Graham Rayman
New York Daily News
NEW YORK — When they are stumped for quality images of possible crime suspects, the NYPD sometimes uses celebrity photos to find a match in its facial recognition system, a new report reveals.
Actor Woody Harrelson’s image was used to try to find a match to a man with similar features who stole beer from a pharmacy on April 28, 2017, according to a study by the Georgetown University Center of Privacy and Technology released Thursday.
The thief was caught on store surveillance video, but the images were too blurry to match. A detective decided the suspect looked like Harrelson, so they ran his photo through the database. In another instance, the NYPD used a picture of a New York Knicks player to identify a man wanted for assault in Brooklyn.
The shoplifting case ended with an arrest, but the authors of the report say the NYPD and other police departments too often blur the line of what is ethical when using the facial recognition software.
“There are no rules when it comes to what images police can submit to face recognition algorithms to generate investigative leads,” the report said. “(But) The stakes are too high in criminal investigations to rely on unreliable — or wrong — inputs. Unfortunately, police departments’ reliance on questionable probe photos appears all too common.”
The NYPD’s analysts routinely edit photos, sometimes adding in different facial features from other pictures. They “enhances” the images in a way that goes beyond lighting adjustments and color correction, the report said.
Among the things the department changes are facial expressions, opening and closing eyes and mouths - which can look like mug shots, the study said.
“These techniques amount to the fabrication of facial identity points: at best an attempt to create information that isn’t there in the first place and at worst introducing evidence that matches someone other than the person being searched for,” the report said.
While NYPD guidelines say that any matches are just unconfirmed possible matches, in practice the department has used face recognition matches to place people in lineups. Cops in Jacksonville, Florida and Washington, D.C. similarly used the software. But on Tuesday, San Francisco banned use of the technology.
The NYPD has made 2,878 busts using facial recognition in some manner, according to the report.
“We cannot sit idly while police recklessly expand the use of Orwellian surveillance technologies like face recognition to secretly track people in real-time as they go about their daily lives,” said Abdullah Hasan, a spokesman for the American Civil Liberties Union. “This is particularly true for police jurisdictions with troubling histories of discriminatory practices.”
In a lengthy statement, the NYPD defended the program, calling a match “merely a lead and not probable cause to arrest.”
“The NYPD has been deliberate and responsible in its use of facial recognition technology,” the department said. “We do not engage in mass or random collection of facial records from NYPD camera systems, the internet, or social media.”
Police credit the software for assisting in the arrest of a man for throwing urine at MTA conductors, the bust of a man who pushed a straphanger on to the subway tracks, and a successful missing persons case involving a woman with Alzheimer’s.
“The leads generated have also led to arrests for homicides, rapes and robberies,” the department said. “The NYPD constantly reassesses our existing procedures and in line with that are in the process of reviewing our existent facial recognition protocols.”
———
©2019 New York Daily News