Object Detection

More Than a Strike: Team StrikeCam Reimagines Airbase Security from the Ground Up 

When Team StrikeCam started their Hacking for Defense journey at USC, the problem they were handed seemed deceptively simple: improve aircraft strike identification and response during base security breaches. But what they uncovered was something deeper—and far more consequential.

StrikeCam wasn’t just responding to physical threats. They were building a new kind of trust infrastructure between humans and autonomous surveillance tools.

What Was at Stake 

As global adversaries increase their use of drones, autonomous ground vehicles, and asymmetric tactics to probe and exploit U.S. airbase vulnerabilities, perimeter defense has become a critical capability gap. The Air Force’s current systems rely heavily on personnel and analog cameras—leaving gaps in visibility, delayed response times, and limited forensic traceability.

StrikeCam’s insight? This wasn’t a hardware problem. It was a data confidence problem.

Prototyping a Smarter Perimeter 

With support from CMP’s Innovation Fund and early feedback from Air Force Security Forces at installations like Nellis and Edwards AFB, the team began building an AI-assisted security camera prototype optimized for perimeter environments.

The design prioritized three capabilities:

1. Automated object and event tagging for rapid identification of strike triggers

2. Lightweight field deployment for austere or temporary perimeter zones

3. Edge computing capacity to reduce latency and dependence on bandwidth-heavy comms

Their MVP integrated commercial-grade sensors with custom firmware trained on a dataset of known threat profiles. By mid-semester, they were running controlled detection simulations with Air Force mentors watching live.

Operators of Another Kind 

Aryan Vats led technical prototyping and integration, balancing performance and ruggedization for base perimeters. 

From Camera to Confidence 

StrikeCam’s system is now under review for testing within a classified defense innovation environment. Their long-term vision isn’t just smarter cameras—it’s a networked security mesh that learns and adapts with every incident.

Advice to Future Students

 “If you want to make a difference, design with the end-user in the room,” Aryan said. “The people watching the feeds are your real customers.”