Related Experiment Video
Updated: Jun 5, 2025
12:45
Detection and Quantification of Tunneling Nanotubes Using 3D Volume View Images
Published on: August 31, 2022
2.7K
High-dimensional anticounterfeiting nanodiamonds authenticated with deep metric learning
Lingzhi Wang1, Xin Yu1, Tongtong Zhang1
1Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, China.
Nature Communications
|December 5, 2024
Summary
This study introduces novel 3D-encoded fluorescent nanodiamond labels for unbreakable anticounterfeiting. A deep metric learning method enhances authentication, improving noise tolerance and applicability to new labels.
Area of Science:
- Materials Science
- Nanotechnology
- Optics
Background:
- Physical unclonable function (PUF) labels offer robust anticounterfeiting solutions.
- Current PUF systems face challenges with limited high-dimensional encoding and inadequate authentication methods.
Purpose of the Study:
- To develop advanced 3D-encoded diamond-based labels for anticounterfeiting.
- To create a robust authentication method with improved noise tolerance and adaptability.
Main Methods:
- Utilized linear polarization modulation of fluorescent nanodiamonds for 3D encoding.
- Implemented a deep metric learning algorithm for image-based authentication.
- Assessed label stability, reproducibility, and authentication performance.
Main Results:
- Demonstrated a 3D encoding scheme for diamond-based labels with a capacity of 10^9771.
- Achieved high distinguishability within a 7.5s readout time.
- The deep metric learning authentication showed superior noise tolerance and applicability to unseen labels compared to traditional methods.
Conclusions:
- Fluorescent nanodiamonds provide stable and reproducible labels for anticounterfeiting.
- The developed deep metric learning authentication is effective for PUF systems.
- This integrated approach lays the foundation for practical, unbreakable anticounterfeiting solutions.
