Need help or advice ? Come to the Icy club ! - Every Wenesday morning from 9h30 to 12h30 - Francois Jacob Building - Main hall - Pasteur
Register

User reviews

This plugin is not rated yet


Please log-in to post a review

SSIM toolbox

by Yoann Le Montagner

The SSIM is an index measuring the structural similarity between two images. It is valued between -1 and 1. When two images are nearly identical, their SSIM is close to 1.

SSIM reference:
Z. Wang, A. C. Bovik, H. R. Sheikh, E. P. Simoncelli (2004),
Image quality assessment: from error visibility to structural similarity,
IEEE Transactions on Image Processing, 13(4), 600-612.

Publication Id
ICY-L4T2K7
See technical details
View complete changelog

Documentation

With this plugin, you can compute a structured similarity index (SSIM) between two sequences.

These functions can be accessed:

  • directly through the GUI (using a EzPlug interface),
  • through the protocol editor provided by the Protocols plugin,
  • from java (for plugin developers) or Javascript (using the Script Editor plugin): in these cases, see the documentation in the SSIMCalculator class for more details.

The SSIM is an index measuring the structural similarity between two images. It is valued between -1 and 1. When two images are nearly identical, their SSIM is close to 1.

Formula computing the SSIM between two sequences seq1 and seq2 at a given pixel or voxel P:

              2*mu1(P)*mu2(P) + C1         2*cov(P) + C2     
  SSIM(P) = ------------------------ x ----------------------
            mu1(P)^2 + mu2(P)^2 + C1   s1(P)^2 + s2(P)^2 + C2

With:

  • mu1(P) and mu2(P): mean value of seq1 and seq2 computed over a small XY window located around P
  • s1(P) and s2(P): standard deviation of seq1 and seq2 computed over the same window
  • cov(P): covariance between seq1 and seq2 computed over the same window
  • C1 = (K1*L)^2: regularization constant (should be as small as possible)
  • C2 = (K2*L)^2: regularization constant (should be as small as possible)
  • K1, K2: regularization parameters (must be >0)
  • L: dynamic range of the pixel values (example: L=255 if the sequence is 8 bit encoded)

The default window is a Gaussian window with standard deviation 1.5 along both the X and the Y axis.

Reference:
Z. Wang, A. C. Bovik, H. R. Sheikh, E. P. Simoncelli (2004),
Image quality assessment: from error visibility to structural similarity,
IEEE Transactions on Image Processing, 13(4), 600-612.

This current implementation sticks as much as possible to the Matlab SSIM implementation provided by these authors at:
https://ece.uwaterloo.ca/~z70wang/research/ssim/