Академический Документы
Профессиональный Документы
Культура Документы
Faculty of Engineering, Department of Civil Engineering, University Putra Malaysia, Serdang, Malaysia
Institute for Cartography, Faculty of Forestry, Geo and Hydro-Science, Dresden University of Technology, 01062 Dresden, Germany
art ic l e i nf o
a b s t r a c t
Article history:
Received 27 May 2013
Received in revised form
13 October 2013
Accepted 17 October 2013
Available online 6 November 2013
Extraction of the ooded areas from synthetic aperture radar (SAR) and especially TerraSAR-X data is one
of the most challenging tasks in the ood management and planning. SAR data due to its high spatial
resolution and its capability of all weather conditions makes a proper choice for tropical countries.
Texture is considered as an effective factor in distinguishing the classes especially in SAR imagery which
records the backscatters that carry information of kind, direction, heterogeneity and relationship of the
features. This paper put forward a computer program for texture analysis for high resolution radar data.
Texture analysis program is introduced and discussed using the gray-level co-occurrence matrix (GLCM).
To demonstrate the ability and correctness of this program, a test subset of TerraSAR-X imagery from
Terengganu area, Malaysia was analyzed and pixel-based and object-based classication were attempted.
The thematic maps derived by pixel-based method could not achieve acceptable visual interpretation
and for that reason no accuracy assessment was performed on them. The overall accuracy achieved by
object-based method was 83.63% with kappa coefcient of 0.8. Results on image texture classication
showed that the proposed program is capable for texture analysis in TerraSAR-X image and the obtained
textural analysis resulted in high classication accuracy. The proposed texture analysis program can be
used in many applications such as land use/cover (LULC) mapping, hazard studies and many other
applications.
& 2013 Elsevier Ltd. All rights reserved.
Keywords:
Texture analysis
Feature extraction
Remote sensing
TerraSAR-X
Pixel based
Malaysia
1. Introduction
Texture is considered as an important characteristic in image
processing which is useful in radar remote sensing and other elds
where it is necessary to interpret gray value images like in the
medical sector (Treitz et al., 1996; Mahmoud et al., 2011). Due to the
surface properties such as roughness, humidity and orientation
every object in SAR scene can have its own unique backscattering
properties (Haack and Bechdol, 2000). The recognition of texture
and object classication are the most challenging problems in the
remotely sensed data processing (Zhang et al., 2007; Hamedianfar
and Shafri, 2013). The main aim in image processing is to convert
the remote sensing (RS) imagery information into tangible information which can be understandable and possibly be used in combination with other data in Geographic Information System (GIS)
environment (Blaschke, 2010). Therefore precision of the data and
the texture analysis method are two main factors that have direct
impact on the level of accuracy and information that can be
achieved (Chica-Olmo and Abarca-Hernandez, 2000).
0098-3004/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.cageo.2013.10.011
35
3. Methodology
The practical and theoretical aspects of texture analysis implemented in this study involve several steps as shown in Fig. 1.
3.1. General information on design and functionality
The texture analysis program was implemented using ArcGIS
ArcObjects 9.2 and Visual Basic for Application (VBA). Fig. 2 shows
the interface of proposed program.
The program can be divided into three parts: raster denition,
matrix framework creation and formula application. The graphic
interface of the program consists of one window containing three
frames. The rst frame of the starting window allows the user to
select a raster le for the texture analysis either by selecting an
already opened le in ArcMap 9.2 or by searching through the
directory. The image le will always be saved as an .img in the
folder of the original image. Depending on chosen direction and
type of analysis, an appendix will be added to the le name, thus
giving it an easy to recognize the description.
The second and third frame makes various texture options for four
different directions available to the user. The texture options which
can be found in the second frame are divided into the three groups
such as contrast group, orderliness group and statistics group. To
achieve these three groups, GLCM should be calculated. The GLCM,
also called gray tone spatial dependency matrix, was rst introduced
by Haralick et al. (1973). To dene this matrix, one has to picture a
rectangular image with Nr rows, Nc columns and with a quantization
of Ng gray levels from the tone of each single pixel. Lr {0,1,2,.,Nr}
and Lc {0,1,2,.,Nc} will be the number of rows or spatial domains
and dened as the set of Ng quantied gray levels. If we create the
36
0; 0
6 1; 0
6
6
Lr Lc : 6
6 2; 0
6
4
N r ; 0
Lr f0; 1; 2; :; N r g
0; 1
0; 2
1; 1
1; 2
2; 1
2; 2
Nr ; 1
0; N c
N r ; 2
1; N c 7
7
7
2; N c 7
7
7
5
N r ; N c
Lc f0; 1; 2; :; N c g
Each pixel is dened as a pair of coordinates and has a gray level G
assigned to itself by a function that can be described as the image
Lr Lc ; I : Lr Lc -G. Now, the relative frequency of occurrence of a
pair of neighboring pixels or resolution cells that have a distance d
between them can be expressed as a matrix P ij . Hereby the variables i
and j describe respective gray levels. This matrix P ij or is called the
GLCM. It is symmetric and denes distance and direction, the so
called angular relationship, of two pixel neighbors as a function
(Haralick, 1979).
37
Fig. 3. Development of GLCM (example): kernel (1), matrix framework Fij (2), symmetrical matrix Vij (3), normalized matrix Pij (4).
V ij
1
N
i;j 0 V ij
Now that the GLCM has been calculated, the formulas can be
applied. As mentioned they can be divided into three groups:
contrast group, orderliness group and statistics group. The previously used symmetric normalized GLCM example P will be used
in the following paragraphs as a demonstration for calculation. The
second matrix will be containing the respective weights resulting
from the given formula. The scalar product of these two matrices
will be the sought for result.
P ij ji jj
N1
P ij i j2
i;j 0
P ij
2
i;j 0 1 i j
N1
P i;j 2 Energy
p
ASM
i;j 0
i;j 0
P i;j ln P i;j
i;j 0
38
i P i;j
i;j 0
N 1
P i;j i 2
i;j 0
Standard deviation : s
p
s2
10
3.2. Analysis
3.2.1. Data and study area
The study is carried out in Terengganu which is situated in
north-eastern part of Peninsular Malaysia, and is bordered in the
northwest by Kelantan, the southwest by Pahang, and the east by
the South China Sea (Zaleha et al., 2006). The data used in this
work was recorded by TerraSAR-X satellite, short TSX-1, from 27th
November 2009. Data was single look, stripmap modus, with three
meters spatial resolution and HH polarization. TerraSAR-X data
used with radiometric resolution of 16 bit thus providing 65,536
different gray values.
3.2.2. Pre-processing
In order to guarantee good results, the images have to be preprocessed (Albinet et al., 2012). Therefore the following steps are
Fig. 5. Original image (upper left) and contrast group results: contrast (upper right), dissimilarity (lower left), homogeneity (lower right).
Fig. 6. Orderliness group results: ASM (upper left), energy (upper right), entropy (lower left), MAX (lower right).
39
40
Fig. 7. Statistics group result: mean (upper left), variance (upper right), Standard deviation (lower left), correlation (lower right).
achieved from texture analysis into one image. This was done
using Composite Bands of ArcMaps Data Management Tools. This
created an image le with 44 bands. After a recalculation of its
statistics and the assignment of the no data value 9999, the
image could be used for segmentation. Scale and color values were
selected, depending on the properties of the image used. Hereby
the resolution of the image object level is dened by the scale
parameter. Thus low values indicate that smaller objects will be
created. The second parameter, color, signies the importance of
image color homogeneity in the segmentation process. For this
work, scale received a parameter value of ten and 0.9 was assigned
as the color parameter. After the execution of the segmentation
process an image containing segmentation polygons was calculated and returned. Nearest neighbor was chosen as the classication method. Color and mutual embedding were both taken into
consideration. Training areas could be dened by assigning certain
segments to either a positive or negative predened output class.
This pre-denition had to be done by the user and consisted of a
name and a color for the later class. For example, the rst
classication cycle contained the positive output class water (blue)
and the negative output class land (green). This way an evaluation
of the texture analysis program and its results can be conducted.
41
Fig. 9. Six layer false color representation of the texture band composite in eCognition with possible elements for later classication (rotated 901 clockwise). (For
interpretation of the references to color in this gure legend, the reader is referred to the web version of this article.).
42
Acknowledgments
The German Aerospace Center (DLR) provided Terra-SAR-X data
under the Science proposal ID: HYD0326. Thanks to Thomas Hahmann for his valuable inputs on the processing of Terra-SAR-X data.
References
Al Fugura, A., Billa, L., Pradhan, B., Mohamed, T.A., Rawashdeh, S., 2011. Coupling of
hydrodynamic model and aerial photogrammetry-derived digital surface model
for ood simulation scenarios using GIS: Kuala Lumpur ood. Malaysia. Disaster
Adv. 4 (4), 2028.
Table 1
Object-based classication accuracy assessment.
Settlement
Water
Settlement and wood
Wood
Woodless
Sum
Producer accuracy %
Settlement
Water
Wood
Woodless
Sum
User accuracy %
46
0
0
1
0
50
98
1
51
2
4
2
37
85
2
0
29
4
0
51
83
0
0
7
33
1
50
80
Overall accuracy
Kappa coefcient
0
0
12
1
30
50
70
49
0
50
43
33
238
94
100
58
77
91
83.6%
0.8
43
Lee, S.J., Serre, M.L., van Donkelaar, A., Martin, R.V., Burnett, R.T., Jerrett, M., 2012b.
Comparison of geostatistical interpolation and remote sensing techniques for
estimating long-term exposure to ambient PM2. 5 Concentrations across the
Continental United States. Environ. Health Perspect. 120 (12), 17271732.
Luo, Y., Zhao, S., Zhou, H., Wang, A., He, K., Tan, L., 2012. A novel classication
method based on texture analysis using high-resolution SAR and optical data.
In: Proceedings of the 2012 Second International Workshop on the Earth
Observation and Remote Sensing Applications (EORSA), IEEE, pp. 186190.
Magee, K.S., 2011. Segmentation, object-oriented applications for remote sensing
land cover and land use classication (Ph.D. dissertation). University of
Cincinnati, USA. (131pp).
Mahmoud, A., Elbialy, S., Pradhan, B., Buchroithner, M., 2011. Field-based landcover
classication using TerraSAR-X texture analysis. Adv. Space Res. 48 (5),
799805.
Materka, A. Strzelecki, M., 1998. Texture Analyisis Methods A Review.
Morris, O.J., Lee, M.D.J., Constantinides, A.G., 1986. Graph theory for image analysis:
an approach based on the shortest spanning tree. Communications, Radar and
Signal Processing. IEE Proceedings, F 133 (2), 146152.
Ouma, Y.O., Josaphat, S.S., Tateishi, R., 2008. Multiscale remote sensing data
segmentation and post-segmentation change detection based on logical modeling: theoretical exposition and experimental results for forestland cover
change analysis. Comput. Geosci. 34 (7), 715737.
Pant, T., Singh, D., Srivastava, T., 2010. Advanced fractal approach for unsupervised
classication of SAR images. Adv. Space Res. 45 (11), 13381349.
Petropoulos, G.P., Kalaitzidis, C., Prasad Vadrevu, K., 2012. Support vector machines
and object-based classication for obtaining land-use/cover cartography from
Hyperion hyperspectral imagery. Comput. Geosci. 41, 99107.
Pradhan, B., 2009. Flood susceptible mapping and risk area estimation using logistic
regression, GIS and remote sensing. J. Spat. Hydrol. 9 (2), 118.
Pradhan, B., Shae, M., 2009. Flood hazard assessment for cloud prone rainy areas
in a typical tropical environment. Disaster Adv. 2 (2), 715.
Pradhan, B., Shae, M., Pirasteh, S., 2009. Maximum ood prone area mapping
using RADARSAT images and GIS: Kelantan river basin. Int. J. Geoinf. 5 (2),
1123.
Pradhan, B., 2010. Remote sensing and GIS-based landslide hazard analysis and
cross-validation using multivariate logistic regression model on three test areas
in Malaysia. Adv. Space Res. 45 (10), 12441256.
Pradhan, B., Youssef, A.M., 2011. A100-year maximum oood susceptibility mapping using integrated hydrological and hydrodynamic models: Kelantan River
Corridor. Malaysia. J. Flood Risk Manag. 4 (3), 189202, http://dx.doi.org/
10.1111/j.1753-318X.2011.01103.x.
Ren, Y., Lehner, S., Brusch, S., Li, X., He, M., 2012. An algorithm for the retrieval of sea
surface wind elds using X-band TerraSAR-X data. Int. J. Remote Sens. 33 (23),
73107336.
Sousa, J.J., Magalhes, L.M., Ruiz, A.M., A.M.R., Cardoso, G., 2013. The viStaMPS tool
for visualization and manipulation of time series interferometric results, series
interferometric results. Comput. Geosci. 52, 409421, http://dx.doi.org/10.1016/
j.cageo.2012.11.012.
Tamura, H., Mori, S., Yamawaki, T., 1978. Textural features corresponding to visual
perception. IEEE Trans. Syst. Man Cybern. 8 (6), 460473.
Tehrany, M.S., Pradhan, B., Jebu, M.N., 2013a. A comparative assessment between
object and pixel-based classication approaches for land use/land cover
mapping using SPOT 5 imagery. Geocarto. Int., http://dx.doi.org/10.1080/
10106049.2013.768300.
Tehrany, M.S., Pradhan, B., Jebur, M.N., 2013b. Spatial prediction of ood susceptible
areas using rule based decision tree (DT) and ensemble bivariate and
multivariate statistical models. J. Hydrol. 504, 6979, http://dx.doi.org/
10.1016/j.jhydrol.2013.09.034.
Treitz, P.M., Soulis, E.D., Howarth, P.J., Kouwen, N., 1996. Texture processing of
synthetic aperture radar data using second-order spatial statistics. Comput.
Geosci. 22 (1), 2734.
Vidal, A., Moreno, M.R., 2011. Change detection of isolated housing using a new
hybrid approach based on object classication with optical and TerraSAR-X
data. Int. J. Remote Sens. 32 (24), 96219635.
Wei, L., Hu, Z., Guo, M., Jiang, M., Zhang, S., 2012. Texture feature analysis in oil spill
monitoring by SAR image. In: Proceedings of IEEE 20th International Conference on Geoinformatics (GEOINFORMATICS), 2012, pp. 16.
Yenugu, M.M., Marfurt, K.J., Matson, S., 2010. Seismic texture analysis for reservoir
prediction and characterization. Lead. Edge 29 (9), 11161121.
Youssef, A., Pradhan, B., Hassan, A.M., 2011. Flash ood risk estimation along the St.
Katherine road, southern Sinai, Egypt using GIS based morphometry and
satellite imagery. Environ. Earth Sci. 62, 611623.
Zaleha, K., Roswati, M.A., Iwasaki, N., 2006. Distribution of some species of
harpacticoid copepods in east coast of Peninsular Malaysia. Coast. Mar. Sci.
30 (1), 141145.
Zan, F.D., Guarnieri, A.M., Rocca, F., Tebaldini, S., 2008. Sentinel-1 radar interferometry applications. In: Proceedings of 2008 7th European Conference on
Synthetic Aperture Radar (EUSAR), VDE, 2008, pp. 14.
Zhang, J., Marszaek, M., Lazebnik, S., Schmid, C., 2007. Local features and kernels
for classication of texture and object categories: a comprehensive study. Int. J.
Comput. Vis. 73 (2), 213238.