While it may not seem like much, an individual’s hand can be a key distinguishing factor, one that researchers have now utilized to their benefit. Because of variations in hand shape and positioning, anatomists note that your hand gives away more than you may think. Still, it’s certainly not a common practice…yet.
“Identifying a person using a small part of the hand is a challenging task, and has, to the best of our knowledge, never been investigated,” Hassanat and his team told the MIT Technology Review.
So in conducting their own investigation, the Jordan-based group took 500 photos of 50 men and women making a V sign with their right hand. All photos were taken against a black background using naught but an eight-megapixel camera phone, as most footage law enforcement officials receive of terrorist activity is filmed on low-resolution devices.
Then, the team began taking key measurements — the end points of the second and third finger, the lowest point between them, and two points in the palm. The Review also notes that the researchers “used a second method to analyze the shape of the hand using a number of statistical measures,” ultimately giving them a total of 16 features to identify different hands.
At this point, machine learning came into play, as they fed an algorithm two-thirds of these images — the remaining third were used to test how successfully the algorithm had “learned.” Surprisingly enough, Hassanat claims that this yielded an accuracy of 90 percent. “There is a great potential for this approach to be used for the purpose of identifying terrorists, if the victory sign were the only identifying evidence,” his team said.
While it may not be ready to be implemented in the field quite yet, this new identification technique is not only a major achievement for technology, but a serious breakthrough in the global war on terror.
Editors' Recommendations
- The best Netflix original movies
- The best Amazon original series available now
- When diagnosis time means life or death, NVIDIA’s advanced AI can save lives
- Machine learning can identify the world’s most overpaid and underpaid soccer stars
- Apple opens digital journal to showcase its machine learning developments
