Blog: Artificial Intelligence Detects IEDs at Military X-Ray Checkpoints – EE Times
The United States Air Force is collaborating with Synapse Technology, an artificial intelligence and defense company, to develop an artificial intelligence (AI) platform that can integrate with the X-ray machines currently deployed at U.S military base entry-control points (ECPs) in order to detect threats like Improvised Explosive Devices (IEDs).
The initiative will be moderated through the Air Force’s AFWERX program. “We look forward to collaborating closely with government agencies to solve their core use cases and help increase safety for our country’s soldiers,” said Madeline Zimmerman, business development lead for Synapse Technology.
The program, which was established in 2017 by the Secretary of the Air Force, is “a catalyst for agile Air Force engagement across industry, academia and non-traditional contributors to create transformative opportunities and foster an Air Force culture of innovation.” “The ultimate aim is to solve problems and enhance the effectiveness of the Air Force,” according to a statement from the group.
Currently, X-ray machines are used to screen visitors coming for threats such as guns, knives, and improvised explosives. “Threats evolve quickly, and AI has the potential to detect emerging IED threats, providing security against current and future threats while giving the warfighter more operational flexibility,” said Synapse CEO Peter Kant.
By integrating an AI component into existing X-ray machines on US military bases, the US Air Force hopes to both raise the level of security and free up highly-trained soldiers to focus on critical duties. “The soldiers at checkpoints are not trained to sit and monitor an x-ray machine…It is outside their core competency. They are trained to be a soldier,” said Ian Cinnamon, president of Synapse Technology. “ The Airforce thinks that if we can alleviate cognitive burden of these takes, there will be a lot of room to reassign the workers to tasks that they are better trained to do.
Founded a decade ago, the idea for an AI-enabled x-ray machine came out of the study of cognitive brain science. “We were trying to understand how the human brain interprets visual data after long periods of focus,” explained Cinnamon. “Sporadic visual search theory points to the simple idea that a human has a hard time finding something while staring at a visual scene when the item appears frequently.”
Today, spotting potential problems at checkpoints is largely left to the human eye, an effort that is prone to error. “We should use software and technology to do it better,” Cinnamon said. “In certain situations, we can even automate part of the process by letting the algorithms do the analysis and let human beings agree or disagree. Machines designed with AI can be smaller, nimbler, and more efficient.”
Further, software would allow the system to be trained quickly to spot new and emerging threats as they evolve. Undertaking that type of training with humans is much harder, Cinnamon added.
Today, IED detection equipment is currently a $5.6 billion market globally, according to Small Business Innovation Research (SBIR). Meanwhile, the artificial intelligence and sensing market in applications both within and outside the Department of Defense (DoD) accounts for $7.5 billion in global sales. “. These include automatic target recognition and Identification Friend-or-Foe (IFF), as well as generalized multi-sensor decision support in consumer, environmental, agricultural, medical, and safety applications,” the SBIR report said. “Other commercialization opportunities include navigation and the artificial vision field, which is itself a $21 billion market.”