IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Five Things to Know About Police Facial Recognition Tech

Facial recognition technology has allowed police departments across the U.S. to compare the faces of criminal suspects against other existing photos, but the tech has also proven controversial.

Facial recognition applied to a crowd of people
Shutterstock/varuna
(TNS) — Facial recognition technology has allowed police departments across the U.S. to compare the faces of criminal suspects against social media photos, driver's licenses and mug shots. But the tech has proven controversial, and it isn't always accurate, as a recent false arrest by the Jefferson Parish Sheriff's Office shows.

Here are a few things to know about the technology:

Here's how it works.

A face shot is submitted to a facial recognition system, which measures distinctive characteristics such as the width of nose and depth of eye sockets. Those data points are run through an algorithm to create a "biometric template" to be stored for future comparison to an image of a criminal suspect. A gallery of likely candidates often results. "Face recognition search results are not considered positive identification and do not establish probable cause, without further investigation; rather, they are advisory in nature as an investigative lead only," according to a policy guide developed by the Bureau of Justice Assistance, under the U.S. Justice Department.

It's not new, but it's getting popular.

The FBI began replacing its fingerprint-based biometric identification system in 2010 to one that incorporates facial recognition. By 2019, the FBI had 21 state partners and searchable access to 641 million photos. The U.S. Government Accountability Office last year found that nearly half of 42 federal agencies that employ law enforcement officers reported owning or using the technology. Six federal agencies reported using it on images filmed during protests after George Floyd's killing by police in May 2020. Three agencies reported using it on images from the U.S. Capitol riots on January 6, 2021, the GAO found. In Louisiana, several requests last year for facial recognition searches through the Louisiana State Analytical and Fusion Exchange came from federal law enforcement agencies.

Regulation is sparse.

Some cities and states have rushed to regulate facial recognition as government use expands. Reports of its use by Chinese officials to track dissidents in public have elevated fears of a surveillance state. By last year, several states and about 20 cities, mostly in liberal bastions, had erected safeguards or bans on government use of facial recognition. But a national surge in crime has led some to reverse course. After less than a year, Virginia lawmakers lifted a sweeping ban on facial recognition by local and campus police. California lawmakers let a 3-year moratorium expire on applying facial recognition to police body-worn camera footage. New Orleans officials this year lifted a two-year ban on the technology, while setting limits. Under new city rules, it can't be used as a surveillance tool or to monitor people at public or political gatherings.

Advocates see it as a powerful crime-fighter.

Backers of the technology liken it to a more reliable eyewitness, pointing to its utility in tracking down missing children as well as violent felons. Among the success stories was the 2017 arrest of Walter Yovany-Gomez, an MS-13 gang member and killer who had ascended to the FBI's Ten Most Wanted Fugitives list after evading authorities for years. Advocates point to safeguards in the federal Privacy Act as well as FBI policies limiting its use by agents to argue against outright bans.

"Thanks to facial recognition, tasks that would take countless hours — like combing through large databases of photos, biometric data, or other personally identifiable information already in the government's lawful possession — can be accomplished in a fraction of that time, at much lower cost," said then-Associate Deputy Attorney General Sujit Raman in 2020. "It goes without saying how such advancements can assist law enforcement."

Police agencies don't like to talk about it.

The New Orleans Police Department only confirmed its use of the technology in late 2020. The Jefferson Parish Sheriff's Office did not respond to several inquiries about its use of the technology and shrugged off a request for public records about it. In most places, there is no law governing police use of it, said ACLU of Louisiana advocacy director Chris Kaiser. "Any law enforcement agency can and already has rolled out this technology in our communities, without any notice or external guidance or guardrails," he said.

© 2023 The Advocate, Baton Rouge, La. Distributed by Tribune Content Agency, LLC.