DHS Funds Research into Machine Learning for Airport Security
Monday, August 31, 2020 | Comments

With a growing need to improve the security, efficiency and accuracy of passenger and baggage screening, the Department of Homeland Security (DHS) Small Business Innovation Research (SBIR) Program is working with a small business to advance explosive detection equipment. Synthetik Applied Technologies received funding to develop machine learning training data that simulates human travelers and baggage object models to support machine learning algorithms.

“As threats to our nation’s airports continue to evolve, we are committed to investing in technologies that will improve the security posture of aviation checkpoints while minimizing the inconvenience to passengers,” said William N. Bryan, DHS senior official performing the duties of the under secretary for science and technology. “We look forward to seeing the technology developed through the SBIR program that supports our vision for a passenger screening process that is reliable, less invasive, and efficient.”

The DHS SBIR program, administered by the DHS Science and Technology Directorate (S&T), selected Synthetik, based in Austin, Texas, to participate in second phase of the program, based on the successful demonstration of feasibility in the first phase for their Synthetic Data Training For Explosive Detection Machine Learning Algorithms technology solution.

In the second phase, Synthetik Applied Technologies will continue their efforts to develop synthetic training data that will enhance machine learning object detection algorithms to improve detection and reduce false alarms. For machine learning algorithms to reach their peak performance, they must be trained on a very large amount of data, and collecting and preparing this data is typically an expensive and time-consuming process. Synthetic data generation creates the capability to generate complete, annotated datasets in a matter of minutes without handling dangerous materials or initiating human subjects’ protocols. This technology would streamline the security screening process, creating an improved passenger experience for the traveling public.

“Synthetik’s work will enable DHS S&T’s Screening at Speed Program to generate high-fidelity training data for machine learning algorithms virtually instantaneously and with very little cost,” said Karl Harris, DHS S&T program manager. “This training data will help us develop faster and more accurate algorithms to improve throughput of passenger bags while protecting the health and safety of Transportation Security Administration (TSA) employees and the traveling public.”

The effort started prior to the COVID-19 pandemic but has become even more relevant as social distancing and other protective measures are put into place in order to minimize the exposure and contact between TSA officers and passengers.

At the completion of the 24-month contract for the second phase, SBIR awardees will have received up to $1 million to develop and demonstrate a prototype to facilitate the pursuit of funding for a third phase. For the third phase, SBIR performers seek to secure funding from a private or non-SBIR government source and pursue technology commercialization resulting from their first and second phase efforts.

Would you like to comment on this story? Find our comments system below.

Post a comment
Name: *
Email: *
Title: *
Comment: *


No Comments Submitted Yet

Be the first by using the form above to submit a comment!


August 2021

15 - 18
APCO Conference and Expo
San Antonio, Texas

More Events >

Site Navigation