Comanche Counter-Terrorism Networks: Automated Inspection of X-Ray Cargo Images utilizing Machine Learning, MongoDB, and Apache Cassandra

• In Event: Roundtable: AI in Criminal Justice

Abstract

Ecumenically, radiographic imaging has become an integral tool for screening and analysis of cargo containers for potential nuclear or radiological threats. We are investigating methods to extract features from radiographic imaging that, when correlated with other measurements and information, are able to classify and detect a potential threat. Furthermore, the analysis of single-energy radiographs presents impediments by the large variety of cargo contents and the overall volume and mass of standard intermodel shipping containers. Our approach is to extract key features that characterize the contents in myriads of ways and then apply machine learning to these features along with feature vectors from other measurements and contextual information. These other features may include spatial profiles, gamma-ray spectra and neutron emissions from radioactive materials, weight, volume, location, condition, origin, shipper, destination, stated contents etc. We report here on analysis of 669 radiographic scans measured by a SAIC VACIS (Vehicle and Cargo Imaging System) at the Port of Oakland. We use Variational Bayesian Factor Analysis (VBFA) methods, machine learning algorithms with Sobel Edge Detection to create features that separate images by overall category which are analyzed and stored in the Apache Cassandra NoSQL Database(s) and the Hadoop Ecosystem.

Author

• Wilbert McClay, Meharry Medical College / University of New Haven Department of Forensics