Researchers have developed a way to identify footsteps with biometric scanning and artificial intelligence.
There are a wide variety of biometric tools available, including fingerprint scans, iris scans and facial recognition. At Springwise, we have seen innovations in biometric scanning that include using biometric data to reveal customers’ ideal holiday destinations and facial recognition used to identify endangered primates. Now, footstep-based biometrics can be added to the list. Researchers at the Indian Institute of Technology in Delhi have developed a biometric system based on footsteps.
The new system uses a geophone. This is a transducer that converts ground movement into electrical signals then analysed by machine learning algorithms. The researchers first had volunteers walk barefoot in a circle around a geophone. They used the data to train the algorithm to differentiate between the steps of different participants by evaluating the time between steps, their frequency, length and their cadence. The researchers found that it was possible to identify a person with 92 percent accuracy using just seven consecutive footsteps.
The researchers believe the system could eventually be used in high-security areas to identify people who are not already logged into an approved footstep database. It could also be used in smart homes. For example, to detect intruders or turn on preset lighting or music when individual family members walk into a room. However, there is still a way to go before this can happen. At the moment, the system can only identify one person at a time, so it can’t be used in crowds.