Plasmonic Anticounterfeit Tags with High Encoding Capacity Rapidly Authenticated with Deep Machine Learning.

Area of Science:

• Materials Science
• Nanotechnology
• Optics

Background:

• Counterfeit goods pose significant economic threats and lead to product failures across industries.
• Developing robust and covert anti-counterfeit measures is crucial for product authenticity and consumer safety.

Purpose of the Study:

• To report a novel covert anti-counterfeit platform utilizing plasmonic nanoparticles (NPs) to generate physically unclonable functions (PUFs).
• To engineer high encoding capacity into these PUFs through controlled NP characteristics and leverage advanced machine learning for authentication.

Main Methods:

• Fabrication of physically unclonable functions (PUFs) by random deposition of anisotropic gold nanoparticles (Au NPs) of varying sizes onto surfaces.
• Analysis of PUF tags using optical microscopy, exploiting color responses from NP size and polarization sensitivity from NP anisotropy.
• Development of a deep machine learning model for rapid and accurate authentication of PUF tags and matching them with product lot numbers via metadata.

Main Results:

• Achieved a high estimated encoding capacity of 270, one of the highest reported for such systems.
• Demonstrated high accuracy and speed in authenticating tags and matching them to specific products and lot numbers.
• Successfully integrated designer plasmonic NPs with deep machine learning for a robust anti-counterfeit solution.

Conclusions:

• The developed platform offers a covert and highly secure method for anti-counterfeiting.
• The combination of plasmonic nanoparticles and deep machine learning provides a scalable and effective solution for product authentication.
• This technology has the potential to significantly mitigate economic losses and product failures caused by counterfeiting.