The software, Silent-Sense , developed by Cheng Bo and his colleagues at the Illinois Institute of Technology, has demonstrated 99% accuracy in tests.
It uses the phone’s built-in sensors to record the unique patterns of pressure, duration , fingertip size and position of each user when interacting with their phone or tablet, Bo said.
Machine learning algorithms then turn this information into a signature that identifies the user — and will lock out anyone whose usage patterns do not match, New Scientist reported.
The system’s accuracy can be further enhanced by enabling the smartphone’s accelerometer and gyroscope to measure how much the screen moves when you are jabbing at it.
They can also pick up on your unique gait as you walk while using the screen.
In tests, a group of 100 users were told to use the smartphone’s touchscreen as they would normally.
SilentSense was able to identify the smartphone’s owner with 99% accuracy after no more than 10 taps. Even with an average of 2.3 touches, the software was able to verify the user 98% cent of the time.
The software stops checking the smartphone user’s identity when apps and mobile games are being used. However, to maintain security, it automatically switches on whenever more sensitive applications, such as email or SMS, are accessed, the reaserchers said.