e of large-scale
'equirements of
ie is not true for
nomic standards
rs because of
to reduce IR
s like free-hand
level of fatigue.
18 to the use of
that will have to
ties of this new
ral Agency for
in Detmold for
sten Meier and
students project
Last accessed
gesturetek.com/
. Last accessed
arge interactive
| Interantional
.177 (2007)
the European
March 2007):
1.2012.
its evolution ,
ne 51, Issue 6.
| framework for
e twenty-sixth
| in computing
Fischer, H.:
, Proc. Mensch
), M., Kase, I:
education. In:
k (2007)
cher, H. 2011.
isaster Control
id Pen-Based-
me-touch.com/;
Y ft.com/surface/
Last accessed
. (Eds): Geo-
2005.
SOFT COMPUTING APPROACH FOR LIQUEFACTION IDENTIFICATION USING
LANDSAT-7 TEMPORAL INDICES DATA
Sengar S.S.^, Kumar A>" Ghosh S. K.?, and Wason H. R.?
‘Indian Institute of Technology, Roorkee, India-(sandydeq, scangfce, wasonfeq) @iitr.ernet.in
"Indian Institute of Remote Sensing, Dehradun, India-anil@iirs.gov.in
KEY WORDS: Earthquakes, Landsat, Accuracy, Temporal Indices Data
Working Group VIII/1
ABSTRACT:
A strong earthquake with magnitude M,,7.7 that shook the Indian Province of Gujarat on the morning of January 26, 2001, caused
widespread appearance of water bodies and channels, in the Rann of Kachchh and the coastal areas of Kandla port. In this work, the
impact of using conventional band ratio indices from Landsat-7 temporal images for liquefaction extraction was empirically
investigated and compared with Class Based Sensor Independent (CBSI) spectral band ratio while applying noise classifier as soft
computing approach via supervised classification. Five spectral indices namely, SR (Simple Ratio), NDVI (Normalized Difference
Vegetation index), TNDVI (Transformed Normalized Difference Vegetation Index), SAVI (Soil-Adjusted Vegetation Index) and
Modified Normalized Difference Water Index (MNDWI) were investigated to identify liquefaction using temporal multi-spectral
images. It is found that CBSI-TNDVI with temporal data has higher membership range (0.968-0.996) and minimum entropy (0.011)
to outperform for extraction of liquefaction and for water bodies extraction membership range (0.960-0.996) and entropy (0.005)
respectively.
1. INTRODUCTION the Bhuj, by calculating the absorption of energy in the NIR
and SWIR region of electromagnetic spectrum. Saraf et al.,
With the advent of operational remote sensing it is now (2002), used picture colour transformation of IRS-1D band 4
possible to efficiently and accurately map earthquake induced data to map liquefaction of Bhuj. Mohanty ef al., (2001) and
ground changes including the appearance of water bodies and Singh et al., (2001), used image differencing to find
changes in soil moisture conditions. During the earthquake, liquefaction.
strong shaking produced liquefaction in the fine silts and sands
below the water table in the Rann of Kachchh. Digital image A lot of work has been done in the field of single class
classification is a fundamental image processing operation to extraction through time series multi-spectral data but while
extract land cover information from remote sensing data and it going through the literature it has been identified that to find
assigns a class membership for each pixel in an image. Often, liquefaction using various indices with noise classifier has not
particularly in coarse spatial resolution images, the pixels may been explored in the past. In this study it has been tried to
be mixed containing two or more classes. Soft classification identify liquefaction using temporal indices.
methods may help in quantifying uncertainties in areas of
transition between various types of land cover. Fuzzy
classifications may be beneficial where a mixed pixel may be 2. INDICES AND CLASSIFICATION APPROACHES
assigned multiple class memberships.
; s : 3 For generating band ratio data, it is important to know various
Till date many researchers in remote sensing field have applied types of band information present in multi-spectral data. Based
time series indices to study cropping pattern. Nianlong ef al, m spectral information of remote sensing data, the user has to
(2010); applied time series NDVI data to identify land Use decide which spectral bands of data are to be used in different
classification. Wardlow and Egbert, (2008), used a hierarchical band ratio functions and require expert knowledge.
crop mapping to classify multi-temporal NDVI data. Lucas ef
al., (2007), studied the use of time-series Landsat sensor data 2.1 CBSI Spectral Band Ratio
using decision rules based on fuzzy logic to discriminate
vegetation type. Based on past research works it indicates that To overcome the need for expert knowledge about remote
researchers have used multi-spectral, hyper-spectral as well a$ ^ sensing data, in this work a CBSI spectral band ratio has been
microwave data for specific land use/land cover (LULC) proposed in eq. (1).
identification while using temporal data sets with importance of
different indices.
CBSI spectral band ratio = min, max/{1...nr0}k (1)
Kumar and Roy, (2010), has worked with add on bands in
multi-spectral dataset of Worldview -2. This work has where; g is grey value, r and c are row and column of a class
proposed class based sensor independent spectral band ratio location respectively, n denotes number of bands, k denotes
NDVI approach for extracting a crop at a time. Ramakrishnan number of classes.
et al., (2006), map the earthquake induced liquefaction around
* Corresponding author
61
