Palmprint De-Identification Using Diffusion Model for High-Quality and Diverse Synthesis

2025-04-11

AI Summary

To mitigate identity privacy leakage risks arising from the public release of palmprint images, this paper proposes a training-free diffusion-based de-identification method. The method leverages semantic-guided embedding fusion and prior interpolation to effectively suppress identity-discriminative features while preserving non-sensitive attributes—such as texture patterns and geometric contours—with high fidelity. It introduces the novel, interpretable metric “de-identification ratio” for quantitative evaluation. The framework enables controllable, stable, and diverse synthesis of anonymized palmprints. Extensive experiments on multiple benchmark datasets demonstrate an average recognition rate reduction exceeding 92%, alongside high visual quality and strong inter-sample diversity in generated images. Moreover, the synthesized palmprints remain compatible with mainstream recognition systems, achieving a favorable trade-off between rigorous privacy protection and practical usability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Privacy
• Natural Language Processing: Generation

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Palmprint recognition techniques have advanced significantly in recent years, enabling reliable recognition even when palmprints are captured in uncontrolled or challenging environments. However, this strength also introduces new risks, as publicly available palmprint images can be misused by adversaries for malicious activities. Despite this growing concern, research on methods to obscure or anonymize palmprints remains largely unexplored. Thus, it is essential to develop a palmprint de-identification technique capable of removing identity-revealing features while retaining the image's utility and preserving non-sensitive information. In this paper, we propose a training-free framework that utilizes pre-trained diffusion models to generate diverse, high-quality palmprint images that conceal identity features for de-identification purposes. To ensure greater stability and controllability in the synthesis process, we incorporate a semantic-guided embedding fusion alongside a prior interpolation mechanism. We further propose the de-identification ratio, a novel metric for intuitive de-identification assessment. Extensive experiments across multiple palmprint datasets and recognition methods demonstrate that our method effectively conceals identity-related traits with significant diversity across de-identified samples. The de-identified samples preserve high visual fidelity and maintain excellent usability, achieving a balance between de-identification and retaining non-identity information.

Problem

• Develop palmprint de-identification to hide identity features.
• Generate diverse high-quality palmprints using diffusion models.
• Balance de-identification with non-sensitive information retention.

Innovation

• Uses pre-trained diffusion models for de-identification
• Incorporates semantic-guided embedding fusion
• Introduces de-identification ratio metric