Facial Recognition Success Stories Showcase Positive Use Cases of the Technology

By Jake Parker, Senior Director of Government Relations, Security Industry Association on July 16, 2020 |

Updated Sept. 19, 2023

SIA’s Core Principles for Facial Recognition Technology Use

The Security Industry Association (SIA) believes facial recognition must be used only for beneficial purposes that are lawful, ethical and nondiscriminatory. Explore our principles for development and deployment of facial recognition tech across safety and security applications.

View the Principles

Advances in computing power, combined with rapid improvements in the quality of photo and video technology developed by the security industry over the last 15 years, have allowed facial recognition technology to develop and mature. Simply put, facial recognition software compares and matches facial images. This is done using an algorithm to compare a presented image with one or more “enrolled” images, after the images are converted to unique numerical versions readable by that specific system and compared for similarity.

Facial recognition has three primary functions, with differing outputs:

Verification: Verifying a person matches an enrolled image associated with their identity
- Output: An automated yes/no decision to authenticate
Identification: Determining that an image matches one or more images enrolled in a database
- Output: Either an automated yes/no decision of some kind OR flagging a likely match for further review
Investigation: Helping determine whether a matching image is in a database for an investigative purpose
- Output: A series of images from a database with the highest similarity scores are provided for review

These functions have proven to be incredibly useful across many types of applications. The purposes can be vastly different from one another, presenting application-specific implementation considerations – from verifying an identity in a remote online transaction to physically opening a door for an authorized person to helping law enforcement investigators narrow down a set of possible matching photos.

This technology is particularly important to the security field because it enhances capabilities of solutions like video security, access control and identity management systems that help our customers secure their facilities, employees and patrons against the threat of violence, theft or other harm.

But use of facial recognition technology has benefited Americans in many other underpublicized ways – helping law enforcement find missing children, fight human trafficking, find dangerous criminals and bring sexual predators to justice. At the same time, it allows individuals to quickly and conveniently prove their identity to enter a venue, board a plane, perform online transactions and seamlessly access personalized experiences.

Here are just a few of the successes.

Law Enforcement Investigations

Facial recognition software is used daily across the U.S. to assist in identifying and capturing the most violent criminals in our country and bringing justice and closure for victims. For over a decade, federal, state and local law enforcement agencies have used it to compare facial images in thousands of investigations. Many public safety officials feel that this technology has become a game-changer for keeping our communities safe, pointing to instances where crimes would have never been solved or prevented without it. Use under appropriate policies and procedures has been endorsed by the nation’s leading law enforcement professional associations.

In U.S. law enforcement, the technology serves as a tool to assist investigators in generating leads regarding the identity of an unknown person of interest in an image (such as a witness, victim, suspect or associate) where needed in a specific ongoing case. In this application, the technology does not confirm an identity or result in automated decisions. If a lead is developed, investigative techniques outside of facial comparison must used to find and confirm information needed to positively identify a person, and if a suspect, establish probable cause to make an arrest or obtain a search warrant.

Under traditional methods, police seek to identify a person of interest manually by looking through hundreds of mugshots with victims, canvassing areas with photos or searching a database using vague suspect descriptions or names that could easily be criminal aliases or fraudulent identities. Facial recognition technology automates and improves the first step of identifying potentially matching photos from a database. Beyond simply improving an otherwise manual process, facial recognition contributes to more accurate identification. The National Institute of Science and Technology has found that forensic examiners performed best when supported by facial recognition technology and the most accurate performance resulted when these efforts are combined.

Proper use of this technology is critical to protecting the innocent, as eyewitness identifications in criminal investigations are notoriously prone to error. According to the Innocence Project, mistaken eyewitness identifications have been the key factor in the vast majority of wrongful convictions in the U.S. later overturned.

The following are reported successes across the U.S., representing just some of many examples.

Exonerating the Innocent

A Florida man falsely accused of vehicular homicide was exonerated only after facial recognition technology made available to public defenders was used to help identify and locate a key witness to the scene of a fatal crash, who confirmed the man was a passenger and not the driver of the vehicle, who was killed in the incident.

A witness in a gang-related assault case in northern Virginia provided cell phone photos of the suspects to police detectives working the case. One of the photos of an unknown suspect was queried against regional booking and arrest photos and an investigative lead was developed. Upon further investigation and confirmation of the identity of the suspect, it was found that the individual was in jail in another jurisdiction at the time of the assault. Use of technology in this case helped quickly clear the individual and avoided unnecessary contact from law enforcement.

Fighting Human Trafficking

Local law enforcement investigators were working to identify a subject suspected of child sex trafficking in Fairfax County, Virginia. Using a photograph from social media of the person believed to be the suspect, a query against regional booking and arrest photos resulted in a lead, aiding in the progress of a critical child sex trafficking investigation.
In California, a law enforcement officer saw a social media post about a missing child from the National Center for Missing and Exploited Children. The officer used the Spotlight investigative tool to return a list of online sex ads featuring the girl. As reported in Wired, the girl who was rescued had been “sold for weeks,” and the officer’s actions initiated a process that “recovered and removed from the girl from trauma.”
Use of facial recognition tools by Kansas law enforcement uncovered the largest forced labor trafficking case in U.S. history, all through identifying cases of driver’s license fraud in the state’s database.

Bringing Child Sexual Predators to Justice

A 15-year-old girl in Scranton, Pennsylvania, was sexually assaulted by an adult male she met online. Beyond seeing her assailant in person, the only additional information she had was from his online profile. Police were able to use facial recognition on one of the digital images to provide some potential matches from a state database, from which the victim was able to identify a likely match. After additional investigative work, authorities obtained a search warrant for the home of the identified suspect, who later admitted to the crime.
A man accused of sexually assaulting a 10-year-old girl was apprehended in Oregon after a 16-year manhunt. Using facial recognition technology, the Federal Bureau of Investigation (FBI) was able to identify the suspect after a positive match was found when the suspect sought to acquire a U.S. passport.
Facial recognition technology was used to help locate and apprehend a convicted pedophile who had been on the run for 14 years, returning him to New Mexico to face justice.
An investigative unit within the U.S. Department of Homeland Security used facial recognition technology to help U.K. authorities identify a Missouri man suspected of sexually exploiting an infant child, as part of an operation to solve thousands of cold cases.
In 2019, New Jersey law enforcement arrested 19 online child predators following an investigation that was aided by facial recognition.

Solving Other Sex Crimes

In Maryland, an unknown subject went to the front door of a residence and began sexually stimulating himself in front of a security camera. The use of facial recognition by local police provided an investigative lead – a person that had conducted similar behavior in years prior. Upon further investigation, the case resulted in a confession by the suspect and criminal charges related to indecent exposure.

In Montgomery County, Maryland, an unconscious person was reported. Responding officers found a disoriented pregnant female subject who was unable to recall anything from the past two days. Eventually, the female victim was able to recall potentially being drugged and later raped by an unknown suspect. Facial recognition was used to generate a lead from a photo of the suspect available from security cameras nearby.

The New York City Police Department (NYPD) arrested a 31-year-old man after he tried to rape and assault a woman on the subway at a station in midtown Manhattan. The NYPD facial recognition team utilized a video taken by a bystander to identify the man, based on a photo taken from a previous arrest.

Catching a Suspected Subway Terrorist

New York City Police Department (NYPD) detectives used facial recognition technology to identify a man who sparked terror by leaving a pair of rice cookers in the Fulton Street subway station. Detectives pulled still images of the suspect from security footage and used facial recognition software to compare them to NYPD’s arrest database. The system returned several hundred potential matches, and after multiple stages of review and confirmation using other methods, the suspect was identified in just one hour.

Finding a Killer Targeting LGBTQ+ Victims

Three members of the LGBTQ+ community were shot and killed by a man at a local home in Detroit, Michigan. The Detroit Police Department used facial recognition, in combination with other investigative tools, to help identify the suspect based on video images from a nearby gas station.

Identifying Victims

Following police response to a shooting/robbery in Maryland, and the victim could not be identified and remained in critical condition. Therefore, notification to his family had not been made. Images obtained from the victim’s cell phone screen were queried, and a lead was developed. Using other known images of the candidate, it was learned the candidate had a birthmark on his temple, this information was shared with investigating officers who confirmed that the birthmark was present. The investigators were then able to contact the victim’s family, and they responded to the hospital. While the victim ultimately succumbed to his injuries, quick work by investigators aided by facial recognition technology enabled the family to make it to the hospital before he passed.

Responding to Health Emergencies

Local law enforcement responded to a health emergency involving an individual at a local airport in College Park, Maryland, with no shirt or shoes, stating they wanted to “fly to outer space/the stars,” but the subject left the area before units arrived. An officer was able to locate the subject after subsequent calls from concerned citizens nearby; however, they had no identification and could not communicate coherently. An image was taken of the subject and queried, producing a potential matching female identity. At first, officers on the scene believed it was not a match because the individual was male. Upon further investigations the lead proved correct, as the transgender man’s identity was confirmed by his father, who had been contacted in another state. The man had reportedly not been the same since taking LSD the previous week. He was reunited with a family member and then taken to a local hospital for evaluation.

An unknown person in Annapolis, Maryland was posting plans to commit suicide on open-source media. Reports were made to the police by concerned people who saw this post. Due to what was written, police believed suicide was eminent and attempted to identify this person using a still image from open sources. This image was used with facial recognition technology and generated a lead through a driver’s license photo. Through further investigation, the suicidal person was identified, and the police and a crisis team were sent to the person’s address. Police were able to locate the suicidal person, who was provided with assistance.

Solving Violent Crime

Local law enforcement investigated a violent assault on a public bus in Baltimore, Maryland. Images of the suspect and the incident were obtained through security camera footage from the bus. Information was disseminated to law enforcement partners seeking assistance with the case. A comparison was made with a law enforcement database, and an investigative lead was developed and provided to the investigating agency. Upon further investigation it led to the arrest of the assailant who was identified by the victim.

In Annapolis, Maryland, Jarrod Ramos, later known as the “Capital Gazette Killer,” was angered by a story the Capital Gazette newspaper ran about him in 2011 and brought a lawsuit against the paper for defamation, which a judge later dismissed. In 2018, Ramos entered the newspaper’s headquarters in Annapolis, Maryland, with a shotgun and killed five employees, leaving two others critically injured. While local police took the suspected gunman into custody, he had no identification and would not speak to investigators, and a fingerprint database was not immediately returning any matches. Detectives obtained an image of Ramos and used facial recognition, which generated a lead in the case. Through further investigation, detectives were able to positively identify Ramos, and search warrants were conducted at his residence. He pleaded guilty in the case and was sentenced to five consecutive life sentences.

Two suspects armed with guns walked into a liquor store in Towson, Maryland, and announced a robbery, taking aim at a 68-year-old clerk. The clerk, fearing for his life, pulled out a gun and shot one of the people robbing the store, who was later pronounced dead at the scene. The second person involved in the robbery escaped but was later identified after social media connections and image analysis using facial recognition led detectives to him. He was successfully prosecuted and convicted of attempted robbery.

A Facebook user claimed on open-source media he was ready to attack and kill law enforcement “tyrants.” The same Facebook user also commented online on a police press release from Montgomery County, Maryland, suggesting he would use hydrofluoric acid containers above entry points to injure law enforcement. The subject later went on Facebook Live and announced his intent to livestream the execution of a law enforcement officer in Texas. Facial recognition was used by local police to quickly generate a lead from open-source photos. Through additional investigation, investigators were able to identify this individual and locate him in Texas. After a lengthy pursuit, he was arrested and charged with terrorist threats against an officer, evading detention with a vehicle and unlawfully carrying a weapon.

In Las Vegas, Nevada, the murder of a man who was shot to death at an Airbnb was reportedly solved with the help of facial recognition as a “secondary tool” after images of the suspect were compared with a database of individuals arrested in the jurisdictions. According to the report, the use of the technology also aided the identification of the gunman in a shooting at a restaurant, an assailant who punched and fractured an individual’s skull at a car show, and those charged in several serious sexual assault cases.

In Detroit, Michigan, a man pleaded guilty to robbing a 53-year-old female victim at gunpoint while she was walking home from a bus stop, taking her purse and book bag. An investigation had been conducted using facial recognition as an investigative tool to assist in his identification from images provided by local security cameras. Other cases were also solved in a similar manner, which involved an alleged suspect wanted in connection with two robberies and a nonfatal shooting, a suspect charged in connection to a police shooting, and a suspect who robbed a business wielding a knife.

Fighting Organized Crime and Gang Violence

Local law enforcement in Maryland requested assistance from a partner agency with a firearms trafficking investigation, providing an image of a suspect. The image was run against a law enforcement database, and a potential lead was developed. Upon further investigation, detectives positively identified the suspect and executed a search warrant that resulted in the seizure of drugs, guns and ammunition.

A retailer reached out to law enforcement with information about an organized theft crew that had been targeting stores throughout Virginia, D.C. and Maryland. An image provided showed a male with unique tattoos on his neck and left hand. Facial recognition was used to generate a lead in the case. Upon further investigation, the individual was subsequently identified and charged.

Local law enforcement conducted a homicide/gang investigation involving a violent group responsible for multiple homicides, drug distribution, kidnapping and robbery in Anne Arundel County, Maryland. Digital images of persons of interest were obtained, and with the assistance of facial recognition, law enforcement was able to generate leads regarding three individuals involved. Through further investigation, individuals were positively identified, and probable cause was established to obtain a wiretap warrant. Though subsequent monitoring of communications, law enforcement was able to prevent at least three shootings, as well as interrupt a kidnapping. As a result of the investigation, over a dozen people were indicted and successfully prosecuted, multiple firearms were recovered, including an assault rifle, and drugs and a significant amount of U.S. currency were also seized.

Solving Firearms Trafficking

Local law enforcement in Maryland requested assistance with a firearms trafficking investigation in Prince George’s County, providing an image of a suspect. The image was run against a law enforcement database, and a potential lead was developed. Upon further investigation, detectives positively identified the suspect and executed a search warrant that resulted in the seizure of drugs, guns and ammunition.

Solving Burglaries

In Crownsville, Maryland, officers responded to a residential burglary captured on a home security camera. Using a facial image from the video, officers queried a law enforcement database using facial recognition, which provided a lead in the case. Upon further investigation, the person in the video was positively identified. He was charged and convicted of burglary and other charges.

Fighting Identity Theft

A string of fraudulent vehicle purchases in Montgomery County, Maryland, were carried out using information obtained via identity theft, harming both the identity victims and dealerships that lost property. The suspects had created false identification documents used to purchase the vehicles, combining their own image with the personally identifiable information from a victim. These images were queried, leads were developed and identities were confirmed through additional investigation, and five arrests were made. Some of the suspects were arrested when they arrived to pick up a vehicle, since by that time they had already provided their false identification with their true image.

A man was caught with the help of facial recognition technology used by the Maine Bureau of Motor Vehicles, after the individual faked his own death and assumed his dead brother’s identity to create fraudulent passports and other identification documents and commit Social Security benefits fraud.

The New Jersey Motor Vehicle Commission‘s (MVC’s) “Operation Facial Scrub” has resulted in charges against hundreds of defendants with identity theft, forgery and document fraud offenses. Using facial recognition software, the division, together with the MVC and State Police, identified individuals who have applied for and obtained driver’s licenses under false names. Numerous defendants had extensive criminal records, including sex offender and DUI convictions. Several defendants possessed valid commercial driver’s licenses under the fraudulent names. This has included even uncovering a vehicle title fraud scheme by two of MVC’s own employees.

Border Security

The U.S. Department of Homeland Security uses facial recognition at over 200 international airports and dozens of land entries and seaports to verify travelers. U.S. Customs and Border Protection (CBP) matches passport photos to a database to verify the identity of thousands of travelers entering and leaving the U.S. each week. In addition to providing more convenience to travelers, it has proven to be an important tool for border security. As of March 2023, more than 1,800 individuals have been intercepted by CBP attempting to enter the U.S. at airports and the land border under a fraudulent identity, after detection using facial biometrics.[2] Beyond simply violating laws against illegal entry, passport fraud fuels many other types of crime with far-reaching harms, including human trafficking and drug smuggling.

Health Care

Facial recognition technology can be used in health care facilities in a variety of ways that protect the health and safety for patients and staff, including securing valuable lab equipment and providing touchless access so that only trained and authorized personnel can enter sensitive areas of a facility like “clean rooms” in order to eliminate risk of contamination.

Many hospitals have already implemented facial recognition to eliminate the need for front-line health care workers to swipe badges or type in codes to verify their credentials. Reducing contact with potentially contaminated surfaces is key when staff must be on high alert around the spread of germs and disease. Additionally, facial recognition can be used to avoid patient misidentification in administering medicine and other treatments.[2] Such technologies also can perform dual security and hygienic functions in more common settings as well, providing touchless access to virtually any workspace and giving an added layer of hygienic security, as opposed to handing over a pass card or touching a fingerprint scanner.