Dynaxion took a huge step forward when we showed our proof of concept system can accurately identify threats and benign materials non-obtrusively and fully automatically. From standard materials such as toothpaste, shower gel and milk powder to simulants for explosives and drugs, all materials were correctly identified. With this important result, we’re ready to initiate the next phase: building a fully integrated prototype.
For the setup of the Proof of Concept, we used a compact particle accelerator system to generate a beam of neutrons. This beam is directed towards the various samples and using several gamma-ray detectors, we gathered spectral information of the content sample. The data was analyzed by a proprietary artificial intelligence algorithm, which provided a 100% accurate identification of all the samples.
Unique aspects of Dynaxion will disrupt the security scanning market
The unique aspects of the Dynaxion scanning system make it potentially revolutionary in the way security screening of parcels and suitcases is performed. We’ve filed a global patent on the scanning mechanism and the way it combines several data sources (spectral and imaging information). In addition, we’ve developed a proprietary way to efficiently create large amounts of training data.
This data is used by a combination of several deep learning networks we developed that provide detailed information about the content of a scanned object. An important enabler of the system is the cost-effective Radio-Frequency Quadrupole (RFQ) based neutron source combined with a compact shielding system. This will result in an overall commercially viable system with comparable size to current advanced X-ray-based CT systems.