>  Docs Center  >  Libraries  >  ASTROLIB  >  MEDARR






      Compute the median at each pixel across a set of 2-d images


      Each pixel in the output array contains the median of the
      corresponding pixels in the input arrays. Useful, for example to
      combine a stack of CCD images, while removing cosmic ray hits.
      This routine has been mostly obsolete since V5.6 with the introduction
      of the DIMENSION keyword to the intrinsic MEDIAN() function. However,
      it is still useful for integer images if bad pixels need to be flagged
      in a mask parameter. (For floating point images, it is much
      faster to set invalid pixels to NaN values.)

Calling Sequence

      MEDARR, inarr, outarr, [ mask, output_mask ]


      inarr -- A three dimensional array [Nx,Ny, N] containing the input
                images. Each image is size Nx by Ny, and there are N

Optional Input

      mask -- Same structure as inarr, byte array with 1b where
                pixels are to be included, 0b where they are to be
                excluded. For floating point images, it is much faster to
                set masked pixels in inarr equal to !VALUES.F_NAN (see below),
                rather than use the mask parameter.


      outarr -- The output array. It will have dimensions equal to the
                first two dimensions of the input array.
      output_mask -- Same structure as outarr, byte array with 1b where
                      pixels are valid, 0b where all the input pixels
                      have been masked out.


        This procedure is *SLOW* when using the Mask parameter because it has
        to loop over each pixel of the image.


      Suppose one wants to combine three floating point 1024 x 1024 bias
      frames which have been read into the IDL variables im1,im2,im3
      IDL> bigim = fltarr(1024,1024,3) ;Create big array to hold images
      IDL> bigim[0,0,0] = im1 & bigim[0,0,1] = im2 & bigim[0,0,2] = im2
      IDL> medarr, bigim, avgbias
      The variable avgbias will be the desired 1024x 1024 float image.


      If the MASK parameter is not set, then MEDARR is just a wrapper for
      MEDIAN(/EVEN, dimension = 3). If the MASK parameter is set,
      a scalar median function over the third dimension is looped over
      each pixel of the first two dimensions. The /EVEN keyword is used
      with MEDIAN (which averages the two middle values), since this avoids
      biasing the output for an even number of images.
      Any values set to NAN (not a number) are ignored when computing the
      median. If all values for a pixel location are NAN, then the median
      is also returned as NAN.

Modification History

      Written by Michael R. Greason, STX, 12 June 1990.
      Don't use MEDIAN function for even number of images.
          W. Landsman Sep 1996
      Mask added. RS Hill, HSTX, 13 Mar. 1997
      Converted to IDL V5.0 W. Landsman September 1997
      Use /EVEN keyword to MEDIAN W. Landsman September 1997
      Rearranged code for faster execution W. Landsman January 1998
      Faster execution for odd number of images W. Landsman July 2000
      V5.4 fix for change in SIZE() definition of undefined variable
                W. Landsman/E. Young May 2001
      Use MEDIAN(/DIMEN) for V5.6 or later W. Landsman November 2002
      Use keyword_set() instead of ARG_present() to test for presence of mask
          parameter D. Hanish/W. Landsman June 2003
      Assume since V5.6 W. Landsman Feb 2004

© 2019 Harris Geospatial Solutions, Inc. |  Legal
My Account    |    Store    |    Contact Us