Facial Privacy Protection with Dynamic Multi-User Access Control for Online Photo Platforms

Future Internet, Год журнала: 2025, Номер 17(3), С. 124 - 124

Опубликована: Март 11, 2025

In the digital age, sharing moments through photos has become a daily habit. However, every face captured in these is vulnerable to unauthorized identification and potential misuse AI-powered synthetic content generation. Previously, we introduced SnapSafe, secure system for enabling selective image privacy focusing on facial regions single-party scenarios. Recognizing that group with multiple subjects are more common scenario, extend SnapSafe support multi-user protection dynamic access control designed online photo platforms. Our approach introduces key splitting control, an owner-centric permission granting revoking regions, request-based mechanism allowing initiate permissions. These features ensure remain protected while maintaining visibility of non-facial general viewing. To reproducibility isolation, implemented our solution using Docker containers. experimental assessment covered diverse scenarios, categorized as “Single”, “Small”, “Medium”, “Large”, based number faces photos. The results demonstrate system’s effectiveness across all test consistently performing encryption operations under 350 ms achieving average decryption times below 286 various sizes. key-splitting maintained 100% success rate configurations, revocation were executed efficiently server processing remaining 16 ms. validate capability managing practical usability contexts.

Язык: Английский

Enhancing Image Security with Memristor Driven Fractional Chaotic Systems and Secretary Bird Optimization

Journal of Machine and Computing, Год журнала: 2025, Номер unknown, С. 1160 - 1173

Опубликована: Апрель 5, 2025

The extensive utilization of information and communication technologies nowadays enhances accessibility underscores the importance data security. Image encryption is a prevalent technique for safeguarding medical on public networks, serving vital function in healthcare sector. Due to their intricate dynamics, memristors are frequently employed creation innovative chaotic systems that enhance efficacy chaos-based techniques. In recent years, methods have surfaced as viable method confidentiality transmitted images. Memristor-based Fractional-order (MFOCS) garnered considerable interest because resilience intricacy. (FOCS) exhibit more dynamics than integer-order systems. Consequently, exploration fractional development picture cryptosystems has gained popularity recently. This research introduces an image framework utilizing memristor-based map conjunction with Secretary Bird Optimization Algorithm (SBOA) improve security against cryptographic threats. suggested utilizes distinctive memory properties high-dimensional system produce unpredictable keys. SBOA utilized essential parameters, guaranteeing superior randomness statistical differential assaults. comprises confusion phase, which pixel positions randomized using sequences, succeeded by diffusion where intensities altered optimal key sequences. Performance evaluation executed entropy analysis, correlation coefficient tests, NPCR, UACI, studies computational complexity. findings indicate attains elevated entropy, minimal correlation, robust sensitivity, rendering it exceptionally resilient brute-force Notwithstanding its computing burden from fractional-order model offers secure efficient appropriate real-time protection applications.

Язык: Английский