Back to search

University of Ioannina

Algorithmic techniques for information hiding and digital object protection

Abstract

dc:description

The widespread adoption of digital technologies and the continuous growth of multimedia content distribution have significantly increased the need for effective mechanisms capable of protecting digital objects against unauthorized modification, forgery, ownership violations, and content manipulation. Images, audio recordings, videos, and electronic documents are extensively utilized in domains such as digital forensics, healthcare, surveillance, and electronic communication, where authenticity and integrity are of critical importance. Consequently, the development of efficient information hiding and watermarking techniques for digital object protection remains an important research challenge. For authenticity preservation, a collection of robust watermarking models is developed for images, audio signals, videos, and PDF documents. For images, watermark information is initially represented as integer sequences and encoded through Self-Inverting Permutations (SiPs), which are subsequently embedded in the frequency domain. The first proposed model employs repetitive embedding of a single SiP watermark across multiple image regions to improve robustness against cropping and partial-content attacks. Building upon this concept, a code-based multiple watermarking framework is introduced, where collections of integers are encoded as multiple SiPs and embedded simultaneously, increasing both watermark capacity and security while maintaining robustness against common signal-processing and geometric attacks. These contributions are further unified within the ROMFAIA framework, which combines robust authenticity preservation with fragile integrity verification for forensic medical images, and are subsequently extended to blockchain-based digital asset protection through the ProtAIN framework for secure NFT deployment. For audio signals, a dual-layer watermarking methodology is introduced by combining image watermarking and audio least significant bit embedding. For videos, a code-based hash-chain watermarking framework is proposed, combining frequency-domain watermarking with cryptographic frame-audio verification. For PDF documents, dynamically generated QR codes together with scrambling and permutation mechanisms are employed to support authenticity verification while preserving document readability and visual quality.For integrity preservation, fragile watermarking techniques are proposed for images, audio signals, and videos. For images, a cryptographic integrity-verification framework is introduced based on inter-block verification relationships, enabling accurate tamper detection and localization of manipulated image areas while maintaining high visual quality. For audio signals, integrity preservation is achieved through the deployment of hash-based verification information embedded within audio samples, allowing the identification and localization of modified audio segments. For videos, two complementary integrity-preservation frameworks are developed. The first employs cryptographic hash chains embedded across video frames and synchronized with the audio stream, while the second introduces a cross-frame permutation framework capable of detecting addition, deletion, and copy-move manipulations. The proposed methodologies support reliable content verification and localization of modified regions while maintaining low computational complexity. The proposed models are assessed in terms of imperceptibility, robustness, computational efficiency, and tamper detection performance using metrics including PSNR, SSIM, BER, NC, Extraction Rate, Precision, Recall, F1-score, Accuracy, and execution time. The obtained results demonstrate high visual quality, strong robustness against common attacks, efficient computational performance, and accurate tamper detection and localization capabilities. Comparative evaluations with representative approaches from the literature further confirm the effectiveness and competitiveness of the proposed methodologies.Overall, this dissertation contributes a unified collection of watermarking and information hiding techniques for authenticity and integrity preservation in digital media. By combining frequency-domain and spatial-domain embedding strategies with combinatorial encoding structures, cryptographic verification mechanisms, and blockchain technologies, the proposed approaches provide effective and scalable solutions for the protection of images, audio signals, videos, electronic documents, and NFT-based digital assets against forgery, manipulation, and unauthorized modification.

Degree

thesis:*
Grantor dc:publisher
University of Ioannina
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Vouronikos, Vasileios
  • Βουρονίκος, Βασίλειος

Subjects

dc:subject × 16

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/62310

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Vouronikos, Vasileios; Βουρονίκος, Βασίλειος. Algorithmic techniques for information hiding and digital object protection. University of Ioannina, 2026. http://hdl.handle.net/10442/hedi/62310