When it comes to recognizing the identity of a person we mostly use fingerprint scans, iris scans, and perhaps even eye gaze scans, but what if I tell you there is another insane way to do this and it might be more accurate and safe.
Scientists have developed Artificial Intelligence (AI) that can identify people by measuring their gait or walking pattern. Yes, you read that right. This new AI can successfully verify an individual simply by analyzing the 3D footstep and time-based data.
The AI system is developed by researchers at the University of Manchester in the UK and the University of Madrid in Spain. The research paper titled “Person Identification using Seismic Signals generated from Footfalls” has published in IEEE Transactions on Pattern Analysis and Machine Intelligence.
To create an AI system that can accurately identify a person by their footsteps, computers need to learn about movements patterns. The team collected the largest footstep database in history, containing nearly 20,000 footstep signals from 127 different individuals. Over the period of a month, they used a geophone to collect roughly 46,000 footfall events from eight barefooted test participants.
“Each human has approximately 24 different factors and movements when walking, resulting in every individual person having a unique, singular walking pattern. Therefore monitoring these movements can be used, like a fingerprint or retinal scan, to recognize and clearly identify or verify an individual,” said Omar Costilla Reyes, from the University of Manchester.
To compile the samples and dataset the team used floor-only sensors (pressure pads) camera and high-resolution cameras. This dataset, called SfootBD, used to develop the advanced computational models needed for automatic footprint biometric verification.
SfootBD is nearly 380 times more accurate than previous methods, and it doesn’t require a person to go barefoot in order to work. It’s less invasive than other behavioral biometric verification systems, such as retinal scanners or fingerprinting, but its passive nature could make it a bigger privacy concern since it could be used covertly.
The AI uses a neural network that can find telltale patterns in a person’s gait that can be used to recognize and identify them with almost perfect accuracy. The machine learning models are capable of distinguishing between steps (and by extension, people), the researchers collected both the time and frequency of footfalls in addition to their length and cadence (the gap between two consecutive footsteps).
An artificially intelligent system called a deep residual neural network scoured through the data, analyzing weight distribution, gait speed, and three-dimensional measures of each walking style. Importantly, the system considers aspects of the gait, rather than the shape of the footprint.
“Focusing on non-intrusive gait recognition by monitoring the force exerted on the floor during a footstep is very challenging,” said Reyes. “That’s because distinguishing between the subtle variations from person to person is extremely difficult to define manually, that is why we had to come up with a novel AI system to solve this challenge from a new perspective.”
The system is basically divided of three layers: things (sensors paired with low-end processors, and embedded processors paired with transceivers); fog (embedded processors and transceivers); and cloud (a server).
The things layer, which in this implementation consists of a Raspberry Pi Zero, a geophone (a ground motion transducer that converts ground movement into voltage), and a long-range transceiver module, automatically extracts the portion of the seismic signal that represents a footfall and compresses it before sending it over ZigBee to the fog layer.
The fog layer — a Raspberry Pi 3 model B — receives the football signal, decompresses it, extracts important features from it, and classifies the signal before passing it onto the cloud over Ethernet or Wi-Fi. Lastly, the cloud performs inference.
In the process of model training, the team found that about 875 footsteps per class — about eight minutes of walking — were required to achieve accuracy greater than 85 percent, but their results beat that baseline in the end.
During testing, the best-performing AI system matched an individual with his or her footsteps 92.29 percent of the time from just seven consecutive footsteps.
“The main advantages of this type of biometric system are [that] seismic sensors can be easily camouflaged; evading detection is impossible because footstep patterns are inimitable; it does not breach an individual’s privacy; less sensitive to environmental parameters and beyond the capacity of an individual to decode and manufacture the raw signal,” they wrote.
This new system also has some limitations. As noted, SfootBD requires the use of floor pads and a high-res camera, so this form of surveillance and identification can’t be used just anywhere.
What’s more, the tool is only as powerful as its database; the only individuals who can be identified are those whose distinctive gaits have been previously recorded and cataloged in the system.
Another notable drawback of the system is its inability to ID more than one person at a time — two or more confuse the system. The researchers leave this to future work, but believe the current iteration could be reliably used to register classroom or workshop attendance, detect intruders, and control home appliances.
More in AI :