Synapse AI Makes Airport Security Faster, More SecureBy Dasmer Singh
Synapse AI Makes Airport Security Faster, More Secure
What’s Wrong with Airport Security?
Airport screenings are frighteningly far from perfect today. Scores of tests over the last few years by the Department of Homeland Security have demonstrated that the TSA fails to detect 95% of all contraband (including weapons and drugs) that go through airport security.
Synapse Technology Corporation is solving this problem by applying artificial intelligence to x-ray security screenings, using deep learning algorithms to train computers to identify contraband instead of relying on humans, who tend to be more error prone.
How It Works
Seed investors include Founders Fund, Responder Ventures, Xfund, Mark Pincus, and Daniel Nadler. The funding supports the growing full time team as well as the acquisition of x-ray machines to refine the detection algorithms. Synapse’s team is looking to grow, and is actively hiring Computer Vision and Full Stack engineers to build on top of the existing Syntech ONE™ product.
Syntech ONE™, Synapse’s software, acts as a “second pair of eyes” and alerts security operators when an item in the x-ray machine looks like a sharp object or weapon. The software focuses on the threatening item with a red bounding box, similar to how smartphone cameras detect and focus on faces when taking photos of people. When a computer has identified possible contraband and displays an alert, the operator opens the bag and inspects more carefully. Not only does this software vastly increase the accuracy with which contraband is detected, but it would also increase the speed at which security operators analyze images. This technology would shorten security lines, require fewer operators at every airport, and reduce TSA costs.
Now that Syntech ONE™ is ready, Bruno Faviero is uber-focused on selling it to potential buyers. Customers include the TSA, private security firms, prisons, schools, hotels and just about any other establishment that uses x-ray security machines. Syntech ONE™ is already undergoing the approval process of multiple organizations and will likely appear in various locations around the globe in 2018.
How They Built Synapse
Synapse was founded by three former MIT graduates Bruno Faviero, CEO; Ian Cinnamon, President; and Simanta Gautam, CTO.
“Computers, unlike humans, don’t get tired, bored or distracted. They are not subject to the optimism bias that humans gravitate towards – the fallacy that since nothing dangerous was found in the first 99 bags, there is probably nothing bad in bag 100. They do not require breaks to operate at a high level, and function consistently across locations,” said Cinnamon, a co-founder and current student at the Stanford Graduate School of Business.
While an undergrad at MIT, Cinnamon studied Brain and Cognitive Sciences and worked with Harvard Medical School Professor Jeremy Wolfe to research the human ability of finding targets that appear sporadically, such as needles in a haystack. Their research conclusion was that fatigue and bias cause humans to be fundamentally limited in finding such targets, explaining the TSA’s high error rate.
Where They Are Today
Synapse Technology Corporation is currently headquartered in a Palo Alto, California office. The small team has a proprietary data collection method that provides mass amounts of X-ray images with and without contraband. Simanta Gautam, Synapse’s CTO and Co-Founder who worked at NASA on hyperspectral computer vision, has tagged these x-ray images as a “high threat” or “safe,” and used them to train Synapse’s proprietary deep learning algorithm. The software is ready for commercial use; Synapse states that detection accuracy is well beyond human capabilities.
The idea for Syntech ONE™ came to Cinnamon in early 2016 while traveling between New York and San Francisco on a weekly basis to support the angel investments of Mark Pincus, Founder and Former CEO of Zynga. Tired of the long and inefficient security lines, Cinnamon emailed his previous mentor, Harvard’s Dr. Jeremy Wolfe, about how someone must be working on implementing AI in the TSA space. “Not really,” Wolfe replied. Cinnamon then reached out about applying computer vision to security screenings to the Under Secretary for Science and Technology at the Department of Homeland Security, who replied within minutes to schedule a meeting with Cinnamon.
Synapse’s team is looking to grow, and is actively hiring Computer Vision and Full Stack engineers to build on top of the existing Syntech ONE™ product.
“It is so exciting this research project turned into a developed product that we can bring to the world,” said Cinnamon.