DN MED - Denoise - Median Filtering

(3/3/11)

PURPOSE

Denoise an image or volume using a median filter.   Example.

SEE ALSO

CE FIT [Contrast Enhancement - FIT the histogram]
CE GNC [Contrast Enhancement - Graduated Non Convex restoration]
DN AD [Denoise - Anisotropic Diffusion]
DN MS [Denoise - Mean Shift]
DU [Dust removal]
FQ [Filter - Quick Fourier, High, low pass, etc., Padded ||]

USAGE

.OPERATION: DN MED

.INPUT FILE: PIC001
[Enter name of image/volume to be processed.]

.OUTPUT FILE: PIC002
[Enter name of file receiving the output image/volume.]

.FILTER LENGTH: 5
[Enter filter length, must be odd number > 2.]

.BOX OR CROSS (B/C): C
[Choose the shape of filter support. See notes for explanation.]

NOTES

  1. Two or three dimensional median filter is applied to the picture. This is a non-linear filter. The longer the filter length the stronger the filtration. The 'Box or Cross' question refers to the shape of filter support. Both of them result in different kind of noise structures being suppressed in the picture. Basicaly a stronger filtration can be achieved by a box filter. The aim of a median filter is to suppresses the 'pink' noise in a picture leaving edges basically unaffected at the same time. A detailed description of median filter features can be found in: Two-dimensional Digital Signal Processing II : Transforms and Median Filters, Editor: Huang)

  2. Implemented by Paul Penczek.

  3. Previously this operation was called: 'CE MED'

SUBROUTINES

MEDIAN, MD2, MD3

CALLER

UTIL2

© Copyright Notice /       Enquiries: spider@wadsworth.org