A group of researchers was able to design a footstep-recognition system that can tell who you are based on your walking style with 100% accuracy.
The system could be the next big thing when it comes to security checkpoints, after fingerprint readers and retinal scanners. It is worth noting that each person’s walking style is unique just like their iris and fingerprints.
- Researchers claim that their newly developed system is 380 times more accurate than previous attempts to identify people by their gait.
- Also, unlike previous systems, the latest technology dubbed SfootBD doesn’t need people to walk barefoot to identify them.
In addition, SfootBD is much less invasive than other security measures, like iris scanners, since people are not even aware that they are being monitored. This is why privacy advocates are concerned that bad actors including governments could use the tech to covertly monitor the population.
Footstep-Recognition System is 100% Accurate
Computer scientist Omar Costilla Reyes, who led the research team that developed SfootBD, explained that a walking pattern has two dozen movements and different factors that differentiate it apart from other walking patterns.
Reyes’ team sifted through a database containing the walking patterns from 120 people. The database is the largest of its kind as it contains 20,000 of signals. Participants’ gait was mapped using pressure pads and sensors.
An artificially intelligent computer was fed with the data to measure gait speed, weight, and overall walking style. The system was able to identify a person based on their gait, not the footprint, with 100% accuracy.
Focusing on non-intrusive gait recognition by monitoring the force exerted on the floor during a footstep is very challenging,
said the researchers.
A study detailing the findings was published in the journal IEEE Transactions on Pattern Analysis and Machine Intelligence.
Image Source: Pixnio
Leave a Reply
You must be logged in to post a comment.