arrow
Return

Robust watermarking for diffusion models using error-correcting codes and post-quantum key encapsulation

delete2026-01-20
delete0
PRE
AI
X
Xianglei Hu
B
Beining Zhang
M
Mawaheb Al-Dossari
N
N. S. Abd El-Gawaad
M
Mira Rakhimzhanova
A
Ahmad Saeed Khan *
DOI:10.3389/fphy.2025.1750515delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Critical infrastructures increasingly rely on AI-generated content (AIGC) for monitoring, decision support, and autonomous control. This dependence creates new attack surfaces: forged maintenance imagery, manipulated diagnostic scans, or spoofed sensor visualisations can trigger unsafe actions, regulatory violations, or systemic disruption. This paper proposes a post-quantum watermarking framework designed for critical infrastructure security. We embed robust provenance markers directly into the latent space of diffusion models, rather than at the pixel level, and reinforce them using error-correcting codes (ECC) to ensure watermark recoverability even after aggressive distortions such as compression, cropping, noise injection, and filtering. To secure watermark keys in transit and at rest, we integrate Kyber, a lattice-based key encapsulation mechanism standardised for post-quantum cryptography, to protect the watermark stream key against quantum-enabled interception. The resulting scheme (i) preserves visual fidelity, (ii) supports reliable forensic attribution and auditability under hostile conditions, and (iii) remains cryptographically secure in the post-quantum era. Experiments show that the proposed ECC-hardened latent watermarking achieves consistently high extraction accuracy across diverse attacks while maintaining image quality, outperforming state-of-the-art diffusion watermarking baselines. We position this watermarking-encryption pipeline as an enabling mechanism for privacy-aware traceability, zero-trust validation, and quantum-resilient content governance in next-generation critical infrastructure.
Keywords:
critical infrastructure security
diffusion model
diffusion models
error-correcting codes
infrastructure resilience
kyber
post-quantum cryptography
privacy and provenance

Journal

F
Frontiers in Physics
IF:
2.1
Papers:
243
Citations:
0

Organization

A
Astana IT University
Scholars:
265
Papers: 140
Citations: 1
K
king khalid university
Scholars:
1.7K
Papers: 1.4K
Citations: 0
O
orebro university
Scholars:
468
Papers: 260
Citations: 0
G
guangdong polytechnic normal university
Scholars:
658
Papers: 318
Citations: 0
researcher View more organizations
Cited Papers

Cited Papers

An Automated and Robust Image Watermarking Scheme Based on Deep Neural Networks
err2021-01-01
err0
errOAAI
errXin Zhong; Pei-Chi Huang; Spyridon Mastorakis; Frank Y. Shih
errShare
errSave
Adversarial text-to-image synthesis: A review
err2021-12-01
err90
errOAAI
errFrolov, Stanislav; Hinz, Tobias; Raue, Federico; Hees, Joern; Dengel, Andreas
errShare
errSave
Robust Blind Video Watermarking Based on Ring Tensor and BCH Coding
err2024-12-15
err0
PREAI
errWang, Jiayan; Zhao, Jing; Li, Li; Wang, Zichi; Wu, Hanzhou; Wu, Deyang
errShare
errSave
errShare
errSave
The Stable Signature: Rooting Watermarks in Latent Diffusion Models
err2023-10-01
err0
errOAAI
errPierre Fernandez; Guillaume Couairon; Hervé Jégou; Matthijs Douze; Teddy Furon
errShare
errSave
errShare
errSave
errShare
errSave
researcher View more