The Metropolitan Police Service Achieved Over 170 Arrests in Six Months:
London’s First Static Live Facial Recognition Infrastructure
“The pilot has led to a significant number of arrests, including individuals suspected of serious offences and violence against women and girls, demonstrating that this pioneering technology is helping to make our streets safer.”
Jason Perry, Executive Mayor of Croydon
Source: Metropolitan Police press release, 19 January 2026
The Metropolitan Police Service (MPS) is the largest police force in the United Kingdom. Covering 32 London boroughs and an area of around 620 square miles, the MPS protects a population of approximately 9 million people, alongside millions of commuters and visitors each year.
London’s dense urban environment, critical national infrastructure, and status as the UK’s political and economic centre present complex policing challenges. The MPS heavily relies on intelligence, specialist capabilities, and advanced technology to police one of the world’s most dynamic cities. This includes extensive CCTV networks, automatic number plate recognition, and advanced facial recognition capabilities.
Following Home Office funding to trial a public space static deployment of Live Facial Recognition (LFR), the MPS sought a powerful static-infrastructure solution. NEC software was identified as the preferred platform, and Bedroq was engaged to design and deliver the supporting architecture required to make this public-space deployment viable.
Complex Technology
LFR had previously only been delivered via specialist vans deployed across London. While operationally effective, van-based deployment limited scalability and created operational overheads. Each deployment required technical on-site configuration and significant officer presence, all while the visible nature of the vans frequently attracted confrontation and diverted resources.
The MPS wanted to transition from short-term, reactive deployments to a static, preventative model mounted on existing lampposts. Croydon’s North End retail district – a high-footfall area with a history of robbery, knife crime, and theft – was selected as the pilot location.
To achieve this, the MPS required a fully designed infrastructure capable of delivering consistent facial recognition accuracy, low latency between detection and officer alert, secure connectivity, and data handling. Furthermore, given the sensitivity surrounding LFR in public space, the system needed to provide demonstrable assurance that equipment could be powered down when not in use.
A Powerful Solution
Bedroq designed and delivered the complete end-to-end infrastructure to support a static LFR deployment in Croydon.
Through detailed onsite modelling and physical scoping of the high street, Bedroq engineered a camera layout capable of achieving the required 80-pixel interocular density to support accurate hits. By using the latest high-resolution 8K Axis cameras and optimising placement, Bedroq reduced the requirement from an initial estimate of more than 60 cameras to just 15 strategically positioned units. This significantly reduced the infrastructure footprint while maintaining full coverage and the highest operational standards.
Furthermore, Bedroq designed and implemented the full networking architecture, including secure switching, resilient connectivity, and supporting data centre infrastructure. Rigorous latency testing across copper and fibre links confirmed that alert notifications could be delivered within seconds of a subject entering a camera’s field of view, meeting the MPS’s operational expectations.
To provide operational control and public accountability, Bedroq developed a bespoke web-based interface enabling granular management of Power over Ethernet (PoE) switches housed within street cabinets. This allows the LFR team to power individual cameras or associated Wi-Fi equipment on or off as required. It also provides clear real-time status indicators across three operational zones. The system ensures full visibility of equipment state and supports transparency across the operation.
Creating Safer Streets
The trial went live in October 2025. Within the first 18 minutes of the first activation, a live alert was generated. Throughout the duration of the 6-month trial, across 24 separate operations, more than 170 wanted criminals have been arrested by the Metropolitan Police. This is the equivalent of one arrest every 35 minutes.
By moving from mobile vans to static infrastructure, the Metropolitan Police Service has reduced deployment complexity and removed the need for visible street-based vehicles in the area, allowing officers to be more conspicuous. This has lowered public disruption and reduced the resource burden associated with van deployments, enabling officers to operate from a dedicated local office space rather than managing technology from the street.
Retailers along Croydon’s North End have reported a significant reduction in incidents throughout the day, including during periods when the system is not actively running. The presence of static infrastructure has delivered an ongoing deterrent effect, contributing to improved safety and stronger working relationships between the MPS and local businesses.
The Croydon pilot demonstrates that static public-space LFR infrastructure can deliver measurable operational impact while maintaining control, accountability, and scalability. It now provides a blueprint for potential expansion into other key locations across the country.
“We will continue using static cameras in Croydon as part of our regular live facial recognition deployments which play a vital part in keeping London safe.”
Lindsey Chiswick, national and Met lead for live facial recognition.
Source: Metropolitan Police press release, 13 May 2026
