Mobile Concealed Weapon Detection

In a world where security challenges are rapidly evolving, traditional screening methods are struggling to keep pace. Fixed metal detectors and bulky infrared scanners often fall short in offering non-intrusive, real-time threat detection. That’s why we embarked on a mission to redefine public safety with our Mobile Concealed Weapon Detection System—a project that blends cutting-edge thermal imaging with advanced AI to transform security screening on the go.

Why We Started

We recognized that modern security demands flexibility and speed. Law enforcement, airport security, and event monitoring require solutions that are not only accurate but also discreet and easily deployable. Our system is designed to be integrated into mobile devices, making real-time threat detection both accessible and scalable. This vision led us to explore the untapped potential of thermal imaging, a technology that can “see” concealed weapons by detecting subtle changes in the body’s heat signature.

The Breakthrough Moment

Our journey began with conventional approaches using standard convolutional neural networks (CNNs). While these early prototypes showed promise, they couldn’t meet the precision and speed needed for on-the-go security. Then, during extensive data analysis, we discovered something unexpected: a concealed handgun consistently disrupted the thermal gradient of the human body. This observation sparked a breakthrough idea.

Inspired by advances in transformer models, we developed the Thermal Transformer Fusion (TTF) algorithm—a novel, dual-stream approach that changed everything. Instead of relying solely on raw pixel data, TTF processes thermal images through two complementary streams:

  • Thermal Stream: A lightweight deep learning model that captures the essential heat signatures and temperature gradients in real time.
  • Attention Stream: A transformer-based module that zeroes in on anomalous regions where the thermal gradient is disrupted, adapting dynamically to varying environmental conditions.

By fusing these two streams, TTF delivers rapid, real-time detection—achieving over 10 frames per second on mobile devices—while ensuring high accuracy and maintaining privacy with a fully GDPR-compliant design.

Key Advantages of Our System

  • Real-Time Threat Detection: Our system processes live thermal data at remarkable speeds, providing immediate alerts and enabling quick decision-making in the field.
  • Non-Intrusive & Privacy-First: Unlike conventional detectors, our solution is non-intrusive and designed with a privacy-first approach, ensuring compliance with European GDPR standards.
  • Versatile & Scalable: Perfect for integration into law enforcement body-worn cameras, airport security systems, and public event monitoring, our solution adapts to a wide range of security applications.
  • Cost-Effective Implementation: Leveraging affordable thermal cameras such as the InfiRay T3 Pro and TS001, we make state-of-the-art security accessible without heavy infrastructure investments.

Moving Forward

We’re currently hard at work refining every aspect of this project—from data collection and model training to on-device optimization. Our focus is on pushing the boundaries of mobile security screening and delivering a solution that not only meets today’s needs but also adapts to future challenges.

This journey is about more than just technology—it’s about creating a safer, more secure environment for everyone. We’re excited to share updates as we progress, and we invite you to follow our journey, share your thoughts, and join us in shaping the future of public safety.