Image Processing Reference
In-Depth Information
Path 1
MDM = Bedrock
Soil and
Bedrock
MDM = Mesic Built Materials
T2 = Low or Medium
Path 2
NDVI = Low or Medium
Land Use Built, Cemeteries ,
Golf Courses, Water, or Agriculture
Fig. 12.1
Schematic example diagram of expert system model described in text. Variable definitions
are: MDM - initial minimum distance to means classification result; T2 - ISODATA classification
for variance texture calculated from ASTER Band 2; NDVI - ISODATA classification for vegetation
index value; Land Use - categories extracted from land use vector dataset (figure published in Stefanov
and Netzband 2005, copyright Elsevier)
variance texture data, and a land use vector polygon dataset.
The land use data were acquired from the Maricopa
Association of Governments (MAG; Maricopa Association
of Governments
2000
) and are contemporaneous with both
the ASTER and MODIS data. The land use data are con-
structed from a combination of survey questionnaires, site
visits, and aerial photograph data. This dataset contains 46
separate land use categories which were aggregated to
seven for use in the expert system model: Open Residential,
Built, Cemeteries, Open Space, Golf Courses, Water, and
Agriculture. Incorporation of land use polygon data provides
additional discriminatory power for spectrally similar
pixels such as asphalt and bedrock. For example, a pixel classified as Asphalt
located within an Open Space polygon would be reclassified as Soil and
Bedrock. A series of decision rules were then constructed to recode misclassified
pixels in the MDM classification product. The MDM classes White Rooftops and
Blue Rooftops were also recoded into one class, Reflective Built Surfaces, within
the expert system model. The expert classification model was run using the area of
overlap of the MDM classification and the MAG land use dataset only (Fig.
12.2
).
The 15-class output of the expert classification model was then further aggregated
to 11 classes prior to accuracy assessment. Classes were aggregated if they were
functionally similar landscape elements (i.e. Canopied and Riparian Vegetation) to
the expert system
approach
facilitates the
integration of
different data
sources or
“knowledge” to
increase
classification
accuracy
