A BRADFORD University professor, who used a face-ageing technique he designed to confirm the true identify of the two Russian Novichok suspects, has spoken of the first real world use for his innovative software.

Professor Hassan Ugail, of Bradford’s Centre for Visual Computing, was asked to run a face matching algorithm on both Salisbury nerve agent suspects Ruslan Boshirov and Alexander Petrov.

Using photographs taken years apart, he was able to confirm that the men identified as Boshirov and Petrov, were actually Anatoliy Chepiga and Alexander Mishkin.

The two men are accused of the attempted assassination of ex-Russian spy Sergei Skripal and were were originally named by the UK authorities as Boshirov and Petrov.

Investigatory website Bellingcat later claimed to have revealed the true identity of the pair. They identified Boshirov as Colonel Anatoliy Chepiga, and Petrov as Dr Alexander Yevgenyevich Mishkin, who both worked for Russia’s GRU intelligence agency.

Professor Ugail told the Telegraph & Argus that the results were conclusive. He described how, despite the photographs being taken years apart, he was able to bring both ages to that of 30 and then compare the two.

“Facial recognition is something that we have been working on for ten plus years, but this is the first time we have used ageing and recognition in a real world application.”

He added that he was quickly able to confirm that the first set of photographs were the same person with a probability in the high 90s, and the second set of photographs was in the low 90s, due to the fact that the older photograph was so grainy.

The men identified as Petrov and Boshirov are believed to have smeared the highly toxic Novichok nerve agent on a door handle at the Wiltshire home of Mr Skripal in March.

Both Mr Skripal and his daughter survived the attack.

Last summer, Professor Ugail unveiled the new face-ageing technique he and his team had been working on as something that would enhance the search for long-term missing people.

The method maps out the key features, such as the shape of the cheek, mouth and forehead, of a face at a certain age. This information is fed to a computer algorithm which then synthesises new features for the face to produce photographic quality images of the face at different ages.

The technique uses a method of predictive modelling and applies it to age progression as well as incorporating facial data from a large database of individuals at different ages thus teaching the machine how humans actually age.

To test their results they ran the algorithm backwards to de-age a photo of a person to a younger age and compared it with an actual photograph taken at the young age.