International Journal of Science and Research (IJSR)

ISSN (Online): 2319-7064

Impact Factor (2012): 3.358

Role of Active Contour in Image Processing

Manjeet Kaur1, Sukhpreet Kaur2
1, 2

Department of Computer Science, Shaheed Udham Singh College of Engineering & Technology, Tangori, Punjab, India

Abstract: With the growing research on image segmentation, it has become important to provide readers with an overview of the
existing segmentation techniques. The active contour is one of the most successful models in image segmentation. Active contours are
used extensively in computer vision and digital image processing to locate object boundaries. It consists of evolving a contour in images
toward the boundaries of objects. Its success is based on strong mathematical properties and efficient numerical schemes based on the
level set method. In this paper, we review and classify active contour models in literature.
Keywords: Segmentation, Image Processing, Active Contour

1. Introduction

2. Related Work

Image processing is a rapidly emergent area of computer

science. Its growth has been uplifted by technological
achievements in digital imaging, computer processors and
mass storage devices. The flexibility and affordability of
various streams are now made path to digital systems.
Examples are medicine, remote sensing and security
monitoring etc.., and the other sources also produce huge
amount of digital image data day to day, which cannot ever
be examined manually. Digital image processing is
concerned primarily to extract useful information from
images. Ideally, this is done by computers, with little or no
human intervention. Image processing algorithms may be
placed at three levels. At the lowest level are those
techniques which deal directly with the raw, possibly noisy
pixel values, with denoising and edge detection being good
examples. In the middle are algorithms which utilize low
level results for further means, such as segmentation and
edge linking. At the highest level are those methods which
attempt to extract semantic meaning from the information
provided by the lower levels [5].

There is much research work in the field of object detection

over the past decades. Some of the work done has been
discussed. Ch. Sri Veera Sagar et al. proposed a new
methodology for the edge detection of complex radar
images. It includes the edge improvisation algorithm with
edge detection. The use of discrete wavelet transform in the
edge improvisation algorithm is justified. Then region based
active contour model is used as edge detection algorithm.
They used distribution fitting energy with a level set function
and neighborhood means and variances as variables. Results
show the accuracy of the proposed scheme using different
images [5]. B. Wang et al. proposed a algorithm based on
background modeling and active contour model for moving
object edge detection. It uses the background modeling to
complete moving object detection, then it uses quad-tree
decomposition method to contain the corresponding to the
foreground image, through the data distribution density of
the sparse matrix, it calculates the seed points corresponding
to the regions which are containing the moving object.
Results show that the proposed algorithm can effectively
obtain the object outlines of multi-moving objects and the
edge detection results are close to the judgment of the human
visual, parallel contour extraction makes the algorithm has
good real-time [1]. B.S Saini et al.compared two methods of
image segmentation based on level sets i.e edge based and
region based active contours. Results show that if iteration is
controlled then edge based can able to locate ROI more
accurately as compared to other method. Region based either
go far beyond cancer periphery or confined within periphery
gives false information regarding the spread of cancer in
body. Quantitative comparison is also done which shows that
edge based is giving far more accurate idea of location and
spread of cancer [2]. Bryan Catanzaro et al. examined
efficient parallel algorithms for performing image contour
detection, with particular attention paid to local image
analysis as well as the generalized eigensolver used in
Normalized Cuts. The efficiency gains they realized enable
high-quality image contour detection on much larger images
than previously practical, and the algorithms they proposed
are applicable to several image segmentation approaches.
Efficient, scalable, yet highly accurate image contour
detection will facilitate increased performance in many
computer vision applications [3]. Yunyun Yang et al.
presented an modified active contour model for fast
multiphase image segmentation based on the piecewise

Image Segmentation is technique used to find objects of

interest from the background. The object pixels would be
black (min intensity) and background pixels white (max
intensity). Segmentation refers to the process of partitioning
a digital image into multiple segments . The goal of
segmentation is to simplify and/or change the representation
of an image into something that is more meaningful and
easier to analyze.
Active contour or Snake model is an extensively-used
method in image segmentation and target detection, a
bordering method based on partial gradient. Firstly, an initial
contour is placed near the interesting object via users or other
automatic processes, and the contour will be changed by the
internal and external energy. The external energy guides the
active contour to move towards the border of the object
while the internal energy helps maintain the smoothing and
topology of the active contour. When the energy reaches its
minimum degree, the active contour will shrink to the border
of the object to be detected. But the active contour model is
always sensitive to the noise. More importantly, the result
produced thereupon usually relies on the initiation and lacks
enough topology adaptation [5].

Paper ID: 020132243

Volume 3 Issue 5, May 2014

www.ijsr.net
Licensed Under Creative Commons Attribution CC BY

1789

International Journal of Science and Research (IJSR)

ISSN (Online): 2319-7064
Impact Factor (2012): 3.358

constant Vese-Chan model and the split Bregman method.

By applying the globally convex image segmentation
technique to the piecewise constant Vese-Chan energy
functional, they defined a new biconvex energy functional to
guarantee fast convergence. Modified model has been tested
with synthetic and real images. Experimental results show
that the modified model can obtain similar results to the
Vese-Chan model but is much more efficient[18]. Ray, N et
al. proposed to minimize a novel region-based energy
functional based on Bhattacharya coefficient involving
histograms of image features. The optimization of the
proposed energy functional simply consists of two very
efficient searches if a crude segmentation such as a bounding
box around the region of change is sufficient. They
illustrated encouraging results on finding bounding box
around abnormality from brain MRI, object detection for
maritime surveillance, and segmenting oil-sand particles
from conveyor belt images [14].

3. Proposed Methodology

a) Initialization n=0, nmax

b) repeat
n++
Compute( c1 , c2)
Evolving the level-set function (outside = -1, inside = 1)
out = find( < 0);
in = find( > 0);
c1 = sum(Img(in)) / size(in);
c2 = sum(Img(out)) / size(out);
for all (x, y)
fx(x, y) = ( (x+1, y)- (x-1, y))/(2*delta_s); where
delta_s is between (0.1, 1.0)
Repeat steps until the solution is stationary, or n>nmax , or
The energy is not changing or contour is not moving

Paper ID: 020132243

In this study, the overview of Various Active Contour

methods applied for image Processing is explained briefly.
This study provides a simple guide to those researchers who
carried out their research study in image segmentation. As
the Active Contour is affected by lots of factors, such as:
energy minimization, spatial characteristics of the image
continuity, texture, image content. Due to these factors
image segmentation is a challenging problem in digital
image processing and is still a pending problem in the world.

5. Acknowledgement
I am thankful to AP Sukhpreet Kaur, Asst.Prof, Shaheed
Udham Singh College of Engineering Technology, Tangori,
for providing constant guidance and encouragement for this
research work.

References

This work is related to find out the optimal solution for

minimization problem of the active contour and algorithm is
given below:

3.1 Flow Graph

4. Conclusion

[1] Bo Wang, Kun Zhang, Liang Shi, Huihui Zhong, An

Edge Detection Algorithm of Moving Object Based on
Background Modeling and Active Contour Model,
International Conference on Systems Engineering and
Modeling, pp. 340-346, 2013.
[2] B.S Saini, G.Sethi, Comparative Analysis of Edge
Based and Region Based Active Contour using Level
Sets and Its Application on CT Images, International
Journal of Research in Engineering and Technology,
vol. 2, pp. 566-573, April 2013.
[3] Bryan Catanzaro, Bor-Yiing Su, Narayanan Sundaram,
Yunsup Lee, Mark Murphy, Kurt Keutzer, Efficient,
High-Quality Image Contour Detection, IEEE
International conference on Computer Vision, pp. 23812388, 2009.
[4] C. Lawrence Zitnick, Devi Parikh, The Role of Image
Understanding in Contour Detection, IEEE, pp. 622629, 2012.
[5] Ch.Sri Veera Sagar, K.Rama Devi, A New Approach
Of Edge Detection In SAR Images Using Region-Based
Active Contours, International Journal of Research in
Engineering and Technology, vol.2, September 2013.
[6] Chunming Li, Fang Li, Chiu-Yen Kao, Chenyang Xu,
Image Segmentation with Simultaneous Illumination
and Reflectance Estimation: An Energy Minimization
Approach, Siemens Corporate Research, Princeton,
ICCV-2009.
[7] Jiang Xin and Zhuangzhi Yan, The Application of
Watershed Algorithm in the Initial Contour Selection of
Mumford-Shah Model, IEEE, pp. 1773-1776,
September 2005.
[8] Junge Sun, Yunhong Wang, Zhaoxiang Zhang, Yiding
Wang School of Computer Science and Engineering,
Beihang University, China School of Information
Engineering, North China University of Technology,
China, Salient Region Detection in High Resolution
Remote Sensing Images, IEEE, 2010.
[9] Kalpana Saini, M.L. Dewal, and Manojkumar Rohit,
Level set based on new Signed Pressure Force Function
for
Echocardiographic
image
segmentation,

Volume 3 Issue 5, May 2014

www.ijsr.net
Licensed Under Creative Commons Attribution CC BY

1790

International Journal of Science and Research (IJSR)

ISSN (Online): 2319-7064
Impact Factor (2012): 3.358

International Journal of Innovation and Applied Studies,

vol.3 No. 2, pp. 560-569, June 2013.
[10] Lei He, Thoma, G.R, "Multiphase Level Set Model with
Local K-means Energy for Histology Image
Segmentation", HISB-2011.
[11] Mahmoud Awadallah, Sherin Ghannam, Lynn Abbott,
and Ahmed Ghanem, Active Contour Models for
Extracting Ground and Forest Canopy Curves from
Discrete Laser Altimeter Data, SilviLaser, pp. 1-8,
October 2013.
[12] Rafael C. Gonzalez, Richard Eugene Woods, Digital
image processing.
[13] Riadh Ksantini, Boubakeur Boufama, Sara Memar, "A
new efficient active contour model without local
initializations for salient object detection", EURASIP
Journal on Image and Video Processing, 2013.
[14] Ray, N., Saha, B.N., Hong Zhan, "Change Detection and
Object Segmentation: A Histogram of Features-Based
Energy Minimization Approach", ICVGIP -2008. 11
[15] Sara Memar, Karen Jin, Boubakeur Boufama, Object
Detection Using Active Contour Model with Depth
Clue, Image Analysis and Recognition, Lecture Notes
in Computer Science, Vol. 7950, 2013.
[16] Siew Wen Chin, Kah Phooi Seng, and Li-Minn Ang,
Lips Contour Detection and Tracking Using Watershed
Region-Based Active Contour Model and Modified
H, IEEE Transactions on Circuits and Systems for
Video Technology, vol. 22 , 2012.
[17] http://en.wikipedia.org/wiki/Image_segmentation
[18] Yunyun Yang, Shenzhen Grad, "Split Bregman method
for minimization of modified Vese-Chan model for fast
image segmentation", ISPA-2013.
[19] Yuqing Miao, Department of Computer Science And
Technology, Guilin University Of Electronic
Technology, Guilin, Dong Yin, School of Information
Science And Technology, University of Science And
Technology of China, Hefei, China, Target Recognition
Application Research With Gradient Magnitude And
Direction, Proceedings of The 2008 IEEE International
Conference on Information And Automation, 2008

Paper ID: 020132243

Volume 3 Issue 5, May 2014

www.ijsr.net
Licensed Under Creative Commons Attribution CC BY

1791