The researchers have developed an AI -based system which can identify “the specific machine” after an inspection of the 3D printed parts it produced. A Grainger College of Engineering The team at the University of Illinois Urbana-Champaign said that the system could be used to manage supply chains, “detect the first problems and check that suppliers follow the processes contained”.
However, talking about “hidden fingerprints” of 3D printers more interested in the implications in criminal investigation of the new AI.
Implications for manufacturing, quality and confidence
Only a photograph of a part printed by a certain 3D printer is necessary to check its “unique signature … or fingerprint”, says the Grainger blog. “We are always surprised that it works: we can print the same part design on two identical machines – the same model, the same process parameters, the same material – and each machine leaves a unique fingerprint as the AI model can find up to the machine,” said Bill King, professor of science and mechanical engineering, and the research project manager. “It is possible to determine exactly where and how something was done. You don’t have to speak of your supplier about anything. ”
Researchers have closely examined the (major) implications for supplier management and quality control. They described how it was important for 3D printers suppliers to guarantee that users “adhere to a specific set of machines, process and factory procedures” to offer expected quality levels throughout the printing / production process.
In the tests, photographs of smartphones of 9,192 pieces made on 21 3D printing machines of six companies, and with four different manufacturing processes, could be fingerprints “with a precision of 98% from only 1 square millimeter of the surface of the room”. It would be interesting to see how precise the AI could be, given a larger area in Anaylze.
Beware of using your 3D printer for “illicit products”
Despite what looks like a fairly narrow accent on the 3D printers used in factory production environments, AI can operate its magic on a much wider pool of devices.
“These fingerprints of manufacturing have hidden in sight,” said King. “There are thousands of 3D printers worldwide and tens of millions of 3D printed parts used in planes, cars, medical devices, consumer products and a multitude of other applications. Each of these parts has a single signature which can be detected using the AI. ” We assume that King’s assertion includes all the printers listed in the update regularly Tom material Best article of 3D printers.
Beyond its used use for the management of the supply chain, researchers say that technology could be applied to “follow the origins of illicit goods”. Apart from that, they do not discuss any involvement for criminal surveys. However, we would like to see the technology used to draw the origin of a 3D printed pistol discovered on a crime scene, for example.
