Computes the estimated covariance matrix of an image using the sliding window forumlation

dst = cv.covarianceEstimation(src, windowSize)

Input

• src The source image. Input image must be of a complex type and floating-point type. Input should be arranged as a 2-channels matrix size(src,3)==2 (corresponding to real and imaginary parts).
• windowSize The number of rows and cols in the window [windowRows,windowCols].

Output

• dst The destination estimated covariance matrix. Output matrix will be of size [windowRows*windowCols, windowRows*windowCols, 2] (channels correspond to real and imaginary parts).

The window size parameters control the accuracy of the estimation. The sliding window moves over the entire image from the top-left corner to the bottom right corner. Each location of the window represents a sample. If the window is the size of the image, then this gives the exact covariance matrix. For all other cases, the sizes of the window will impact the number of samples and the number of elements in the estimated covariance matrix.

Algorithmic details of this algorithm can be found at [1]. A previous and less efficient version of the algorithm can be found [2].

References

[1]:

O. Green, Y. Birk, "A Computationally Efficient Algorithm for the 2D Covariance Method", ACM/IEEE International Conference on High Performance Computing, Networking, Storage and Analysis, Denver, Colorado, 2013

[2]:

O. Green, L. David, A. Galperin, Y. Birk, "Efficient parallel computation of the estimated covariance matrix", arXiv, 2013