273.

[Cover] [Contents] [Index]

Page 44

Figure 1.27 (a) Test SAR image showing speckle noise, (b) Filtered image using 5×5 mean filter, (c) Filtered image using window size of 9×9.

(1.36)

where zij is the noise affected image pixel at location (i, j), xij is the signal that one wishes to recover, and vij is multiplicative noise. In order to simplify the expression, the subscripts are omitted in the following discussion. Normally, the mean μv of noise v takes a value of unity (Lee, 1986), and the noise standard deviation σv can be obtained from a local window located over a homogeneous area of the image, such as a water surface. The idea of extracting σv in this way is based on the second assumption, that the parameters x and v in Equation (1.36) are independent. The mean of a noise fading signal μz can be expressed by:

(1.37)

where μx and μv are the means of x and v, respectively. Using Equation (1.36), the variance σz2 of z can be expressed as:

[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