General Discussion
Related: Editorials & Other Articles, Issue Forums, Alliance Forums, Region ForumsJefferson Parish Sheriffs used facial recognition technology to arrest a man. The tech was wrong.
https://www.nola.com/news/crime_police/jpso-used-facial-recognition-to-arrest-a-man-it-was-wrong/article_0818361a-8886-11ed-8119-93b98ecccc8d.htmlBut a recent attempt by the Jefferson Parish Sheriffs Office to nab a high-end purse thief via facial recognition ended badly for a Georgia man who was jailed for almost a week over a false match, his lawyer says.
A detective took the algorithm at face value to secure a warrant to arrest Randal Reid, 28, in the June theft of luxury purses from a Metairie consignment shop, attorney Tommy Calogero said.
A Baton Rouge Police Department detective then adopted JPSOs identification of Reid to secure an arrest warrant alleging he was among three men involved in another luxury purse theft the same week at a shop on Jefferson Highway, court records show.
bucolic_frolic
(43,196 posts)Local populations, and thus genealogy, can be quite stable over hundreds of years. There are local look-alikes even without facial recognition technology.
getagrip_already
(14,764 posts)But nobody will be surprised by that.
The vendor who supplies fr software to the state does not make any claims that a target "match" is indeed the source photo.
In fact, they don't even use that term. What the sw provides is a score indicating how many similarities exist between the analyzed images. The higher the score, the more mathematical points match.
But is then the responsibility of a human analysts to use tools to further compare and analyze similar images. It is as much art as science.
Some things that can make matches harder are lighting, skin tone, head tilt and skew, and many other conditions.
For a Leo to see a high match and then move directly to arrest is as abusive as doctoring mug shot galleries to influence a witness.