A new technique to identify people by looking at parts of their faces and without seeing the full facial features of the targeted person is under testing.
Using artificial intelligence techniques, a team of researchers at the University of Bradford has achieved 100 percent recognition rates for both three-quarter and half faces, the German News Agency reported.
The Science Daily website cited Researcher Hassan Ugail from the University of Bradford, saying: "The ability humans have to recognize faces is amazing, but research has shown it starts to falter when we can only see parts of a face. Computers can already perform better than humans in recognizing one face from a large number, so we wanted to see if they would be better at partial facial recognition as well."
The team used a machine learning technique known as a 'convolutional neural network', and fed the system with 2800 photos of 200 students and staff from FEI University in Brazil, with equal numbers of men and women.
The computer recognized full faces 100 percent of the time, but the team also had 100% success with three-quarter faces.
However, the bottom half of the face was only correctly recognized 60 percent of the time.
The team, then, trained the model using partial facial images as well. This time, the scores significantly improved achieving around 90% correct identification.
"We've now shown that it's possible to have very accurate facial recognition from images that only show part of a face, which opens up greater possibilities for the use of the technology for security or crime prevention," Ugail said.