AI Gun Detection System Monitors Crowds for Firearms

At last weekend’s Miami Grand Prix, all attention was on Lando Norris, who secured his first Formula One victory for McLaren. Meanwhile, behind the scenes, AI-powered software monitored the crowds for firearms.

AI-based gun detection platform from ZeroEyes was part of the on-site security team at the Miami International Autodrome, where it scanned the race’s 275,000 attendees, alerting authorities within seconds of identifying a weapon.

ZeroEyes’ Miami deployment came as part of a partnership with Louisiana-based Wireless Guardian to demo its AI detection software at a large-scale event.

Technology from the two companies combined to ensure the race was safe for drivers and spectators.

All Eyes on Firearms

ZeroEyes’ AI gun detection solution was trained on hundreds of thousands of images and videos of individuals holding weapons.

ZeroEyes was founded by a group of former Navy SEALs in response to the Parkland high school shooting where a former student shot and killed 17 people in 2018. The founders created the startup “to protect people from having to face similar circumstances.”

Their software analyzes more than 36,000 images per second from existing camera feeds to monitor for firearms. If the AI detects a possible firearm, the images are sent to ZeroEyes’ operation center where a human specialist will confirm the report.

The company has operating centers in Philadelphia, Pennsylvania and Honolulu, Hawaii where firearms experts including military veterans and former law enforcement personnel are notified if the software detects a firearm.

The experts verify incoming reports before sending the information to law enforcement and the property’s security staff through a variety of communication methods including app-based notifications and phone calls.

The AI software provides actionable intelligence for officials including a description of the shooter, their last known location and the weapons they’re carrying.

Sam Alaimo, ZeroEyes co-founder told AI Business the whole process can be completed in three to five seconds.

“The human loop is going to verify the gun and the client will not see it until a human has eyes on it. We want to make sure we don’t send a false positive,” he said. “Sometimes it might take a little bit longer if there’s some sort of lag with the cameras or a bad internet reception. But the vast majority of the time it’s three to five seconds.”

At large-scale sporting events like the F1, Alaimo said ZeroEyes’ technology wasn’t just scanning crowds watching the event, but external sites like parking lots.

“If you look at the majority of mass shootings within commercial and education settings, the majority start outside and then they move inside,” Alaimo said. “So you’re looking for guns in a parking lot, not just waiting for the person to walk through a metal detector at the front door. That’s where we can have a tremendous amount of value out there in the parking lots, the rear loading docks and the hallways because there’s a lot of ways to get around metal detectors.”

The company’s technology has already been recognized by the Department of Homeland Security, becoming the first video analytics system to receive DT&E certification under the SAFETY Act.

Continual Training and Privacy Support

To continually update its software, ZeroEyes has built a dedicated unit to create training footage.

With a sizable green screen room not dissimilar from a Hollywood movie set, the company can train its software on a variety of cameras that would be used in environments where the software might be deployed.

ZeroEyes green screen room where its team creates training images | Credit: ZeroEyes

“All day, every day, we have an inventory of hundreds of different weapons and we have people walking around in high light, low light, at high and low angles, coming in and out of vehicles,” Alaimo said. “Every variation we can try to conceive of, to make sure that in the real world, we’re never surprised by the orientation of the gun, the way it’s being held, what the background is, the speed at which it’s moving. We’re constantly varying it and we’re constantly trying to project to see what can happen and then prepare for it ahead of time.”

ZeroEyes designed its software solution with privacy in mind.

Despite continuously scanning for firearms, its AI technology is only trained to detect guns and not potentially invasive incursions like facial recognition and voice recording.

“We deliberately built the company not to be able to recognize faces and not to store biometric data,” Alaimo said. “We don’t even store the live video feed.”

“We don’t want to invade privacy. We’re all patriots who founded the company. We want nothing to do with facial recognition or biometric data. It’s just looking for that inanimate object… we’re not taking anybody’s guns away, we want you to know if there’s an assault rifle in front of an elementary school.”

Beyond Firearms?

ZeroEyes started out being deployed in schools but has since expanded to industries across 40 states, including health care settings, manufacturing sites and shopping malls, as well as college campuses, commercial property units and Fortune 500 corporate campuses

Its technology has been used beyond the U.S., with clients in the U.K., the Bahamas and South Africa.

There’s also a government-focused part of the company that’s split off, providing AI solutions to the Defense Department to detect boats and drones, though that tech is not being offered commercially at present.

Expanding its commercial-focused software beyond firearms to knives, for example, is a request Alaimo said ZeroEyes has received but tempered its feasibility, for now.

“That’s something we’re not able to work on at the moment because a knife is too similar to so many other things, the false positive load to be through the roof with technology the way it is, but for the future, we’re going to keep doing what we’re doing.”

Monitoring an F1 Grand Prix race is a larger task than monitoring cameras in a school and the U.S. is set to co-host one of the largest sporting events in the world in 2026: The FIFA Soccer World Cup.

Alaimo said AI-powered firearm detection systems could help security teams at future large-scale sporting events. He said the Miami demo was a great way to verify its software works on a much larger scale.

“We need to internally figure out how that’s going to work, we typically do long-term contracts, not pop-up events,” Alaimo said. “But I think that there’s an easier way to make that work and I hope we have the opportunity to do that.”

Combining Sensors With AI

Wireless Guardian, ZeroEyes’ partner in Miami, was founded in 2017. Its sensor-based solutions are designed to safeguard the public in large-scale spaces like events, as well as confined environments like retail stores.

Its technology can detect potential weapons and can even triangulate all radio frequency devices.

Wireless Guardian’s solutions can be used to detect people and devices in a crowd and alert authorities in real time. The company’s website said its systems can perform facial captures and audit both individuals and vehicles.

Wireless Guardian said before the race that its technology “empowers security personnel to proactively address potential threats and maintain a secure environment” throughout high-profile events like the F1.