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
Melgani, F., and L. Bruzzone, 2004. Classification of
machines, IEEE Transactions on Geoscience and Remote
Sensing, 42(8):1778—1790.
Platt, J., 2000. Probabilities for SV machines, in Advances in
Large Margin Classifiers, Smola, A., Bartlett, P., Scholkopf, B.
and Schuurmans, D., Eds. Cambridge, MA: MIT Press.
hyperspectral remote sensing images with support vector
Spiegelhalter, D. J. and Knill-Jones, R.P., 1983. Statistical and
Knowledge-Based Approaches to Clinical Decision-Support
Systems, with an Application in Gastroenterology. Journal of
the Royal Statistical Society. Series A (General) Vol. 147, No. 1
(1984), pp. 35-77.
Table 1 Classification results using different data for Beijing area (all in percent)
SAR alone TM alone Combined
classification by
WOE
PA UA PA UA PA UA
Agricultural 37.62 41.87 91.90 71.37 94.60 74.13
Builtup area 74.64 67.81 81.40 74.67 91.23 75.34
Woodland 54.21 43.87 69.84 91.08 68.86 91.56
Water 44.49 81.71 96.40 93.24 93.64 98.88
Bareland 42.15 42.11 78.06 85.41 79.78 92.06
Overall Accuracy 51.77 81.71 84.28
Kappa 38.44 76.94 80.16
Table 2 Classification results using different data for Hengshui area (all in percent)
ETM- alone SPOT 5 alone Combined
classification by
WOE
PA UA PA UA PA UA
Residential 87.63 88.95 88.05 89.85 90.73 90.56
Bareland 60.14 41.11 73.46 71.67 73.20 70.04
Cotton 86.47 98.05 90.71 99.56 92.69 99.55
Water 98.88 90.98 99.85 89.78 99.96 91.00
Wheat 88.45 92.06 89.00 84.90 95.69 93.50
Reed 85.54 81.77 83.79 58.77 89.11 78.25
Grass 47.34 59.79 48.41 75.10 48.91 80.01
Orchard 72.15 81.04 42.48 63.23 79.11 84.49
Overall accuracy 84.20 84.98 90.12
Kappa 80.59 81.45 87.67
Figure 1 Portions of land cover classification results using difference data combinations: (a) SAR data alone; (b) Landsat TM alone;
(c) Combined TM and SAR data using WOE.
