Machine Vision and Applications
DOI 10.1007/s00138-012-0475-8
ORIGINAL PAPER
Change detection for moving object segmentation with robust
background construction under Wronskian framework
Badri Narayan Subudhi · Susmita Ghosh ·
Ashish Ghosh
Received: 24 September 2011 / Revised: 18 September 2012 / Accepted: 3 December 2012
© Springer-Verlag Berlin Heidelberg 2013
Abstract Although background subtraction techniques
have been used for several years in vision systems for moving
object detection, many of them fail to provide good results
in presence of noise, illumination variation, non-static back-
ground, etc. A basic requirement of background subtraction
scheme is the construction of a stable background model
and then comparing each incoming image frame with it so
as to detect moving objects. The novelty of the proposed
scheme is to construct a stable background model from a
given video sequence dynamically. The constructed back-
ground model is compared with different image frames of
the same sequence to detect moving objects. In the proposed
scheme the background model is constructed by analyzing a
sequence of linearly dependent past image frames in Wron-
skian framework. The Wronskian based change detection
model is further used to detect the changes between the con-
structed background scene and the considered target frame.
The proposed scheme is an integration of Gaussian averaging
and Wronskian change detection model. Gaussian averaging
uses different modes which arise over time to capture the
underlying richness of background, and it is an approach
for background building by considering temporal modes.
Similarly, Wronskian change detection model uses a spa-
tial region of support in this regard. The proposed scheme
B. N. Subudhi · A. Ghosh (B )
Machine Intelligence Unit, Indian Statistical Institute,
Kolkata 700108, India
e-mail: ash@isical.ac.in
B. N. Subudhi
e-mail: subudhi.badri@gmail.com
S. Ghosh
Department of Computer Science and Engineering,
Jadavpur University, Kolkata 700032, India
e-mail: susmitaghoshju@gmail.com
relies on spatio-temporal modes arising over time to build
the appropriate background model by considering both spa-
tial and temporal modes. The results obtained by the pro-
posed model is found to provide accurate shape of moving
objects. The effectiveness of the proposed scheme is verified
by comparing the results with those of some of the existing
state of the art background subtraction techniques on public
benchmark databases. We found that the average F-measure
is significantly improved by the proposed scheme from that
of the state-of-the-art techniques.
Keywords Background subtraction · Object detection ·
Temporal analysis · Gaussian mixture models
1 Introduction
An important problem in computer vision is the detection
of moving objects from a video sequence [7]. Proper detec-
tion of moving objects is crucial for many computer vision
and artificial intelligent systems. Moving object detection
has been widely applied in the fields like visual surveillance
[23, 27], face and gait-based human recognition [29, 37],
activity recognition [17], robotics [28], etc.
Background subtraction (BGS) method [22] is very pop-
ular for video object detection. It needs to handle different
situations like illumination variation, noise, non-static back-
ground, and shadow of the moving objects in the scene. The
effects of noise and illumination variation are very com-
mon in daily life video scenes [4]. In this regard a robust
BGS scheme using running Gaussian average was proposed
by Wren et al. [36]. Here, the authors have modeled the
background independently at each pixel location by fitting
a Gaussian probability distribution function (pdf) over a
sequence of previous/past image frames. The parameters of
the Gaussian pdf (i.e., the mean and the variance) at each
123