Live Facial Recognition (LFR) technology

Live Facial Recognition (LFR) compares a live camera feed of faces against a predetermined watchlist in real-time. It is used to locate persons of interest by generating an alert when a possible match is found.

It analyses key facial features and generates a mathematical representation of these features. It then compares them against the mathematical representation of known faces in a database, generating possible matches.

Live Facial Recognition (LFR) is a tool that has been in development for a considerable amount of time. We are now at a stage where we are confident in its ability to accurately identify missing people or people of interest and support with wider policing aim to keep our community safe.

Live Facial Recognition (LFR) is a powerful proactive policing tool that can support both the prevention of crime and resolution if investigations.

We adhere to strict guidelines to ensure privacy protection. Data collected via Live Facial Recognition (LFR) is handled according to data protection laws and regulations. Personal information is only retained if a match is made and deemed necessary for investigation purposes.

This means that all captures of individuals that do not match someone from our pre-established watchlist are immediately deleted. This is an automatic function and the data is irretrievable.

In relation to captures and data that do cause an alert, this is only kept for 24 hours. Again, this is an automatic function and the data is irretrievable.

All CCTV footage generated from a mobile CCTV deployment is deleted within 31 days. This information with only be kept in the following instances:

• In accordance with the Data Protection Act 2018, MOPI and the Criminal Procedures and Investigations Act 1996; and/or
• In accordance with Hertfordshire Constabulary's complaints/conduct investigation policies

We will never pass biometric data to third party agencies.

Historically there have been issues with Facial Recognition technology and potential gender and ethnic bias. As the technology has developed over time this bias has reduced greatly.

An independent study conducted by the National Physics Laboratory (NPL) found that for a watchlist featuring 10,000 people of interest, the false positive rate is less than 1 in 10,000. A smaller watchlist of just 1,000 people had a false positive rate of less than 1 in 100,000.

The practical application of this tactic by other police forces and accompanying data and results suggests the technology is more reliable than false positive alert rate forecast by the NPL study.

The full results are of the NPL’s commissioned report: ‘Facial Recognition Technology in Law Enforcement Equitability Study’ and further information about the testing can be found in the Test Strategy for Facial Recognition Technology in Law Enforcement Equitability Study.

There are settings the algorithm can be operated at where there is no statistical significance between demographic performance.  

With that said, to add a further layer of mitigation, all matches are reviewed by specially trained officers who will provide final confirmation on whether the match is correct. Only then will an officer engage with the individual and conduct a further assessment on the match before taking appropriate action. 

Our use of Live Facial Recognition (LFR) will be overseen by independent bodies and subject to regular audits to ensure accountability and compliance with legal and ethical standards. 

All alerts are first reviewed by a specially trained officer.

In the rare instance of a misidentification, our aim will always be to rectify the matter as quickly as possible, ensuring minimal distress to the individual and wider public.

All deployments are reviewed and assessed with the aim of establishing any improvements that can be made.

With each deployment of Live Facial Recognition (LFR), a new watchlist is collated.

These lists are made up of people wanted by Hertfordshire Constabulary or the courts and/or are subject to strict court orders that they have failed to adhere to.

Live Facial Recognition (LFR) is also a vital tool for safeguarding vulnerable individuals. We may, at times, utilise Live Facial Recognition (LFR) in the search for high risk missing people, to locate victims of crime or people that have been identified as potentially having crucial information relating to a serious crime investigation.

Unless you are on the watchlist you cannot be matched. ​

All deployments of Live Facial Recognition (LFR) will be communicated in advance via our website and social media channels.

This is unless advance publication would unnecessarily impact the operational purpose of the deployment. For example, if someone wanted for a serious crime is expected to be in a certain area, publicising the deployment beforehand might stop them from showing up, making it harder to arrest them.

The results of each deployment will also be published on our website within five working days of its use.

In public spaces where Live Facial Recognition (LFR) is used, it is generally not feasible to opt-out.

Legally, people are allowed to avoid walking past a Live Facial Recognition (LFR) camera or have their face covered without being stopped or arrested by an officer.

The only reasons an officer may intervene, is in the normal way if camera's weren't there; that an individual's behaviour appears suspicious, or in the very specific and rare circumstances where Section 60AA of the Criminal Justice and Public Order Act 1994 is in place for that area, meaning face coverings aren't allowed for the specified time it's live.

In either circumstance, officers are expected to use their normal policing powers and judgement.

Individuals have the right to enquire about how their data is being used and to challenge any misuse or inaccuracies.

Concerns or complaints can be sent via email.

To maintain operational integrity, we will publish the location and date of any Live Facial Recognition (LFR) deployment seven days in advance in adherence with College of Policing guidelines and the force standard operating procedure.

The next deployment will be in Stevenage town centre on Friday, 01 May 2026.

Information on previous LFR deployments:

• Slam Dunk Festival, Hatfield (1) - 23 May 2026
• Slam Dunk Festival, Hatfield (2) - 23 May 2026
• Slam Dunk Festival, Hatfield (3) - 23 May 2026
• Slam Dunk Festival, Hatfield (4) - 23 May 2026