Bsoft function

img_correlation_coefficient_radial

Source:

src/img/img_combine.c

Author:

David Belnap

Description:

Calculates the correlation coefficient between two images and
applies a radial filter.

Algorithm:

	Calculates correlation coefficient between two images using the 

	equation:



                      Sum(Ai*Bi) - [Sum(Ai)*Sum(Bi)]/n

 CC = -----------------------------------------------------------------

      SQRT[Sum(Ai^2) - (Sum(Ai))^2/n] * SQRT[Sum(Bi^2) - (Sum(Bi))^2/n]



	where Ai and Bi are pixel values of images A and B, respectively,

	and n is the total number of pixels sampled.

	   A radial filter is applied.  The center of the image is needed and

	is read from the pA Bimage structure.  The two images must have the 

	same center (apply a translation shift if necessary).  The program 

	tests that each pixel is within the radii range.  If the pixel is

	within the range, then it is included in the CC calculation.

Arguments:

Bimage* pA	first image.
Bimage* pB	second image.
float radius_min	minimum radius for CC calculation, in pixels
float radius_max	maximum radius for CC calculation, in pixels

Returns:

float correlation coefficient.

Prototype:

float img_correlation_coefficient_radial(Bimage* pA, Bimage* pB, float radius_min, float radius_max)

Functions used:

balloc

Other objects included:

struct Bimage

Generated by bdoc.pl on Mon Jun 15 11:47:34 2009

Back to the Bsoft home