How I helped unmask two Russian spies
Hassan Ugail, Professor of Visual Computing at the University of Bradford, describes his central role in identifying the two men at the heart of the Salisbury novichok poisoning case.
I HAD complete confidence in our age progression and facial recognition software, so when the opportunity arose to contribute to one of the biggest international stories of the year, I was ready.
Two pictures, taken many years apart, were published by the Bellingcat website and Russian-based site The Insider identifying Ruslan Boshirov, one of the suspects in the Salisbury novichok case, was in fact Colonel Anatoliy Chepiga, a known agent of the Russian secret service, the GRU.
I was asked to compare both images. I undertook computer-based face recognition to see the probability of a match between the two images. To compensate for the significant differences in age, I ran an automatic ageing algorithm to bring the two individuals in the picture to the point where they were directly comparable.
I then ran the same face recognition algorithms to compare between the two aged photos. The result was that the three algorithms found the percentage probability of a match ranged from 97.7 percent up to 98.4 percent. This meant that I could say with confidence that the images of Anatoliy Chepiga and that of Ruslan Boshirov were of the same man.
Computer-based face recognition systems use facial images captured from a camera to analyse and store in digital form features such as the shape of the eyes, nose, cheeks and the mouth. The computer is capable of taking a very close look at a given facial image to analyse in great detail the various facial features as well as facial colour and the texture. This gives us a unique digital face print or the face template of an individual, and any new images of the individual captured at different locations and different times can then be compared with this stored template.
Having confirmed the real identity of Boshirov as Colonel Chepiga, Bellingcat asked me to run the same analysis on the man known as Alexander Petrov, who had visited Salisbury with Chepiga, but was suspected to be Alexander Mishkin, also a member of the GRU. I ran the same algorithms over an earlier passport picture of Mishkin and a cover passport in the name of Petrov. Again, I could say with confidence that they were the same man – Mishkin.
Bellingcat, using my analysis, delivered the findings in a final report on the case and at a briefing at the Houses of Parliament, widely reported in the media. The world now knows the real identity of the two men at the heart of the Salisbury poisoning case.