Automated production of batched unclonable micro-patterns anti-counterfeiting labels with strong robustness and rapid recognition speed

TL;DR

This paper presents a rapid, robust, and high-capacity fluorescent anti-counterfeiting label technology using laser etching and machine learning, enabling fast recognition and high accuracy for secure product authentication.

Contribution

The work introduces a novel fluorescent anti-counterfeiting label with high throughput, robustness, and rapid recognition, significantly advancing pixel-level anti-counterfeiting methods.

Findings

Achieved etching speed of 1,200 labels in 3 seconds.

Recognized over 51,966 labels with 98.7% accuracy.

Recognition time per label is under 42 milliseconds.

Abstract

Anti-counterfeiting technologies are indeed crucial for information security and protecting product authenticity. Traditional anti-counterfeiting methods have their limitations due to their clonable nature. Exploring new technologies, particularly those based on pixel-level textures is a promising avenue to address the clonable issue due to high encoding capacity. However, research in this field is still in its infancy. This work introduces a new fluorescent anti-counterfeiting label technology with four key characteristics: efficient laser etching, high-throughput fabrication and segmentation, robustness aided by data augmentation, and an exceptionally high recognition speed. To be specific, the etching achieves a speed of 1,200 labels/3s, the high throughput yields a rate of 2,400 labels/4 min, and a total count of 51,966 labels. The number of labels is further augmented to 5,196,600 by implementing arbitrary rotation and brightness variation to enhance the robustness in the recognition procedure. We divide these labels into 44 categories based on differences in patterns. Utilizing machine learning methods, we have achieved a total recognition (including extraction and search process) time per label averaging 421.96 ms without classification, and 40.13 ms with classification. Specifically, the search process with classification is nearly fiftieth times shorter than the non-classification method, reaching 8.52 milliseconds in average. The overall recognition time is much faster than previous works, and achieve an accuracy over 98.7%. This work significantly increases the practicality of pixel-level anti-counterfeiting labels.