Nikiforov
Vladimir O.
D.Sc., Prof.
ON DISTORTION ASSESSMENT OF IMAGES MASKING WITH M-MATRICES
Read the full article
';
Abstract
The article deals with results of the first experiments on images processing coded with unique M-matrices of the 22 nd order. General data on M-matrices and their place among known types of structured orthogonal matrices are given, and benefits from their implementation for image protection against unauthorized access are briefly substantiated. The extent of distortions influence, caused by lossy compression algorithms implementation and by modeling of a part of information flow loss is estimated as a result of carried out experiments. Peak signal to noise ratio metric and structure similarity index are used as objective figures of arising changes in recovered images after distortion. The main purpose of the paper is to define overall appropriateness of M-matrices implementation for images masking during their transmission or storage in contemporary technical systems.
