BSIAF: A Blockchain-Augmented Forensic Authentication Framework for Tamper-Evident Image Provenance
Journal of Intelligent Decision Making and Information Science · 15.09.2026 · 3 (19s) · 287-309
DOI: 10.17762/jidmis.v3.6168 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 15.09.2026. Verfasst von Koushik R. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The proliferation of sophisticated image manipulation tools and AI-generated synthetic media has significantly challenged the reliability and provenance of digital images used as evidence. Conventional blockchain-based image authentication approaches that rely solely on cryptographic file hashing are highly sensitive to benign transformations, such as JPEG re-encoding, which can lead to false authentication alerts. To address this limitation, this paper proposes the Blockchain-Based Secure Image Authentication and Forensics (BSIAF) framework, integrating dual-layer SHA-256 and perceptual hash (pHash) verification, on-chain EXIF metadata anchoring, multi-signal forensic analysis, and a Merkle-rooted tamper-evident audit trail. The framework combines exact file-integrity verification with perceptual similarity assessment to distinguish benign image transformations from potential manipulations. Additionally, metadata anchoring enables the identification of metadata modifications independently of pixel-content changes, while the forensic fusion pipeline integrates Error Level Analysis, SIFT-RANSAC, CNN-based analysis, and metadata comparison to support image tampering detection and localization. A blockchain-based smart contract maintains image integrity records and supports verifiable provenance, while Merkle-rooted audit logs strengthen the integrity of forensic investigation records. The proposed framework is evaluated using publicly available image-forensics datasets and experiments involving image tampering, JPEG compression, metadata modification, splicing, deepfake detection, and blockchain performance. The findings demonstrate the potential of integrating blockchain-supported provenance with complementary image-forensics techniques to improve the reliability, traceability, and auditability of digital image authentication. The study also identifies limitations in deepfake classification under compression and highlights opportunities for improving forensic robustness and scalability in future work.
Zitieren
APA
R, K. (2026). BSIAF: A Blockchain-Augmented Forensic Authentication Framework for Tamper-Evident Image Provenance. Journal of Intelligent Decision Making and Information Science, 3(19s), 287-309. https://doi.org/10.17762/jidmis.v3.6168
Vancouver
R K. BSIAF: A Blockchain-Augmented Forensic Authentication Framework for Tamper-Evident Image Provenance. Journal of Intelligent Decision Making and Information Science. 2026;3(19s):287-309. doi:10.17762/jidmis.v3.6168
BibTeX
@article{R2026,
author = {R, Koushik},
title = {BSIAF: A Blockchain-Augmented Forensic Authentication Framework for Tamper-Evident Image Provenance},
journal = {Journal of Intelligent Decision Making and Information Science},
year = {2026},
volume = {3},
number = {19s},
pages = {287--309},
doi = {10.17762/jidmis.v3.6168}
}
EUVIMED