139.

[Cover] [Contents] [Index]

Page 222

Figure 5.21 (a) Test image derived from Brodatz (1966). (b) After noise fading process with SNR=5 (see text for explanation).

ture, and Carr (1996) proved a Fortran computer program to compute the semivariogram-based measures of texture.

The second approach to semivariogram-based texture definition is to select a particular model to fit the semivariogram, and then to use selected parameters of the model as measures of texture. For example, Ramstein and Raffy (1989) use an exponential model for the semivariogram, and derive a parameter which they call a′ from this model. This parameter is used as a measure of texture. Atkinson and Lewis (2000) comment that the automatic model-fitting approach is unreliable, as the choice of model may not be appropriate for the region covered by the image. Further details

[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