Talk to us

Islamic Society of Baltimore: safer crowds, with no facial recognition.

The Islamic Society of Baltimore (ISB) serves thousands of community members every day. Traffic at peak prayer times, especially Jummah, is hard to manage. The security team is small, and volunteers spent hours counting donations by hand. We built one system to help with all three.

No facial recognition and no personal data. How privacy works

Talk to us

Community center in Baltimore

Four problems, four fixes.

ISB serves thousands of people through prayer services, educational programs and community events. Safety and operations have to keep up, with limited security resources.

  1. 01

    Traffic at peak times

    Real-time traffic analytics help make better use of parking and cut congestion at peak times.

  2. 02

    A small security team

    Computer vision tracks how crowded each area is and flags unusual activity. Each alert reaches the security team instantly, so they can coordinate emergency responses.

  3. 03

    Donations counted by hand

    Donations go through the Mohid system. Tracking and financial reports run automatically, so volunteers no longer count by hand.

  4. 04

    No live view of the building

    A data hub connects the security, donation and facility systems. Real-time dashboards show facility usage and help with planning.

Privacy comes first.

ISB needed safety without giving up privacy or community values. This is what the system does and doesn’t do.

What it does

  • Measures how crowded each area is, in real time
  • Flags unusual activity and alerts the security team
  • Runs its AI on site, without depending on the cloud

What it does not do

  • Use facial recognition
  • Collect personal data

The results, in numbers.

24/7

Automated monitoring, with instant alerts

50%

Fewer security incidents

40%

Less traffic congestion during Jummah

100%

Automated donation tracking and reporting

15+

Hours saved every week on manual work

Staff also get real-time crowd management at large events, and better coordination in emergencies.

What an agent could add next.

These ideas use the same data. They are not part of the results above.

  • A daily summary

    Sends crowd, traffic and donation numbers to staff each morning, so nobody has to open a dashboard.

  • Staffing from the data

    Suggests when to add security and volunteers, based on past peak times and how staff were deployed.

  • Event-day planning

    Before a large event, it reviews past traffic and drafts a plan for the team to approve.

  • Alerts that get followed up

    If an alert goes unanswered, the agent reminds the next person on the list and keeps a record.