are identical. Table 3 shows that by this criterion the Segmentation-
based Fusion has the highest rank.
R G B I1 I2 I3
PCA 026 | 027 | 027 | 0.18 | 0.17 | 0.23
Gram Schmidt | 0.26 | 0.26 | 0.27 | 0.19 | 0.19 | 0.26
Wavelet 0.49 | 0.34 | 0.32 | 0.45 | 0.15 | 0.42
Seg.-based 0.51 | 0:50 | 0[50 | 0:50 | 0.50 | 0.25
Table 3: Universal Quality Index according to (Wang and Bovik
2002) for different channels of the pansharpened images
5-4 Classification Results
As we use hyperspectral data mostly for material classication, we
prefer to compare the pansharpening methods with respect to the
result of a supervised classification which they provide. Here we
present only first results of an ongoing research.
Before classifying the different pansharpened images, training re-
gions for different surface materials were defined. We have used
18 training regions for 12 roof materials plus vegetation. In or-
der to determine the accuracy of the classification, the results are
compared with ground truth information. As test regions for the
evaluation we use the training regions plus 33 additional regions.
The ground truth has to be improved in the future, however.
een
Fo male
HH Poo mera 3
x iat
(a) RGB Image (b)
" =
(e) Wavelet Fusion
(f) Segmentation-based Fusion
Figure 9: Classification Example
Figures 9 gives an impression about the results of a SAM clas-
sification. As expected, the Segmentation-based Fusion yields
a good homogeneity of the segments. More significant are the
evaluation measures which are reported in Table 4. The Overall
Accuracy gives the ratio between correctly classified pixels and
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
the total number of pixels. The so-called kappa coefficient addi-
tionally takes into account possible chance correct classifications
(Cohen 1960). The results in Table 4 favour the Wavelet and the
Segmentation-based Fusion — inspite of their minor visual qual-
ity.
Fusion K Overall
Method Accuracy
PCA 0.70 | 74.7
Gram Schmidt | '0.72 | 76.7 %
Wavelet 0°82 | 853%
Seg.-based 0.80 | 84.0%
Table 4: Accuracy Measures of the Classification Results based
on different Fusion Methods
6 CONCLUSIONS AND FUTURE WORK
Pansharpening in general not only affects the overall brightness
of hyperspectral data, but also the shape of the hyperspectral sig-
natures, which is important for material classification. In order
to distort the original data as few as possible, we propose to per-
form in a first step a segmentation of the panchromatic image,
employing map vector data in addition if available. In a second
step the data are resampled on the finer grid, whereby the inter-
polation is performed only by means of data pixels which are
located completely in the respective segment. A comparison of
different pansharpening methods reveals very different rankings
dependent on the quality criteria. We think that our method has a
good potential for the pansharpening for classification purposes.
Further investigations will be based on better ground truth in-
formation and more sophisticated classification methods. They
should also comprise a more detailed scrutiny of misclassified
regions.
REFERENCES
Cohen, J., 1960. A coefficient of agreement for nominal scales.
Educational and Psychological Measurement, 20, 37-46
Hirschmugl, M., Gallaun, H., Perko, R., and Schardt, M. 2005:
*Pansharpening"-Methoden für digitale, sehr hoch auflósende
Fernerkundungsdaten. In: Beitráge zum 17. AGIT Symposium,
Salzburg, Austria, July 06-08 2005.
Laben, C.A. and Brower, B.V. 2000: Process for enhancing the
spatial resolution of multispectral imagery using pansharpening.
United States Eastman Kodak Company (Rochester, New York).
US Patent 6011875.
Mallat, S., 2009. A Wavelet Tour of Signal Processing. Elsevier
Inc.
Pohl, C. and van Genderen, J.L. 1998: Multisensor image fusion
in remote sensing: concepts, methods and applications. Interna-
tional Journal of Remote Sensing, vol.19, no.5, pp.823- 854, Mar.
1998.
Ranchin T., Wald L., 2000: Fusion of high spatial and spectral
resolution images: the ARSIS concept and its implementation.
Photogrammetric Engineering and Remote Sensing, 66(1), 49-61
Wang, Z. and Bovik, A. 2002: A Universal Image Quality Index.
IEEE Signal Processing Letters, vol. 9 no. 3 p. 81-84 2002
Ecognition User Guide http://www.gis.unbc.ca/help/software/-
ecognition4/ELuserguide.pdf visited in April 2012