C
1 : flv )23.0
-2.0 p
fA*(v p) = Rs 2 20 : 2.0«fiv «3.0 (2)
0 : flv 0:20
where f(v,) = rainfall in centimeters of the pixel in a
raster image of rainfall.
Both of these examples have been developed on the basis of
a raster geometric image in which the pixel values can be
assigned real number values in the closed interval [0,1].
4. CONCLUSIONS
The development of representational methods for fuzzy
features offers the potential to improve the accuracy of
analytical results from GIS analysis. While the examples
presented are simplistic cases relying on spatial and
thematic attributes of geographic datasets in raster formats,
more complex fuzzy set functions can be formed in a
similar manner by changing the fuzzy interpolation
functions from linear to polynomial and other non-linear
functions. The use of spatial position as the defining
parameter of fuzziness allows the same function to be used
in a variety of image representations of the same feature.
Use of thematic values as the defining parameter necessarily
limits the function to a single image type. The use of a
comprehensive framework which supports multiple
representations and multiple data sources appears to hold
promise for applications requiring analysis of many datasets.
While the current trend toward feature-based
implementations (sometimes using object-oriented computer
science approaches) of GIS holds promise for better user
interfaces and the shielding of users from the complexities
of geometric manipulation of geographic features, the
strength of these approaches appear to be in allowing
multiple representations within a GIS for a single
geographical entity. Fuzzy sets offer one approach to
representing geographic features with indeterminate
boundaries with spatial, thematic, and in the future,
temporal attributes and relationships defining the extent of
the fuzzy set function.
Future research should include development of specific
applications to test the fuzzy set representations in the
feature-based approach. Among such applications,
representation of agricultural phenomena with fuzzy spatial
extents and fuzzy functions defined by thematic parameters
such as phosphorus, potassium, nitrogen, soil ph, and other
attributes show promise.
S. REFERENCES
Altman, D., 1994. "Fuzzy Set Theoretic Approaches for
Handling Imprecision in Spatial Analysis," International
Journal of Geographical Information Systems, Vol. 8, pp.
271-289.
883
Berry, B.J.L., 1964. Approaches to Spatial Analysis: A
Regional Synthesis. Annals of the Association of American
Geographers, Vol. 54, pp. 2-11.
Burrough, P.A., 1989. "Fuzzy Mathematical Methods for
Soil Survey and Land Evaluation," Journal of Soil Science,
Volume 40, pp. 477-492.
ERDAS, 1995. ERDAS Imagine 8.2 User Manuals,
ERDAS, Inc., Atlanta, Georgia.
Katinsky, M., 1994. Fuzzy Set Modelling in Geographic
Information Systems, Unpublished Master's Thesis,
University of Wisconsin-Madison, 46 p.
Sinton, D., 1978. "The Inherent Structure of Information as
a Constraint to Analysis: Mapped Thematic Data as a Case
Study," In G. Dutton, (ed), Harvard Papers on Geographic
Information Systems, Volume 6, Addison Wesley, Reading,
MA.
Usery, E.L., 1993. "Category Theory and the Structure of
Features in GIS," Cartography and GIS, Vol. 20, No. 1,
pp. 5-12.
Usery, E.L., 19942. "Display of Geographic Features from
Multiple Image and Map Databases," International Archives
of Photogrammetry, Volume 30-4, Commission IV
Symposium on Mapping and Geographic Information
Systems, Athens, Georgia, pp. 1-9.
Usery, E.L., 1994b. "Implementation Constructs for Raster
Features," Proceedings, ASPRS/ACSM Annual Convention,
Reno, Nevada, pp. 661-670.
Usery, E.L., 1996a. "A Conceptual Framework and Fuzzy
Set Implementation for Geographic Features," In Burrough,
P.A. and A.U. Frank, (eds.), 1996. Geographic Objects
with Indeterminate Boundaries, Taylor and Francis, Ltd,
London, In Press.
Usery, E.L., 1996b. "A Feature-Based Geographic
Information System Model," Photogrammetric Engineering
and Remote Sensing, In Press.
Wang, F., 1990. "Improving Remote Sensing Image
Analysis through Fuzzy Information Representation,"
Photogrammetric Engineering and Remote Sensing, Vol. 56,
No. 8, pp. 1163-1168.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996