SciTransfer
LETS-CROWD · Project

AI-Powered Crowd Monitoring and Risk Assessment Tools for Safer Mass Events

digitalPilotedTRL 6

Imagine you're responsible for security at a huge concert or football match with tens of thousands of people. How do you spot trouble before it starts? This project built smart camera software that watches crowd density and movement in real time, plus a risk assessment system that helps security teams plan ahead. They also created tools that scan social media for early warning signs and help police forces across Europe share best practices — all tested at real events with 7 different law enforcement agencies.

By the numbers
7
Law enforcement agencies involved in practical demonstrations
11
Use cases tested during demonstrations
18
Consortium partners across Europe
8
Countries represented in the consortium
2
Computer vision software tools developed (crowd monitoring + person re-identification)
24
Total project deliverables produced
The business problem

What needed solving

Managing security at mass gatherings — concerts, sports events, public celebrations — is a constant challenge. Security teams rely on manual camera monitoring that misses emerging threats, and risk planning is often based on gut feeling rather than data. When something goes wrong in a crowd of thousands, response time and early detection are the difference between a controlled situation and a disaster.

The solution

What was built

The project built two computer vision software tools — one for real-time crowd monitoring and one for person re-identification and search — plus a dynamic risk assessment methodology and a policy-making toolkit with an empirical database for security planning. These were tested through 11 use cases with 7 law enforcement agencies.

Audience

Who needs this

Private security companies managing large event venuesStadium and arena operators responsible for crowd safetySmart city technology vendors building public safety platformsNational police forces and law enforcement modernizing surveillance capabilitiesEvent insurance companies assessing crowd risk
Business applications

Who can put this to work

Event Security & Management
mid-size
Target: Private security firms managing large-scale events

If you are a security company responsible for protecting concerts, sports matches, or festivals — this project developed two computer vision tools for real-time crowd monitoring and person re-identification that were demonstrated with 7 law enforcement agencies across 11 use cases. These tools can help your operators detect dangerous overcrowding or locate specific individuals faster than manual camera scanning.

Smart City & Public Safety Technology
enterprise
Target: Companies developing video surveillance or public safety platforms

If you are a technology vendor building smart city surveillance solutions — this project created crowd behaviour forecasting software and semantic intelligence tools for social media monitoring. Tested by 7 LEAs across 8 countries, these components could be integrated into your existing platform to add predictive crowd analytics and automated threat detection capabilities.

Venue & Stadium Operations
enterprise
Target: Stadium operators and large venue management companies

If you are a venue operator dealing with crowd safety compliance and liability — this project built a policy-making toolkit with an empirical database and analytical tools for security planning. The dynamic risk assessment methodology, validated through 11 use cases, can help you model different event scenarios and deploy the right security measures before doors open.

Frequently asked

Quick answers

What would it cost to license or deploy these crowd monitoring tools?

The project data does not include pricing or licensing terms. As a publicly funded RIA project with 18 consortium partners including 8 SMEs, commercialization terms would need to be negotiated with the coordinator ETRA Investigacion y Desarrollo (Spain) or relevant technology partners. Based on available project data, no commercial pricing has been published.

Can these tools scale to large venues with 50,000+ attendees?

The project specifically targeted mass gatherings and tested solutions through practical demonstrations with 7 law enforcement agencies following 11 use cases. The computer vision tools were designed for crowd monitoring at scale, though specific capacity benchmarks are not detailed in the available deliverables.

Who owns the intellectual property and can I license it?

This was a Research and Innovation Action (RIA) funded by the EU with 18 partners across 8 countries. IP ownership typically follows the EU grant agreement, where each partner owns the results they generate. Contact the coordinator ETRA Investigacion y Desarrollo SA in Spain for licensing discussions.

Does this comply with GDPR and privacy regulations?

The project explicitly states it assessed how security measures affect citizens and respected EU fundamental rights. The tools are described as 'human-centred,' indicating privacy-by-design principles were applied. However, deploying person re-identification technology will require a Data Protection Impact Assessment under GDPR.

How long would integration into existing security systems take?

Based on available project data, the computer vision tools were built as two separate software modules — one for crowd monitoring and one for person re-identification and search. The deliverables describe them as prototypes, so integration into production security systems would likely require additional engineering and customization.

What kind of technical support or training is available?

The project involved 7 law enforcement agencies in practical demonstrations across 11 use cases, generating operational know-how. The consortium includes 8 industry partners and 3 universities that could potentially provide training. Contact the coordinator for current support availability.

Is this technology already deployed anywhere?

The project ran from 2017 to 2019 and conducted practical demonstrations with 7 LEAs. The deliverables describe the computer vision tools as 'prototypes,' indicating they were demonstrated but not yet commercially deployed at the time the project closed.

Consortium

Who built it

The LETS-CROWD consortium brings together 18 partners from 8 countries (Belgium, Germany, Spain, Israel, Italy, Portugal, Romania, UK), with a strong operational focus — 44% are industry partners and 8 are SMEs. The coordinator, ETRA Investigacion y Desarrollo SA from Spain, is a private company (not an SME), suggesting commercial capacity. With 7 additional non-industry/non-university partners (likely law enforcement and public sector bodies), this consortium was built around end-users rather than purely academic research. The mix of 3 universities and 8 industry players indicates technology was developed with real deployment in mind, not just publications.

How to reach the team

ETRA Investigacion y Desarrollo SA, Spain — a private R&D company. Look for their public safety or smart city division.

Next steps

Talk to the team behind this work.

Want to explore how LETS-CROWD's crowd monitoring and risk assessment tools could fit your security operations? SciTransfer can connect you directly with the right consortium partner for your needs.