5. CONCLUSIONS
In this paper, the integration of FS using GA method and the
MCS techniques with DS algorithm has been proposed and
implemented for classifying different combined datasets of
multi-date ENVISAT/ASAR, ALOS/PALSAR and Landsat 5
TM- images. Results of classification revealed that the
proposed (FS-GA-DS model) method outperform both classical
non-FS and FS-GA method. The FS-GA-DS model always gave
significantly higher accuracy than any single best classifier. The
FS-GA-DS method produced the highest overall classification
accuracy of 88.29% for the largest combined dataset consisted
of original SAR and optical images, their textural information
and NDVI. The FS-GA method is also an effective method for
classification of multisource remote sensing data, though it is
less accurate than the FS-GA-DS method. Finally, the
classification results confirm the advantages of multisource
remote sensing data, particularly the integration of SAR and
optical data, for land cover mapping. The combination approach
often performed better than any single typed dataset.
6. REFERENCES
Benediktsson, J.A., Chanussot, J. and Fauvel, M., 2007.
Multiple Classifier Systems in Remote Sensing: From Basics to
Recent Developments: MCS 2007, LNCS 4472, Springer
Verlag, Berlin 2007, pp. 501-512.
Bruzzone, L., Marconcini, M., Wegmuller, U., Wiesmann, A.,
2004. An advanced system for the automatic classification of
multitemporal SAR images. [EEE transaction on Geoscience
and remote sensing, 42(6), pp. 1321 — 1334.
Chu, H.T., and Ge, L., 2010. Land cover classification using
combination of L- and C-band SAR and optical satellite images.
In: 31” Asian conference of Remote Sensing (ACRS), Hanoi, 1-5
November.
Dixon, B., and Candade, N., 2008. Multispectral land use
classification using neural networks and support vector
machines: one or the other, or both. International Journal of
Remote Sensing, 29(4), pp. 1185-1206.
Du, P., Zhang, W., and Sun, H., 2009. Multiple Classifier
Combination for Hyperspetral Remote Sensing Image
Classification. MCS 2009, LNCS 5519, Springer-Verlag Berlin
Heidelberg. pp.52-61.
Erasmi, S., and Twele, A., 2009. Regional land over mapping in
the humid tropics using combined optical and SAR satellite data
— a case study from Central Sulawesi, Indonesia. International
Journal of Remote Sensing, 30(10), pp. 2465-2478.
Huang, C.L. and Wang, C.J, 2006. A GA-based feature
selection and parameters optimization for support vector
machines. Expert Systems with Applications, 31, 231-240.
Kavzoglu, T., and Mather, P. M., 2003. The use of
backpropagating artificial neural network in land cover
classification. International journal of Remote sensing, 24(23),
pp. 4907-4938.
Kavzoglu, T., and Colkesen, L, 2009. A kernel functions
analysis for support vector machines for land cover
classification, International journal of Applied Erath
Observation and Geoinformation, 11, pp: 352 — 359.
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
Lee, S., and Lathrop, R. G., 2006. Subpixel analysis of Landsat
ETM- using Self-Organizing Map (SOM) neural networks for
urban land cover characterization. IEEE Transactions on
Geoscience and Remote Sensing, 44(6), pp. 1642-1654.
Lu, D., and Weng, Q., 2007. A survey of image classification
methods and techniques for improving classification
performance. International Journal of Remote Sensing, 44(5),
pp: 823-870.
Salah, M., Trinder, J. and Shaker, A., 2009. Evaluation of the
self-organizing map classifier for building detection from lidar
data and multispectral aerial images. Journal of Spatial
Science, 54, pp. 15-34.
Salah, M., Trinder, J., Shaker, A., Hamed, M., and Elsagheer,
A., 2010. Intergrating Multiple Classifiers with Fuzzy Majority
Voting for Improveed Land cover Classifiation. In: /APRS, Vol.
XXXVIII, Part 3A — Saint-Mandé, France, September 1-3.
Sheoran, A., Haack, B., and Sawaya, S., 2009. Land cover/use
classification using optial and quad polarization radar imagery.
ASPRS 2009 Annual Conferencce, Baltimore, Maryland, USA,
9 — 13 March.
Soria-Ruiz, J., Fernadez-Ordonez, Y., and Woodhouse, I. H.,
2010. Land-cover lassification using radar and optical images: a
case study in Central Mexico. International Journal of Remote
Sensing, 31(12), pp. 3291-3305.
Trinder, J. C., and Salah, M., 2010. Combining statistical and
neural classifiers using Dempster-Shafer theory of evidence for
improved building detection. In: Proceeding of 1 3
Australasian Remote Sensing & Photogrammetry Conference
(15^ ARSPC), Alice Springs, Australia, 13-17 September.
Tso, B., and Mather, P., 2009. Classification Methods for
remotely sensed data, Second edition, CRC Press, Taylo &
Francis Group.
Waske, B., and Benediktsson, J.A., 2007. Fusion of Support
Vector Machines for classification of multisensor data, /EEE
Transactions on Geosciences and Remote Sensing, 45(12) pp.
4842-4845.
Zhou, M., Shu, J. and Chen, Z., 2010. Classification of
hyperspetral remote sensing image based on Genetic Algorithm
and SVM. Remote Sensing & Modeling of Ecosystems for
Sustainability VII, Proc. of SPIE Vol.7809, 78090A,
doi:10.1117/12.860153.
7. ACKNOWLEDGEMENT
ALOS/PALSAR Level 1.1 products were processed by
ERSDAC, Japan. Copyright of raw data belongs to METI and
JAXA.
ENVISAT/ASAR were kindly provided by the European Space
Agency (ESA).
Landsat 5 TM Level 1T products were processed and provided
by the USGS Earth Resources Observation and Science Center.