extracted, together with ignoring small details of target
objects; but these adverse segmented components of our model are tiny compared with the
whole segmentation results. This means that our model is insensitive to initial curve
placements due to the local information from both ILFI and ISFI, and robust to some extent
Author Manuscript

against Gaussian white noise in segmentation, relative to other existing models shown in
Fig. 6.

6. Conclusions
In this study, a novel region-based active contour model was developed for accurate image
segmentation. Its novelty lies in the introduction of a novel local hybrid region intensity
fitting energy based on the LBF and LIF models, and this energy functional was constructed
by using the original image and its two local fitted versions drawning upon the concept of
Kullback–Leibler divergence, and minimized in a variational level set framework to drive the
motion of the contour curves to desirable boundaries. Experiments demonstrate that our
proposed model is capable of segmenting synthetic and real images, as well as clinically
Author Manuscript

medical images, and has a relatively higher performance as compared to available region-
based models.

Acknowledgments
This work is supported, in part, by Chinese Academy of Sciences Research, Development Equipment Project (Grant
No.YZ201313, YZ201445), National Natural Science Foundation (Grant No.11505281), National Institutes of
Health (NIH) (Grant No.R21CA197493, R01HL096613).

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 11

References
Author Manuscript

1. Ali H, Badshah N, Chen K, Khan G. A variational model with hybrid images data fitting energies
for segmentation of images with intensity inhomogeneity. Pattern Recognit. 2016; 51:27–42.
2. Chan T, Vese L. Active contour without edges. IEEE Trans Image Process. 2001; 10(2):266–277.
[PubMed: 18249617]
3. Chen F, Yu H, Hu R. Shape sparse representation for joint object classification and segmentation.
IEEE Trans Image Process. 2013; 22(3):992–1004. [PubMed: 23144032]
4. Cremers D, Rousson M, Deriche R. A review of statistical approaches to level set segmentation:
integrating color, texture, motion and shape. Int J Comput Vis. 2007; 72(2):195–215.
5. Dai L, Ding J, Yang J. Inhomogeneity-embedded active contour for natural image segmentation.
Pattern Recognit. 2015; 48(8):2513–2529.
6. Estellers V, Zosso D, Lai R, Osher S, Thiran J, Bresson X. Efficient algorithm for level set method
preserving distance function. IEEE Trans Image Process. 2012; 21(12):4722–4734. [PubMed:
22692909]
7. He C, Wang Y, Chen Q. Active contours driven by weighted region-scalable fitting energy based on
Author Manuscript

local entropy. Signal Process. 2012; 92(2):587–600.

8. He L, Peng Z, Everding B, Wang X, Han C, Weiss K, Wee W. A comparative study of deformable
contour methods on medical image segmentation. Image Vis Comput. 2008; 26(2):141–163.
9. Jayadevappa D, Kumar S, Murty D. Medical image segmentation algorithms using deformable
models: a review. IETE Tech Rev. 2011; 28(3):248–255.
10. Ji Z, Xia Y, Sun Q, Cao G, Chen Q. Active contours driven by local likelihood image fitting energy
for image segmentation. Inf Sci. 2015; 301(20):285–304.
11. Jifeng N, Chengke W, Shigang L, Shuqin Y. NGVF: an improved external force field for active
contour model. Pattern Recognit Lett. 2007; 28(1):58–63.
12. Kass M, Witkin A, Terzopoulos D. Snakes: active contour models. Int J Comput Vis. 1987; 1(4):
321–331.
13. Lankton S, Tannenbaum A. Localizing region-based active contours. IEEE Trans Image Process.
2008; 17(11):2029–2039. [PubMed: 18854247]
14. Li, C., Kao, C., Gore, J., Ding, Z. Implicit active contours driven by local binary fitting energy.
Author Manuscript

IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2007. p. 339-345.
15. Li C, Kao C, Gore J, Ding Z. Minimization of region-scalable fitting energy for image
segmentation. IEEE Trans Image Process. 2008; 17(10):1940–1949. [PubMed: 18784040]
16. Li C, Wang X, Eberl S, Fulham M, Feng D. Robust model for segmenting images with/without
intensity inhomogeneities. IEEE Trans Image Process. 2013; 22(8):3296–3309. [PubMed:
23693130]
17. Li C, Wang X, Eberl S, Fulham M, Yin Y, Chen J, Feng D. A likelihood and local constraint level
set model for liver tumor segmentation from ct volumes. IEEE Trans Biomed Eng. 2013; 60(10):
2967–2977. [PubMed: 23771304]
18. Li C, Xu C, Gui C, Fox M. Distance regularized level set evolution and its application to image
segmentation. IEEE Trans Image Process. 2010; 19(12):3243–3254. [PubMed: 20801742]
19. Li D, Li W, Liao Q. Active contours driven by local and global probability distributions. J Vis
Commun Image Represent. 2013; 24(5):522–533.
20. Liu S, Peng Y. A local region-based Chan-Vese model for image segmentation. Pattern Recognit.
Author Manuscript

2012; 45(7):2769–2779.
21. Mabood L, Ali H, Badshah N, Ullah T. Absolute median deviation based a robust image
segmentation model. J Inf Commun Technol. 2015; 9(1):13–22.
22. Min H, Jia W, Wang X, Zhao Y, Hu R, Luo Y, Xue F. An intensity-texture model based level set
method for image segmentation. Pattern Recognit. 2015; 48(4):1547–1562.
23. Mylona E, Savelonas M, Maroulis D. Automated adjustment of region-based active contour
parameters using local image geometry. IEEE Trans Cybern. 2014; 44(12):2757–2770. [PubMed:
24771604]

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 12

24. Pablo A, Michael M, Charless F, Jitendra M. Contour detection and hierarchical image
segmentation. IEEE Trans Pattern Anal Mach Intell. 2011; 33(5):898–916. [PubMed: 20733228]
Author Manuscript

25. Posirca I, Chen Y, Barcelos C. A new stochastic variational PDE model for soft Mumford-Shah
segmentation. J Math Anal Appl. 2011; 384(384):104–114.
26. Ranjith, U., Caroline, P., Martial, H. A measure for objective evaluation of image segmentation
algorithms. IEEE Computer Society Conference on Computer Vision and Pattern Recognition
(CVPR); 2005. p. 34-34.
27. Song Y, Wu Y, Dai Y. A new active contour remote sensing river image segmentation algorithm
inspired from the cross entropy. Digital Signal Process. 2016; 48:322–332.
28. Swain M, Ballard D. Color indexing. Int J Comput Vis. 1991; 7(1):11–32.
29. Tomoshige S, Oost E, Shimizu A, Watanabe H, Nawano S. A conditional statistical shape model
with integrated error estimation of the conditions: application to liver segmentation in non-contrast
ct images. Med Image Anal. 2014; 18(1):130–143. [PubMed: 24184436]
30. Wang B, Gao X, Tao D, Li X. A nonlinear adaptive level set for image segmentation. IEEE Trans
Cybern. 2014; 44(3):418–428. [PubMed: 23797311]
31. Wang H, Huang T, Xu Z, Wang Y. An active contour model and its algorithms with local and
Author Manuscript

global gaussian distribution fitting energies. Inf Sci. 2014b; 263(1):43–59.

32. Wang H, Liu M. Active contours driven by local gaussian distribution fitting energy based on local
entropy. Int J Pattern Recognit Artif Intell. 2013; 27(6):1073–1089.
33. Wang L, He L, Mishra A, Li C. Active contours driven by local gaussian distribution fitting energy.
Signal Process. 2009; 89(12):2435–2447.
34. Wang L, Pan C. Robust level set image segmentation via a local entropy-based k-means clustering.
Pattern Recogn. 2014; 47(5):1917–1925.
35. Wang L, Shi F, Li G, Gao Y, Lin W, Gilmore J, Shen D. Segmentation of neonatal brain mr images
using patch-driven level sets. NeuroImage. 2014a; 84(1):141–158. [PubMed: 23968736]
36. Wang X, Huang D, Xu H. An efficient local Chan–Vese model for image segmentation. Pattern
Recognit. 2010; 43(3):603–618.
37. Wu H, Appia V, Yezzi A. Numerical conditioning problems and solutions for nonparametric i.i.d.
statistical active contours. IEEE Trans Pattern Anal Mach Intell. 2013a; 35(6):1298–1311.
[PubMed: 23599049]
Author Manuscript

38. Wu Y, Wang Y, Jia Y. Adaptive diffusion flow active contours for image segmentation. Comput
Vision Image Understanding. 2013b; 117(10):1421–1435.
39. Xu C, Prince J. Snakes, shapes, and gradient vector flow. IEEE Trans Image Process. 1998; 7(3):
359–369. [PubMed: 18276256]
40. Yang X, Gao X, Li J, Han B. A shape-initialized and intensity-adaptive level set method for auroral
oval segmentation. Inf Sci. 2014; 277(2):794–807.
41. Yezzi, J., Tsai, A., Willsky, A. Medical image segmentation via coupled curve evolution equations
with global constraints. IEEE Workshop on Mathematical Methods in Biomedical Image Analysis;
2000. p. 359-369.
42. Yezzi J, Tsai A, Willsky A. A fully global approach to image segmentation via coupled curve
evolution equations. J Vis Commun Image Representation. 2002; 13(2):195–216.
43. Yue X, Miao D, Zhang N, Cao L, Wu Q. Multiscale roughness measure for color image
segmentation. Inf Sci. 2012; 216(20):93–112.
44. Zhang K, Song H, Zhang L. Active contours driven by local image fitting energy. Pattern Recognit.
Author Manuscript

2010; 43(4):1199–1206.
45. Zhang K, Zhang L, Song H, Zhou W. Active contours with selective local or global segmentation: a
new formulation and level set method. Image Vis Comput. 2010; 28(4):668–676.
46. Zhao Y, Rada L, Chen K, Harding S, Zheng Y. Automated vessel segmentation using infinite
perimeter active contour model with hybrid region information with application to retinal images.
IEEE Trans Image Process. 2015; 34(9):1797–1807.

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 13

Appendix A. Derivation of the gradient flow

Author Manuscript

The minimization of energy functional E(ϕ) is equal to search for the steady-state solution of
the following gradient flow formulation:

where is the Gateaux derivative [37,38,43] of E(ϕ) with respect to ϕ. In our proposed
model, the derivations of regularization terms of ϕ in Eq. (12) have been already analyzed in
previous paper [33], we only need to solve the derivative of ELHIF in the manner: ϕ̃ = ϕ + ηϱ
with a small variation η. Fixing fi(x), i = 1, 2 and letting ϱ → 0, we have
Author Manuscript

Thus the corresponding Euler–Lagrange equation can be given by:

The Gateaux derivative of ELHIF is derived as follows

Author Manuscript

Combining the derivative of all the three terms in our proposed model, the total variational
formulation of ϕ is as below
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 14
Author Manuscript
Author Manuscript

Fig. 1.
The segmentation results of the proposed model for both synthetic and real images displayed
in different columns by changing default parameter values (i.e., σ = 5, Δt = 0. 01, ε = 1. 0
and υ = 0.0085 × 255 × 255 for the first image; σ = 5 and ε = 5 for the second image, λ2 =
0.05, σ = 7, Δt = 0.01 and ε = 7 for the third image). Row 1: initial rectangle contours and
their final evolutions. Row 2 and Row 3: intermediate results with 50 and 100 iterations,
respectively.
Author Manuscript
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 15
Author Manuscript
Author Manuscript
Author Manuscript

Fig. 2.
The segmentation results of our proposed model for four medical images by replacing
several default parameter values from left to right (i.e., λ2 = 0.07, σ = 5, Δt = 0.01 and ε =
1.0 for the first image; σ = 19, Δt = 0.01 and ε = 7 for the second image; σ = 5, Δt = 0.35
and ε = 1 for the third image; and σ = 8, Δt = 0.01 and ε = 1 for the last image). Row 1:
initial rectangle contours and their final evolutions. Row 2 and Row 3: fitted images ILFI and
DLFI. Row 4 and Row 5: fitted images ISFI and DSFI.
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 16
Author Manuscript
Author Manuscript
Author Manuscript

Fig. 3.
Comparisons of the CV, LBF, LIF, LGDF, LRCV, CEM and our models for the segmentation
of four images displayed in different columns based on the same initial green contour
curves. Row 1 to Row 7: results of the CV, LBF, LIF, LGDF, LRCV, CEM and our models,
respectively. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 17
Author Manuscript

Fig. 4.
Author Manuscript

The segmentation results of our model by merely changing (λ1, λ2) to (0.85, 0.1), (1.0, 0.1)
and (1.0, 0.25) for images shown in different columns.
Author Manuscript
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 18
Author Manuscript
Author Manuscript

Fig. 5.
The segmentation results of our models for a real image based on different initial contour
curves. Row 1: initial rectangle contours and their final evolutions. Row 2: the estimated
level set functions.
Author Manuscript
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 19
Author Manuscript
Author Manuscript
Author Manuscript

Fig. 6.
Comparison of the LBF, LIF, LGDF, LRCV and our models based on the same green initial
curves for the image corrupted by Gaussian white noise with variances of 0.005, 0.01, and
0.015, respectively. Rows 1–5: the results of the LBF, LIF, LGDF, LRCV and our models,
respectively. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Wang et al. Page 20

Table 1

Comparison of our proposed model with the CV, LBF, LIF, LGDF, LRCV and CEM models based on DSC for
Author Manuscript

synthetic and real images shown in Fig. 1.

Models Image1 Image2 Image3 Mean

CV 0.533 0.813 0.935 0.760
LBF 0.969 0.949 0.972 0.963
LIF 0.972 0.940 0.966 0.959
LGDF 0.970 0.948 0.948 0.955
LRCV 0.883 0.934 0.891 0.903
CEM 0.672 0.611 0.797 0.693
LHIF 0.987 0.959 0.981 0.976
Author Manuscript
Author Manuscript
Author Manuscript

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 2

Comparison of our proposed model with the CV, LBF, LIF, LGDF, LRCV and CEM models based on DSC for clinically medical images shown in Fig. 2.

Models Image1 Image2 Image3 Image4 Mean

Wang et al.

CV 0.412 0.904 0.754 0.752 0.705

LBF 0.898 0.976 0.951 0.979 0.951
LIF 0.890 0.955 0.956 0.951 0.938
LGDF 0.909 0.948 0.964 0.977 0.950
LRCV 0.796 0.962 0.963 0.954 0.919
CEM 0.451 0.827 0.783 0.705 0.692
LHIF 0.919 0.981 0.969 0.981 0.963

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
Page 21
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 3

Comparison of our model and the CV, LBF, LIF, LGDF, LRCV and CEM models in terms of iterations (Iter) and CPU time for synthetic and real images
given in Fig. 1.
Wang et al.

Models Image1 Image2 Image3

96 × 127 pixels 75 × 79 pixels 93 × 118 pixels

Iter Time(s) Iter Time(s) Iter Time(s)

CV 100 0.438 50 0.235 100 0.420
LBF 200 1.689 600 2.392 200 2.454
LIF 1000 2.510 800 1.859 200 3.056
LGDF 400 7.329 400 7.315 200 2.509
LRCV 800 9.932 360 4.217 100 4.462
CEM 500 2.945 400 2.214 150 0.889
LHIF 300 1.649 150 0.825 300 1.942

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
Page 22
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 4

Comparison of our model and the LBF, LIF, LGDF, and LRCV models in terms of PRI for randomly chosen 15 images from BSDS500 data set.

Image ID LBF LIF LGDF LRCV LHIF

Wang et al.

15088 0.773 0.858 0.862 0.797 0.876

28096 0.920 0.951 0.969 0.941 0.971
35070 0.659 0.776 0.754 0.713 0.793
66075 0.707 0.790 0.801 0.749 0.859
80099 0.802 0.826 0.888 0.911 0.924
108004 0.773 0.768 0.850 0.713 0.817
113044 0.605 0.601 0.723 0.701 0.788
135069 0.987 0.986 0.988 0.986 0.987
147091 0.735 0.755 0.863 0.724 0.869
175043 0.835 0.805 0.797 0.803 0.832
196027 0.810 0.843 0.885 0.829 0.891
207056 0.735 0.701 0.755 0.747 0.805
227092 0.796 0.779 0.750 0.755 0.810
296059 0.738 0.854 0.827 0.742 0.865
317080 0.784 0.791 0.787 0.777 0.839
Mean 0.777 0.806 0.833 0.793 0.862

Inf Sci (Ny). Author manuscript; available in PMC 2018 January 04.
Page 23