Materials identification, classification and threat detection using nonlinear analysis of X-ray diffraction data
Airport and transportation security require that a high volume of luggage and packages be assessed for their potential as explosive threats. While existing X-ray diffraction (XRD) technology is capable of accurately determining the material composition of a sample, it can be extremely time consuming due to the required number of photons received at the detectors for a high-quality signal.
This technology enables material classification and automated explosive threat detection with low-quality XRD data achievable with airport and transportation scanners. Using a database of lab-quality XRD measurements of explosive and non-explosive materials commonly screened for, this technology compares the sample under test using physically significant classification features that are robust to noise and lack of training data.
This innovation improves the robustness of material classification to noise and low-resolution data