Researchers have found a new way to quickly break into smartphones which employ the pattern unlock method. By using a computer vision algorithm to trace a person’s finger on a phone display, the software can correctly identify a pattern usually within five attempts.
Pattern unlocks consist of a grid of dots on a device’s lock screen which users connect in a unique way to gain access to the phone. It’s popular unlocking method — it’s estimated that 40 percent of Android owners use it as opposed to a PIN or a password — but it’s also one of the less secure.
The study was carried out in a collaboration between Lancaster University, the University of Bath and Northwest University in China, who filmed smartphone users while they drew the unlock patterns onto their smartphone screens.
The researchers tested 120 unique patterns from 215 users and discovered that the tracking software was able to correctly identify the pattern within five attempts in 95 percent of the cases. The process is said to work even when the camera can’t see the device display and from up to nine metres away when recorded on a DSLR camera.
Ironically, it also seemed that the more complicated pattern, the easier it was to crack.
Most Android users are probably aware that patterns are less secure than PINs or text passwords; many Android devices make this fact known to the user when choose an unlock method. Further, the use of a camera and dedicated software may seem like a convoluted approach when you can often look at someone draw a pattern and simply memorize it.
Finding ways around Android lock security screen isn’t a new practice. Facial recognition locks can be tricked with a photograph, and fingerprint scanners can be bypassed using sticky tape. This research, in essence, is just more evidence that patterns are a lesser form of lock screen security.
If you want to help keep your device safe, you could try to ensure you unlock your phone away from prying eyes or change to a more secure unlock method like a PIN or password.
What are your thoughts on the research? Let us know in the comments.