Banned-Fan Watchlists
Real-time matching at entrances against the exclusion list the club or the league maintains, with an alert to stewards and police before the turnstile.
See Ethical Facial Recognition
A stadium is a crowd problem with a watchlist problem inside it, compressed into ninety minutes twice a week. Both have to work at the same time, on the same cameras.
In Production
IREX has become the standard security platform for all national stadiums in the Premier League of Bosnia and Herzegovina. The system identifies banned football fans attempting to enter matches by matching attendees against watchlists in real time, which helps prevent violence and keeps events safe. Additional modules cover other public-safety threats within and around stadium perimeters on match days.
The two capabilities that matter here pull in opposite directions. Watchlist matching needs to work at an entrance under load, in poor light, with people moving. Crowd analytics needs to see density across a queue or a concourse and raise an alert early enough for a steward to act. IREX runs both on the same estate.
Crowd analytics estimates up to 3,000 people within a user-defined region of a camera view, and provides both real-time dangerous-crowding alerts and precise post-event counting for review.
Match Day
Real-time matching at entrances against the exclusion list the club or the league maintains, with an alert to stewards and police before the turnstile.
See Ethical Facial Recognition →Density alerts across queues, concourses and entry points, plus precise counting per area for post-event review.
See Crowd Management →Intrusion and restricted-zone monitoring around the stadium footprint on match days.
See Perimeter Intrusion Detection →Visual and acoustic threat detection at entries and inside the ground.
See Weapon & Gunshot Detection →A bag left in a concourse or against the perimeter fence, raised while the ground is still filling.
See Abandoned Object Detection →Early detection across large enclosed and open areas, ahead of point sensing.
See Fire & Smoke Detection →On Camera
Both ramps carry the same crowd toward the same gate, and only one of them is a problem. Crowd analytics reads density per region rather than counting heads through a turnstile, so an alert arrives while a steward can still hold a flow or open a second gate — and after the match the same regions give a precise count for the safety review, estimating up to 3,000 people in a defined region of a camera view.
Crowd Management →
That is the production use case in Bosnia and Herzegovina, across every Premier League national stadium. The engine is trained on a proprietary 40-million-image dataset optimized for real-world CCTV conditions, and how strict a match has to be is set entrance by entrance and tuned on your own turnstiles during the pilot.
Up to 3,000 within a user-defined region of a camera field of view. Larger areas are covered by defining regions across several cameras.
It depends on your jurisdiction and on the legal basis for the exclusion list, which is exactly why the Case ID gate exists: the operator records the lawful grounds before a privacy-sensitive action, and the record is auditable. IREX also runs background checks on prospective clients and declines business it judges high-risk.
With a pilot: one site, three to five use cases, six to twelve weeks, and two or three measurable success criteria agreed in writing before it begins. Accuracy is benchmarked on your own cameras, and the measured numbers go into the contract.
Keep Reading
Crowd and watchlist analytics can only be judged under load. Pick your biggest match.