218.

[Cover] [Contents] [Index]

Page 294

Figure 7.4 Method of crossover used by GA to generate new search strings.

(1989) suggests that a mutation rate of 0.001 can achieve a relatively good result. This value is therefore used in the study reported here.

All GA operations discussed above have the same purpose, which is to test different yet high performance similarity templates or schema. A schema is a subset of strings with similarities at certain string positions (Holland, 1975). For example, suppose there are two individuals; individual A has a binary coded value 1010, and individual B contains the coding value 1111. One possible schema for these two individuals can therefore be 1*1*, where the symbol * means ‘don’t care’. One interesting and

Table 7.3 Information classes and number of ground reference pixels

Class

Information class

Number of pixels

1

Metarhyodacids (rhd)

429

2

Granodiorities

43,496

3

Gneiss basitic

5,807

4

Metavolcanics

8,157

5

Gneiss acidic

1,998

6

Pluvial flood plain

703

7

Quaternary deposits

2,529

8

Granites

385

[Cover] [Contents] [Index]


Classification Methods for Remotely Sensed Data
Classification Methods for Remotely Sensed Data, Second Edition
ISBN: 1420090720
EAN: 2147483647
Year: 2001
Pages: 354

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net