Have you ever said of a person, "You can see it in the way they walk"? Well, if it was their identity that you were referring to, then you were right. To that end, scientists have now created an artificial intelligence-based system that identifies people via their footsteps.

Created by researchers from the University of Manchester and the University of Madrid, the new biometric verification system simply requires people to walk normally across a pressure pad that's placed on the floor.

As they do so, their gait pattern (including the amount of pressure they exert with each step) is recorded – it doesn't matter whether or not they're wearing shoes. This recording is compared to a gait pattern that's already on file for the individual in question, to see if the person walking across the mat is really who they claim they to be.

"Each human has approximately 24 different factors and movements when walking, resulting in every individual person having a unique, singular walking pattern," says Manchester's Dr. Omar Costilla Reyes, who led the research. "Therefore monitoring these movements can be used, like a fingerprint or retinal scan, to recognize and clearly identify or verify an individual."

To train the AI used by the system, the scientists created the largest footstep database in history. Known as SfootBD, it's made up of almost 20,000 footstep signals obtained from 127 people.

The resulting system has been tested in settings including airport security checkpoints, workplaces, and home environments. In all cases, it was able to non-intrusively verify individuals' identity with almost 100 percent accuracy, telling "genuine" people apart from imposters pretending to be them. That said, there are other possible applications for the technology.

"The research is also being developed to address the healthcare problem of markers for cognitive decline and onset of mental illness, by using raw footstep data from a wide-area floor sensor deployable in smart dwellings," says Reyes. "Human movement can be a novel biomarker of cognitive decline, which can be explored like never before with novel AI systems."

A paper on the research has been published in the journal IEEE Transactions on Pattern Analysis and Machine Intelligence.

Source: University of Manchester via AlphaGalileo