Photo illustration of how the technology works. Source: Wenyao Xu, University at BuffaloAccording to a new study led by researchers from the University at Buffalo, 3D printers are like fingerprints, in that no two are exactly alike.

In what is thought to be the first accurate technique for tracing a 3D-printed object to the machine it was printed from, researchers are calling the method “PrinTracker.” The team believes that the technique could assist law enforcement and intelligence agencies in tracking the origin of 3D-printed goods such as guns and counterfeit products.

"3-D printing has many wonderful uses, but it's also a counterfeiter's dream. Even more concerning, it has the potential to make firearms more readily available to people who are not allowed to possess them," said the study's lead author Wenyao Xu, Ph.D., associate professor of computer science and engineering in UB's school of engineering and applied sciences.

Understanding the new technique requires knowledge of the 3D-printing process. 3D printers move back and forth like a standard inkjet printer as it prints an object. A filament-like plastic is discharged from the nozzle instead of ink, usually in layers until the 3D-printed object is formed.

In each of those layers are tiny wrinkles called in-fill patterns which are intended to be uniform. Yet a combination of the printer’s model type, nozzle size and other factors is responsible for subtle imperfections. These patterns are both unique and repeatable - like fingerprints are to a person - and can trace the product back to the 3D printer that created it.

"3-D printers are built to be the same. But there are slight variations in their hardware created during the manufacturing process that lead to unique, inevitable and unchangeable patterns in every object they print," Xu said.

To test PrinTracker’s accuracy, researchers created five different door keys using 14 common 3D printers — 10 of which were filament deposition modeling (FDM) printers and four of which were stereolithography (SLA) printers.

Using an inkjet scanner, the team made digital images of each key, enhancing and filtering each image so that they could identify elements of the in-fill pattern. The team then created an algorithm capable of aligning and calculating the variations of the different keys, verifying the authenticity of each key’s fingerprint.

The team had created a fingerprint database of the 14 3D printers that they used for matching the keys to their respective printers. The technique worked 99.8% of the time. Ten months later, the test was repeated with the same outcome.

The team also tested for keys that had been intentionally damaged to obscure their identities. In those instances, PrinTracker was 92% accurate.

"We've demonstrated that PrinTracker is an effective, robust and reliable way that law enforcement agencies, as well as businesses concerned about intellectual property, can trace the origin of 3-D-printed goods," Xu said.

The study is being presented in Toronto at the Association for Computing Machinery's Conference on Computer and Communications Security, which runs from Oct. 15 to 19.