as were identified
tour lines. Further
> finding using the
E., Valenti, S. and
multi-factor spatial
existence of illegal
vhical Information
| landfill Site 2007
tawa, Canada.
itellite images and
landfill. Ryerson
2010. Selection of
g GIS and multi-
g and Assessment,
dar remote sensing
infield University,
lated health effects
landfill. American
184—190.
od for the remote
s: formulation and
Sensing, 29(4), pp.
d Haack. B. 2010.
azardous water: a
| Yan, W.Y. 2010.
ques in monitoring
satellite data. A/-
op. 542-551.
011. Public health
ed near municipal
ical Practice and
S
blic Authority of
rmation they have
al thanks also are
epean, and Dillon
rmation that they
research work is
Discovery Grant.
DETECTING SLUMS FROM QUICK BIRD DATA IN PUNE USING AN OBJECT
ORIENTED APPROACH
Sulochana Shekhar
Central University of Karnataka, Gulbarga, India
WG, Theme or Special Session: VIII/8: Land
KEY WORDS: Slums, Quick bird data, Object oriented Analysis, eCognition, Pune
ABSTRACT:
We have been witnessing a gradual and steady transformation from a pre dominantly rural society to an urban society in India and
by 2030, it will have more people living in urban than rural areas. Slums formed an integral part of Indian urbanisation as most of
the Indian cities lack in basic needs of an acceptable life. Many efforts are being taken to improve their conditions. To carry out
slum renewal programs and monitor its implementation, slum settlements should be recorded to obtain an adequate spatial data
base. This can be only achieved through the analysis of remote sensing data with very high spatial resolution. Regarding the
occurrences of settlement areas in the remote sensing data pixel-based approach on a high resolution image is unable to represent
the heterogeneity of complex urban environments. Hence there is a need for sophisticated method and data for slum analysis. An
attempt has been made to detect and discriminate the slums of Pune city by describing typical characteristics of these settlements,
by using eCognition software from quick bird data on the basis of object oriented approach. Based on multi resolution
segmentation, initial objects were created and further depend on texture, geometry and contextual characteristics of the image
objects, they were classified into slums and non-slums. The developed rule base allowed the description of knowledge about
phenomena clearly and easily using fuzzy membership functions and the described knowledge stored in the classification rule base
led to the best classification with more than 80% accuracy.
1. INTRODUCTION
1.1 Urbanisation and Slums
Today, half the world's population lives in urban areas and by
the middle of this century all regions will be predominantly
urban, and according to current projections, virtually the whole
of the world's population growth over the next 30 years will be
concentrated in urban areas (UN-HABITAT, 2010). Above all,
this rapid urban growth has been strongly associated with
poverty and slum growth. It is felt that slums represent the
worst of urban poverty and inequality. The increasing
concentration of the urban population in slum areas is
generally equated with increasing urban poverty a process
recognized as the urbanization of poverty. According to new
estimates presented in UN-HABITAT’s report, between the
year 2000 and 2010 over 200 million people in the developing
world will have been lifted out of slum conditions. But in the
course of the same years the number of slum dwellers will be
increased by six million every year. Based on these trends it is
expected that the world’s slum population will continue to
grow if no corrective action is taken in the coming years (UN-
HABITAT, 2010).
Defining slum raises several conceptual issues, making it
difficult to precisely estimate the slum population living in
urban areas. Concepts and definitions of slums vary from
country to country and even in the same country, slum
settlements may be known by different names (Kohli.D, 2011)
In order to carry out the urban planning and development tasks
necessary to improve living conditions for the poorest world-
wide an adequate spatial data basis is needed (Mason, O.S and
Fraser, C.S., 1998) and this can only be obtained through the
analysis of remote sensing data (Hofmann,P.,2001). Since
traditional methods demand more labour, money and time,
alternative methods that include sophisticated techniques to
extract the information from remote sensing data of very high
resolution (VHR) and thus could reduce subjectivity, time and
labour (Naga Jyothi., 2008 et al.) and provide more reliable
data are need of the hour.
1.2 Related work on detecting slums
Many studies used census and field survey as the basis for
studying about slums and formed the database for GIS-based
mapping (Joshi Pratima, Sen Srinanda and Hobson Jane, 1998;
Sliuzas and Kuffer, 2006). Recently very high resolution
remote sensing based methods for mapping slums are getting
popular among the scientific community (Mason& Fraser,
1998; Sliuzas, Kerle and Kuffer, 2008; Hofmann .P., 2001;
Hofmann. P., 2004 et al. ) but there are only very few studies
based on Indian situation (Ujjwal Sur, 2004).
In most of the remote sensing based studies visual
interpretation of data has played major role in identification of
slums (Angeles et al., 2009; Sliuzas R.V 2004.,Sliuzas and
Kuffer 2008; Baud, Kuffer, Pfeffer, Sliuzas, and Karuppannan
(2010). Visual interpretation performed by interpreters familiar
with local conditions provides a flexible and useful approach to
slum mapping, though it does have shortcomings for repetitive
surveys of very large cities due to difficulties in controlling
quality over time and between interpreters. Later pixel based
image classification is widely used in slum analysis and it also
helped to understand the patterns over time and space (Jain,
Sokhi and Sur, 2005; Jain, 2007; Weeks et al., 2007). But
pixel-based approach on a high resolution image is unable to
represent the heterogeneity of complex urban environments. In
most cases by using only pixels’ spectral information to
