Despite advances in biometric identification technologies, such as fingerprint, eye and facial recognition, the humble handwritten signature is still the most commonly used form of biometric used to verify someone's identity. Researchers have developed software that blends the old and the new by leveraging the motion detection capabilities of a smartwatch to verify a signature as it is written.
While dedicated hardware, such as digital pads or special electronic pens, has dragged the signature into the digital age, these systems can be cumbersome and expensive. New software, developed by researchers from Tel Aviv University (TAU) and Ben-Gurion University of the Negev, uses motion data captured by a smartwatch as the wearer scribbles their John Hancock to confirm the veracity of said signature.
The motion data captured during the signing process comes courtesy of the smartwatch's accelerometer or gyroscope, which produces what the researchers say is a unique identifier that allows the signature to be classified as genuine or a forgery.
"Using a wrist-worn device such as a smartwatch or a fitness tracker bears obvious advantages over other wearable devices, since it measures the gestures of the entire wrist rather than a single finger or an arm," says study co-author Dr. Erez Shmueli of TAU's Department of Industrial Engineering. "While several other recent studies have examined the option of using motion data to identify users, this is its first application to verify handwritten signatures – still a requirement at the bank, the post office, your human resources department, etc."
To test the accuracy of the system, the researchers enrolled the services of 66 TAU undergraduate students. With all of them wearing smartwatches, they each provided 15 signature samples on a tablet using a digital stylus. The students then watched video of people signing their signature and were tasked with forging five of those signatures after having time to practice and being rewarded for "exceptional forgeries."
Despite the researchers reporting the verification software achieved an "extremely high level" of accuracy in detecting forgeries, they admit there is a major downside to this approach – the fact that most people (66 percent according to a recent survey cited by the researchers) wear a watch on their non-dominant hand. This means that the majority of people would need to swap the smartwatch to their dominant hand when signing their signature, significantly affecting the user friendliness of the system.
However, the team has applied for a patent with plans to eventually commercialize the technology. But before then, they are looking to improve upon it.
"Next we plan to compare our approach with existing state-of-the-art methods for offline and online signature verification," said Dr. Shmueli. "We would also like to investigate the option of combining data extracted from the wearable device with data collected from a tablet device to achieve even higher verification accuracy."
The team's paper on the software can be found here.