Researchers from University of York in the UK examined the ability of both human viewers and smartphone face recognition software to identify a face morph as distinct from the two faces contributing to the morph.
Both humans and smartphones are unable to accurately differentiate between 'real' faces and photos that are morphed on fraudulent identity cards, say scientists.
Researchers from University of York in the UK examined the ability of both human viewers and smartphone face recognition software to identify a face morph as distinct from the two faces contributing to the morph.
They took two 'real' face photos and digitally blending them to make a new, but similar, face that both contributing faces can use as false ID. Human participants and smartphone software were asked to decide if a pair of faces matched. Sometimes, one of the pair was a morph photo and the other was one of the contributing faces.
Results showed that both humans and smartphone software are frequently unable to distinguish face morph photos from the two faces contributing to the morph.
Initially, human viewers were unable to distinguish a 50/50 morph photo from its contributing photos 68 per cent of the time.
However, after simply briefing the viewers to look out for manipulated, 'fraudulent' images, the error rate dropped greatly to 21 per cent. Researchers also looked at smartphone software, which achieved similar results to briefed human viewers, with an error rate of 27 per cent.
These rates, however, are still significantly higher than error rates when comparing two photos of entirely different people, researchers said. Although, the participants in this study are unlikely to be as motivated or as skilled as a professional at spotting fraudulent photos, this study indicates that humans and smartphones may not naturally identify face morphs, a weakness that could be exploited by fraudsters, researchers said.
"It is encouraging, however, that armed with the knowledge of morphed photo IDs, the risk of fraudulent activity being missed is significantly reduced," said Mike Burton from the University of York.
"Raising awareness of this type of fraud and including it in training schemes for frontline staff can help overcome these issues, and with new technologies coming on line, it should be a challenge that can be tackled with some success," Burton added. The study was published in the journal PLOS ONE.
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
 "University of York, UK, smartphones, morph photos, PLOS ONE, Web Exclusive")
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
)
){kind=small}
){kind=small}
)
)
)
)
)
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}