INTECHAIGHT: A Blockchain-Integrated Framework for Authorship Verification and Originality Protection of Digital Images

¹ Department of CSE, Sathyabama Institute of Science and Technology, Chennai, India
² Department of CSE, Sathyabama Institute of Science and Technology, Chennai, India
³ Department of CSE, Sathyabama Institute of Science and Technology, Chennai, India

This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

The fast development of the digital image distribution has worsened the issues related to copyright violations, unlicensed use, and derivative plagiarism. At the same time, the existing centralized protection systems remain vulnerable to manipulations and rights claims. The framework presented in this paper is known as Intechaight, an implementation of blockchain in an automated digital image copyright protection system that is proposed to check image originality. The suggested system organizations a hybrid plagiarism detection model, combining perceptual hashing (pHash) to analyze structural similarities and CLIP-based embedding to identify semantic similarity and EfficientNet-B0 features to detect visual similarity, ensuring effective exercise to copy, transform or semantically derived images. In order to achieve traceability of the ownership, a frequencydomain DCT-based watermarking method that uses ReedSolomon error correction is utilized; using this allows the embedded ownership information to resist common image manipulations. A smart contract is used to store cryptographic hash of the original image, extracted fingerprint and embedded watermark immutably, stored on the zkSync blockchain along with the wallet address of the creator and the time they were created. It leads to the creation of a decentralized registry that provides tamper-resistant, verifiable authorship evidence, and supports automated adjudication and tracking and is able to produce legal evidence. As it is illustrated by experimental results, the proposed solution is able to attain a good balance between robustness, accuracy, and deployability, which makes it suitable to be used in real-life implementations of copyright protection of digital images.