FSTDDEV

Description

FSTDDEV applies a standard deviation speckle filter on a SAR image. This filter is primarily used on SAR data to remove high-frequency noise (speckle), while preserving high-frequency features (edges).

Parameters

fstddev(fili, filo, dbic, dbib, dboc, flsz, imagefmt)

Name	Type	Caption	Length	Value range
FILI *	str	Input detected SAR image	1 -
FILO	str	Output filtered image	0 -
DBIC *	List[int]	Input raster channel	1 -	1 -
DBIB	List[int]	Input bitmap	0 - 1
DBOC *	List[int]	Output filtered image channels	1 -	1 -
FLSZ	List[int]	Filter size (pixels, lines)	0 - 2	1 \| 3 \| 5 \| 7 \| 9 \| 11 \| 13 \| 15 \| 17 \| 19 \| 21 \| 23 \| 25 \| 27 \| 29 \| 31 \| 33 Default: 7,7
IMAGEFMT	str	Image units	0 - 9	Amplitude, Power Default: Amplitude

* Required parameter

Parameter descriptions

FILI

Specifies the name of the input file that contains the image data to be filtered.

FILO

Specifies the name of the file that will hold the results. FILO could specify an existing file. If not specified, FILO is assumed to be the same as FILI. If the output file does not exist, it is created. If a new output file is specified, DBOC must be left blank.

DBIC

Specifies the channels in the input detected SAR image to be processed.

DBIB

Optionally specifies the bitmap under which the input raster should be processed. If no bitmap is provided, the entire window is processed.

DBOC

Specifies the output channel numbers to receive the filtered results. This channel must already exist within the file specified by the FILO parameter. If this parameter specifies the same channels as those defined as the input (DBIC), DBOC overwrites the input channels. When FILO specifies a file that does not exist, DBOC must be empty.

FLSZ

Optionally specifies the horizontal and vertical dimensions of the filter, in pixel and line units. Acceptable values are odd integers from 1 to 33, inclusively. The filter need not be square. Minimum filter sizes are 1x3 or 3x1; the default size is 7x7.

IMAGEFMT

Optionally specifies the image format of the SAR image, which defines how the correction factor should be computed. By default, the image is assumed to be in amplitude units.

Image units can be "Amplitude" (or equivalently "AMP") and "Power" (or equivalently "POW").

Details

FSTDDEV applies a standard deviation speckle filter on a SAR image. It is designed to smooth out noise while retaining edges or shape features in the image.

Different filter sizes will greatly affect the quality of processed images. If the filter is too small, the noise filtering algorithm is not effective. If the filter is too large, subtle details of the image will be lost in the filtering process. A 7x7 filter usually gives the best results.

The standard deviation filter model requires that the signal represent power. If the input image is in amplitude format, each pixel value will be squared to derive power, and the square root will be applied to the filtered result so that the output still has amplitude values.

Example

A standard deviation filter is applied to channel 1 of irvine.pix.

from pci.fstddev import fstddev

fili = 'irvine.pix'
filo = 'irvine.pix'
dbic = [1]               #Uses the elevation data.
dbib = [12]              #Process under bitmap 12.
dboc = [3]
flsz = []                #Defaults to 7x7 filter.
imgtyp = ''              #Defaults to Amplitude AMP

fstddev(fili, filo,dbic, dbib,dboc, flsz, imgtyp)

Algorithm

FSTDDEV performs spatial filtering on each individual pixel in an image using the gray-level values in a square window surrounding each pixel. The values for the filter size must be odd integers, and can be 1 to 33 pixels. All pixels are filtered. To filter pixels located near edges of the image, edge-pixels are replicated to provide sufficient data.


        +----------+
        | a1 a2 a3 |
        | a4 a5 a6 |     <---  Filter window 3 X 3
        | a7 a8 a9 |
        +----------+

The resulting gray-level value R for the smoothed pixel is:

        R = (Sum2 - nValidPixelsInWindow*Mean2)/( nValidPixelsInWindow -1)

where:

Sum2 = Addition of squares of valid pixels in a window.
nValidPixelsInWindow = Number of valid pixels in a window.
Mean2 = Square of the mean value in a window.

References

Jong-Sen Lee, "Digital Image Enhancement and Noise Filtering by Use of Local Statistics", IEEE Transactions on Pattern Analysis and Machine Intelligence , Vol. PAM1-2, No. 2, March, 1980.

J.S.Lee, "Refined Filtering of Image Noise Using Local Statistics" Computer Graphic and Image Processing 15, 380-389 (1981)

D.T. Kuan, A.A. Sawchuk, T.C. Strand, and P. Chavel, "Adaptive restoration of images with speckle," IEEE Trans. ASSP., Vol. 35, no. 3, pp. 373-383, March 1987.

A. Lopes, R. Touzi and E. Nezry, "Adaptive speckle filters and Scene heterogeneity", IEEE Transaction on Geoscience and Remote Sensing , Vol. 28, No. 6, pp. 992-1000, Nov. 1990.

V.S. Frost, J.A. Stiles, K.S. Shanmugan, and J.C. Holtzman, "A model for radar images and its application to adaptive digital filtering of multiplicative noise," IEEE Trans. Pattern AnalysisThe and Machine Intelligence , vol. 4, no. 2, pp. 157-166, March 1982.

A. Lopes, E. Nezry, R. Touzi, and H. Laur, "Structure detection and statistical adaptive speckle filtering in SAR images", International Journal of Remote Sensing, Vol. 14, No. 9, pp. 1735-1758, 1993.