specific user. These segmentations are tailored to the users
since they commit fewer errors involving classes thematically
different from the users’ perspective. Critically, the standard,
geometric-only, metric M
g
indicated the segmentation derived
with a scale parameter of a value taken to be optimal regard-
less of the user needs in terms of land cover representation.
Therefore, it is advantageous to combine a thematic method
such as the
TSI
with a geometric method like Möller
et al.
’s
(2013) since together they find a new balance between over
and under-segmentation, where thematic errors (under-seg-
mentation) are considered according to the perspective of a
particular user.
The segmentation selected had a marked impact on the
accuracy of a land cover map produced following an object-
based approach to classification. The results show that the
incorporation of user-specific thematic information in image
segmentation quality assessment resulted in an increase of
classification accuracy for both users. Therefore, the as-
sessment of image segmentation quality with a geometric-
thematic method contributes to the production of land cover
maps tailored to specific user needs. Furthermore, as with
the geometric-only method, the proposed geometric-thematic
method can be used to assess candidate segmentations for use
in a classification whatever the segmentation method used
(e.g., single- and multi-scale).
Acknowledgments
The authors are grateful to Helena Rio-Maior for helping to
build the thematic similarity matrix for a wolf researcher
and to Markus Möller for providing R code. The authors
also thank to the anonymous referees for the comments that
helped improve the manuscript.
LISS-III
data were provided
by the European Space Agency. Hugo Costa was supported
by the Ph.D. Studentship number SFRH/BD/77031/2011 from
the “Fundação para a Ciência e Tecnologia” (
FCT
), supported
by the “Programa Operacional Potencial Humano” (
POPH
) and
the European Social Fund.
References
Ahlqvist, O., and M. Gahegan, 2005. Probing the relationship between
classification error and class similarity,
Photogrammetric
Engineering & Remote Sensing
, 71(12):1365–1373.
Albrecht, F., 2010. Spatial accuracy assessment of object boundaries
for object-based image analysis,
Proceedings of the Ninth
International Symposium on Spatial Accuracy Assessment in
Natural Resources and Environmental Sciences
, 20-23 July,
Leicester, United Kingdom (University of Leicester), pp. 13–16.
Anderson, J.R., E.E. Hardy, J.T. Roach, and R.E. Witmer, 1976.
Land Use and Land Cover Classification System for Use with
Remote Sensor
, U.S. Geological Survey Professional Paper 964,
Washington, D.C.
Arvor, D., L. Durieux, S. Andrés, and M.-A. Laporte, 2013. Advances
in Geographic Object-Based Image Analysis with ontologies:
A review of main contributions and limitations from a remote
sensing perspective,
ISPRS Journal of Photogrammetry and
Remote Sensing
, 82:125–137.
Baatz, M., and A. Schäpe, 2000. Multiresolution segmentation:
An optimization approach for high quality multi-
scale image segmentation,
Angewandte Geographische
Informationsverarbeitung XII
, Beitr ge zum AGIT-Symposium
Salzburg 2000 (Karlsruhe: Herbert Wichmann Verlag), pp. 12–23.
Blaschke, T., G.J. Hay, M. Kelly, S. Lang, P. Hofmann, E. Addink, R.
Queiroz Feitosa, F. van der Meer, H. van der Werff, F. van Coillie,
and D. Tiede, 2014. Geographic Object-Based Image Analysis -
Towards a new paradigm,
ISPRS Journal of Photogrammetry and
Remote Sensing
, 87:180–191.
Bouchon-Meunier, B., M. Rifqi, and S. Bothorel, 1996. Towards
general measures of comparison of objects,
Fuzzy Sets and
Systems
, 84(2):143–153.
Caetano, M., V. Nunes, and A. Nunes, 2009.
CORINE Land Cover 2006
for Continental Portugal
, Instituto Geográfico Português.
Clinton, N., A. Holt, J. Scarborough, L.I. Yan, and P. Gong, 2010.
Accuracy assessment measures for object-based image
segmentation goodness,
Photogrammetric Engineering & Remote
Sensing
, 76(3):289–299.
Costa, H., and G.M. Foody, 2013. Incorporating user requirements
into the assessment of image segmentations used in object-
based classification,
Proceedings of the Remote Sensing and
Photogrammetry Society Annual Conference (RSPSoc 2013):
Earth Observation for Problem Solving
, 04-06 September,
Glasgow, Scotland.
DeFries, R.S., and S.O. Los, 1999. Implications of land-cover mis-
classification for parameter estimates in global land-surface
models: An example from the simple biosphere model (SiB2),
Photogrammetric Engineering & Remote Sensing
, 65(9):1083–1088.
EEA, 2007.
CLC2006 Technical guidelines
, EEA Technical Report
17/2007, European Environment Agency, Copenhagen.
Foody, G.M., 2009. Classification accuracy comparison: Hypothesis
tests and the use of confidence intervals in evaluations of
difference, equivalence and non-inferiority,
Remote Sensing of
Environment
, 113(8):1658–1663.
Gao, Y., J.F. Mas, N. Kerle, and J.A.N. Pacheco, 2011. Optimal region
growing segmentation and its effect on classification accuracy,
International Journal of Remote Sensing
, 32(13):3747–3763.
Körting, T.S., L.M.G. Fonseca, and G. Câmara, 2013. GeoDMA -
Geographic Data Mining Analyst,
Computers & Geosciences
,
57:133–145.
Liu, C., P. Frazier, and L. Kumar, 2007. Comparative assessment of the
measures of thematic classification accuracy,
Remote Sensing of
Environment
, 107(4):606–616.
Mas, J.F., 2005. Change estimates by map comparison: A method to
reduce erroneous changes due to positional error,
Transactions
in GIS
, 9(4):619–629.
Möller, M., J. Birger, A. Gidudu, and C. Gl ßer, 2013. A framework
for the geometric accuracy assessment of classified objects,
International Journal of Remote Sensing
, 34(24):8685–8698.
Möller, M., L. Lymburner, and M. Volk, 2007. The comparison
index: A tool for assessing the accuracy of image segmentation,
International Journal of Applied Earth Observation and
Geoinformation,
9(3):311–321.
Müller, R., T. Krauss, M. Lehner, P. Reinartz, J. Forsgren, G. Rönnb ck,
and Å. Karlsson, 2009.
IMAGE 2006 European Coverage,
Methodology and Results
, DLR Report, 55 p.
Nosofsky, R.M., 1986. Attention, similarity, and the identification -
Categorization relationship,
Journal of Experimental Psychology:
General
, 115(1):39–57.
Quinlan, J.R., 1993.
C4.5: Programs For Machine Learning
, Morgan
Kaufmann Publishers, Inc., San Francisco, California, 302 p.
Rio-Maior, H., R. Godinho, M. Nakamura, and F. Álvares, 2012.
Comportamento social e espacial de um núcleo de 5 alcateias
no noroeste de Portugal,
Proceedings of the
III Iberian Wolf
Congress
, 23–25 November, Lugo, Galicia, Spain.
Smits, P.C., S.G. Dellepiane, and R.A. Schowengerdt, 1999. Quality
assessment of image classification algorithms for land-cover
mapping: A review and a proposal for a cost-based approach,
International Journal of Remote Sensing
, 20(8):1461–1486.
Stehman, S.V., 1999. Comparing thematic maps based on map value,
International Journal of Remote Sensing
, 20(12):2347–2366.
Tversky, A., 1977. Features of similarity,
Psychological Review
,
84(4):327–352.
Whiteside, T.G., S.W. Maier, and G.S. Boggs, 2014. Area-based
and location-based validation of classified image objects,
International Journal of Applied Earth Observation and
Geoinformation
, 28:117–130.
Zhang, Y.J., 1996. A survey on evaluation methods for image
segmentation,
Pattern Recognition
, 29(8):1335–1346.
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
June 2015
459
