The Department of Homeland Security (DHS) Science and Technology Directorate (S&T) has launched a groundbreaking dataset of 250 synthetic carry-on baggage images to accelerate the development of advanced airport screening systems. This initiative directly addresses the long-standing challenge of limited access to high-quality data, which has hindered innovation in creating faster and more accurate baggage screening technologies.
The new dataset supports presidential executive orders aimed at enhancing public safety and improving trade and travel. Pedro Allende, DHS Under Secretary for Science and Technology, emphasized the importance of this development, saying, "Detection algorithms are key to reducing false alarms that cause hands-on bag inspections and longer checkpoint wait times. Enabling rapid algorithm development from the best innovators not only will increase detection accuracy at the checkpoint but also improve the traveler experience." Dr. John Fortune, Screening at Speed Program Manager, added that current access to curated bag images for algorithm training is limited and often requires a prohibitive vetting process.
Developed by Cignal LLC of Reedsville, Pennsylvania, for S&T’s Screening at Speed Program, the dataset comprises computed tomography (CT) X-ray images that accurately replicate scans from the Transportation Security Administration's latest detection systems. These images are available upon request from Cignal’s website in Digital Imaging and Communication in Medicine (DICOM) format, a standard familiar to algorithm developers. Future plans include releasing larger, more complex bag sets and synthetic millimeter wave images to train passenger screening systems, further advancing homeland security capabilities.