Facial Recognition Is Getting Better
May 25, 2013

Analyzing Facial Recognition Technology Using The Boston Bombing Suspects

redOrbit Staff & Wire Reports - Your Universe Online

A test of three different facial recognition systems using one of the Boston Marathon bombing suspects as a test subject has helped demonstrate the value of the technology as a law enforcement tool, Michigan State University (MSU) researchers claim in a new study.

Working in the East Lansing, Michigan university´s Pattern Recognition and Image Processing laboratory, MSU computer science and engineering professor Anil Jain and research scientist Josh Klontz used actual law-enforcement video from the bombing and analyzed them using each of the three different types of facial recognition technology.

They found that one of the three systems was able to provide a “rank one” identification (essentially, a direct match) of suspect Dzokhar Tsarnaev. However, the other suspect — Tsarnaev´s older brother Tamerlan (who was ultimately killed in a shootout with police) — “could not be matched at a sufficiently high rank, partly because he was wearing sunglasses,” Jain explained in a statement Friday.

“This study was revealing in that facial recognition technology can successfully handle some cases in which facial images extracted from a video were captured under favorable conditions,” he added.

Under controlled conditions, when a suspect´s face is angled towards a camera and the lighting is good enough, this technology can be upwards of 99 percent accurate, the researchers explained.

According to the researchers, automatic face recognition searches a large database of facial images and can often quickly match a face to a name by locating the closest match.

This type of technology is often used by law enforcement agencies to create mug shot databases, but Jain said that it is currently unknown what type of facial recognition technology was used during the Boston bombing investigation.

“Some algorithms are better suited than others for face recognition in uncontrolled video,” the university explained. “While the technology has made great strides in recent years, it doesn´t mean that improvements aren´t needed. Also, more police agencies have to put the technology to use.”

“If you use an automatic system, it speeds up the process,” Jain added. “Sometimes police get bad tips so innocent people are questioned. Such situations can be avoided with a robust and accurate face-recognition system.”

Jain, Klontz and their colleagues, who have made their research available online, have been internationally recognized in the field of identification technology, according to the university. They have reportedly devised methods to match forensic facial sketches with mug shots, as well as a tattoo-matching system used by police to help identify criminal suspects.