Assessing Automated Facial Recognition Technology
This articles features the use cases of facial recognition technology in threat detection and augmented intelligence.
Join the DZone community and get the full member experience.Join For Free
Facial recognition technology has become an increasingly deployed tool in law enforcement, whether checking people as they navigate the airport or even in stop and search. A recent report from Cardiff University explores how effective these technologies have been.
The research evaluated the use of automated facial recognition (AFR) at numerous sport and entertainment events by the South Wales Police force over the course of a year. The analysis found that while the technology can help the police identify suspect individuals they might otherwise have missed, it requires a large investment and considerable changes in standard operating procedures to enable consistent results to be achieved.
The technology the police used works in two main ways. The first of these provides a live and real-time facial assessment based upon CCTV footage from an area. It compares the faces it detects with a database of facial images of known individuals.
The second method takes still images of people previously unidentified by the police and compares them against images from the custody database. This is done to create more investigative leads to follow up on. This was a pretty haphazard method, however, as the research found that 68 percent of the submissions generated weren’t of high enough quality for the system to work.
On a more positive note, however, the data showed a considerable improvement in the performance of the technology over time, both in terms of the capabilities of the technology, but also the ability of the police officers to use it effectively. Indeed, the data showed that 18 arrests were made using the real-time analyses, with over 100 arrests made after people had been identified by the system.
As with so many applications of AI, the team believes that the best way to think of the technology is not in terms of replacing human officers, but rather augmenting them and their work. The technology is, in essence, a decision support tool, with the ultimate decisions always being made by police operators.
“There is increasing public and political awareness of the pressures that the police are under to try and prevent and solve crime. Technologies such as Automated Facial Recognition are being proposed as having an important role to play in these efforts. What we have tried to do with this research is provide an evidence-based and balanced account of the benefits, costs, and challenges associated with integrating AFR into day-to-day policing,” the researchers say.
It’s a technology that has gained broad acceptance from the police, with leaders from South Wales Police revealing that they were happy with the results, and indeed from the learning that was achieved through the course of the work.
“South Wales Police remains committed to the continuous use of the technology in a proportionate and lawful way to protect the public, while also remaining open and transparent about how and when we use it,” they say.
Published at DZone with permission of Adi Gaskell, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.