Contextual Analysis for CBRN Clutter Alarm Mitigation and Lead Generation

Physical Science Inc. (PSI) proposes to develop advanced computer vision algorithms to automatically extract relevant contextual information from video captured during chemical, biological, radiological, nuclear and counter explosive (CBRN) detections. PSI produces the PERM-OWL and PERM-Mobile systems which provide a gamma and neutron radiation sensor suite, synchronized contextual video, and state-of-the-art algorithms for source detection, identification, localization, and attribution. Ongoing development to integrate chemical and biological sensors is expected to complete in the next 1-2 years. The proposed Contextual Object Real-time Analysis for Lead generation (CORAL) algorithms will analyze attributed source of CBRN and counter explosive materials including people, vehicles, and business to extract relevant information for alert prioritization and adjudication. This capability will significantly reduce the operational burden of deploying CBRN sensors. The majority of sources detected by these sensors are caused by normal background variability, including commercial and industrial activity associated with well-labeled vehicles and businesses. By automatically extracting this information and correlating with other law enforcement datasets, many alerts can be deprioritized or potentially automatically adjudicated. CORAL algorithms will be directly integrated with operationally deployed PERM-Mobile systems to solicit operator feedback and guide capability refinement.