Image Processing Reference
In-Depth Information
Fig. 6.15 Accuracy of the neural network classification of eight V-I-S subtypes (Ridd et al. 1997 )
quite successful in characterizing urban/peri-urban environments even where
nearby mountains are in the data set. What is not known is the degree to which the
success stems from the hyperspectral data with a narrow band width and 20 m cell
size or the neural network classification.
6.5
Environmental Process Application of the V-I-S Model
The V-I-S model has been shown to be a suitable foundation for physical process
models. The two examples summarized here represent energy and moisture flux,
the two primary drivers in the urban ecosystem (and ecosystems generally).
6.5.1
ATLAS Data: Thermal Emittance per Land Cover Class
Gluch et al. ( 2006 ) utilized ATLAS (Advanced Thermal Land Applications Sensor)
10 m airborne data to determine daytime thermal emittance from eight V-I-S cover
types over Salt Lake Valley, Utah: two vegetation types, two impervious types, two
soil types, shade, and water. Clearly the V-I-S model of land cover is superior to
models or classifications of land use for energy analysis. Energy dynamics are tied
 
Search WWH ::




Custom Search