AI Case Study
Patriot One are developing concealed weapon identification from deploying machine learning on radar-generated images
Patriot One aims to use radar located at key traffic points - transport system turnstiles, reception desks, entry doors - to capture images of carried, and concealed, devices. Machine learning then is deployed to identify potential weapons. Theoretically this enables widely deployed, potentially fully automated, security. In reality, issues with false positives (depending on the required confidence level trade-offs) remain a challenge.
Public And Social Sector
Accoding to Atlantic:
'Patriot One’s approach relies upon a specialized radar technology—developed in partnership with McMaster University—that can be hidden behind security desks or in the walls near a building’s main access points. The radar waves bounce off a concealed object and return a signal that reveals its shape and metal composition. From there, a deep-learning tool recognizes the radar patterns that match weapons including handguns, long guns, knives, and explosives. So far, one of the biggest challenges is training the tool to ignore the usual clutter in a student’s backpack, such as wadded-up gym clothes, a textbook, or a pencil case.'
Patriot One claim that the Best Detection Performance to-date is:
• Sensitivity (true positive rate) 91.6%
• Specificity (true negative rate) 94.4%
• Accuracy (overall) 93.0%.
Commentators have noted the high risk of false positives - including for balled up clothing or pencil cases. Part of the challenge is that field deployment would likely significantly increase execution challenges.
Security issues - ranging from terrorism to school shootings - have increased the demand for simple ways to flag potential armed risks to high profile locations and the individuals inside.