Researchers have developed a new system where they made use of artificial intelligence to identify a person through their footsteps.

With the help of a lot of data on footsteps, researchers at Indian Institute of Technology developed a new system where they used AI to identify a person just by analyzing their footsteps.

In order to gather the data, a device known as geophone was used that converts ground movements into electrical signals. Eight volunteers were asked to each walk barefoot in a circle with a geophone placed in the center, coming as close to the device as 3.2ft and as far as 8.2ft from it, reported Futurism.

Chinese school uses facial recognition to observe inattentive students

About an hour of walking data was collected for each individual. They then added up to 46,489 footsteps, a record they believe is the largest footstep database ever gathered. The data was then used to train an algorithm to differentiate between the steps of individual volunteers by evaluating the time, length, and rhythm between the steps.

After the experiment, the AI was able to detect a person with 92% accuracy after recording only seven consecutive footsteps. The researchers claim that their system can be used in high-security areas such as military bases, classrooms, and future smart homes.

However, there are still more improvements to be done. For instance, currently the system is unable to identify more than one person at a time, hence useless in crowded places. For the future, the researchers are hopeful that it can ultimately replace other biometric identification systems like the fingerprint or retina scanners.

“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 individual’s privacy; [and it’s] less sensitive to environmental parameters and beyond the capacity of an individual to decode and manufacture the raw signal,” researchers wrote.