Name

---

CORREL_IMAGES

Purpose

---

      Compute the 2-D cross-correlation function of two images

Explanation

---

      Computes the 2-D cross-correlation function of two images for
      a range of (x,y) shifting by pixels of one image relative to the other.

Calling Sequence

---

      Result = CORREL_IMAGES( image_A, image_B,
                        [XSHIFT=, YSHIFT=, XOFFSET_B=, YOFFSET_B=, REDUCTION=,
                        MAGNIFICATION=, /NUMPIX, /MONITOR )

Inputs

---

      image_A, image_B = the two images of interest.

Optional Input Keywords

---

      XSHIFT = the + & - shift to be applied in X direction, default=7.
      YSHIFT = the Y direction + & - shifting, default=7.
      XOFFSET_B = initial X pixel offset of image_B relative to image_A.
      YOFFSET_B = Y pixel offset, defaults are (0,0).
      REDUCTION = optional reduction factor causes computation of
                      Low resolution correlation of bin averaged images,
                      thus faster. Can be used to get approximate optimal
                      (x,y) offset of images, and then called for successive
                      lower reductions in conjunction with CorrMat_Analyze
                      until REDUCTION=1, getting offset up to single pixel.
      MAGNIFICATION = option causes computation of high resolution correlation
                      of magnified images, thus much slower.
                      Shifting distance is automatically = 2 + Magnification,
                      and optimal pixel offset should be known and specified.
                      Optimal offset can then be found to fractional pixels
                      using CorrMat_Analyze( correl_images( ) ).
      /NUMPIX - if set, causes the number of pixels for each correlation
                      to be saved in a second image, concatenated to the
                      correlation image, so Result is fltarr( Nx, Ny, 2 ).
      /MONITOR causes the progress of computation to be briefly printed.

Outputs

---

      Result is the cross-correlation function, given as a matrix.

Procedure

---

      Loop over all possible (x,y) shifts, compute overlap and correlation
      for each shift. Correlation set to zero when there is no overlap.

Modification History

---

      Written, July,1991, Frank Varosi, STX @ NASA/GSFC
      Use ROUND instead of NINT, June 1995, Wayne Landsman HSTX
      Avoid divide by zero errors, W. Landsman HSTX April 1996
Remove use of !DEBUG W. Landsman June 1997
      Subtract mean of entire image before computing correlation, not just
          mean of overlap region H. Ebeling/W. Landsman June 1998
      Always REBIN() using floating pt arithmetic W. Landsman Nov 2007