Universal Method for Detecting Violations in the Integrity of a Digital Image Based on Analysis of Blocks of its Matrix

The requirement to provide the content expertise (particularly digital video) in real time is be-coming critical. Thus, the aim of the work is to increase the efficiency of identifying the fact of violation of image integrity by developing a universal expert method with low computational complexity....

Full description

Saved in:
Bibliographic Details
Main Authors: Bobok I., Kobozeva A.
Format: Article
Language:English
Published: Academy of Sciences of Moldova 2023-11-01
Series:Problems of the Regional Energetics
Subjects:
Online Access:https://journal.ie.asm.md/assets/files/08_04_60_2023.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The requirement to provide the content expertise (particularly digital video) in real time is be-coming critical. Thus, the aim of the work is to increase the efficiency of identifying the fact of violation of image integrity by developing a universal expert method with low computational complexity. This aim was achieved by using a new approach developed by the authors earlier that based on the properties of the dependence of the frequency index of the singular vector of the image matrix on its number, adapted for the case of a block organization of expertise. The most important theoretical result of the work is the higher rate of growth of the linear approxi-mation of the dependence of the block-average values of the frequency indexes of singular vec-tors on its number, which was established for the original content, as compared to the non-original content. The significance of the obtained results is that the developed expert method, being a block one, has insignificant computational complexity – operations for an image matrix, which makes it promising for working with digital content, in particular with vid-eo, in real time. At the same time, the algorithmic implementation of the method made it possible to increase the efficiency of detecting violations of image integrity by reducing type I errors by 2% compared to the best analogue; type II errors remained at the same level. The versatility of the method is confirmed by its high efficiency regardless of the specifics of the perturbation ef-fect, including in conditions of minor disturbances.
ISSN:1857-0070