Join or Sign in

Register for your free membership or if you are already a member,
sign in using your preferred method below.

To check your latest product inquiries, manage newsletter preference, update personal / company profile, or download member-exclusive reports, log in to your account now!
Login asmag.comMember Registration

How facial recognition can be used in law enforcement agencies?

How facial recognition can be used in law enforcement agencies?
There is a loud debate going on regarding the use of face recognition technology by law enforcement agencies.
There is a loud debate going on regarding the use of face recognition technology by law enforcement agencies. Just recently the British civil liberties organization Big Brother Watch lodged a claim in the British High Court against the Metropolitan Police’s use of face recognition cameras. The group called the police’s use of such technology “Orwellian” — a reference to George Orwell’s 1984, in which society is subjected to omnipresent government surveillance.
Still, despite public concern the law enforcement segment is expected to take up the largest portion of the face recognition market in terms of use cases through 2022, according to a report by MarketsandMarkets. Globally, the face recognition market is expected to reach nearly US$7.8 billion by 2022, up from $4 billion in 2017.
Roger Rodriguez,
Director of Client Relations,
Vigilant Solutions
So, how can law enforcement agencies deploy face recognition in a way deemed acceptable to the public? 

Best practices for face recognition by law enforcement

Roger Rodriguez, Retired NYPD Detective and Director of Client Relations at Vigilant Solutions shared a five-step investigative workflow the company recommends law enforcement agencies use when deploying face recognition technology. The workflow includes the following five elements:
  1. Identify the image. A face examiner needs to ask the following when vetting images for quality: Does the image meet the criteria for facial recognition? Does the image need enhancements? Does the image get rejected? Then apply data filters to the search to narrow the returned list to levels of specificity for higher accuracy rates.
  2. Run the search.
  3. Facial identification. Lower quality probe images return matches deeper in the candidate list. Expand the list to return 250 to 500 candidates. When additional profile images become available, utilize them for a comparative analysis. Definitive markings found by the analyst make possible match candidates stronger choices during the subjective facial analysis.
  4. Verify your choice. There are two levels of verification:
    • First-level verification. If physical characteristics have been met and a possible match candidate is selected from the gallery, conduct an immediate background investigation. Check incarceration status. Check addresses (i.e., residence vs. proximity to the crime) and check modus operandi and prior arrest history.
    • Second-level verification. Present your case to three to five people. Show that all physical similarities and background check validations are complete. Pitch the match candidate by discussing physical similarities and background validations.
  5. The possible match (a.k.a. “the investigative lead”). Facial recognition search results are investigative leads only. Agencies should not make an arrest based on a possible match report. They should use other standard law enforcement procedures to verify a person’s identity.

The road ahead

While there are benefits to having law enforcement agencies use face recognition technology, people wonder if the benefits truly outweigh the costs. With privacy concerns growing, the use of face recognition by law enforcement is also drumming up concerns regarding racial profiling and the infringing of civil rights. It will be a tough sell for those already fed up with mass surveillance; however, society also expects full measures to be taken to ensure its safety. Regardless, one thing is for certain: the need for more education and more accountability by those using face recognition technology.

Product Adopted:

Share to: