Singular Value Decomposition

'Singular value decomposition, or SVD, consists of the reduction of a singular matrix into the product of orthogonal and square matrices.

Discovered in the 19th century, this technique has recently discovered applications in signal error-reduction and in watermarking.