Video Utility for Likelihood-based Change ANalysis (VULCAN)

Metron proposes to design, develop, test and demonstrate algorithms for automatically detecting and tracking human activity from low contrast, monochrome imagery, such as from airborne mid-wave infrared (MWIR) cameras, in which the human objects of interest may be as small as a single pixel. There are three principle components: (1) image rectification and registration algorithms to align consecutive frames from a moving camera, (2) algorithms for extracting contacts consistent with humans from the imagery, and (3) detection and tracking algorithms that fuse these contacts over time to produce tracks. These algorithms have the potential to provide dramatic improvements in speed and size of problem (both geographic area and number of active tracks) with respect to detecting and tracking dismounted humans. BENEFIT: There are a number of potential commercial, military and government opportunities for technology that can detect and track humans given low resolution, low contrast mid-wave infrared imagery. In theater, this can be used to provide force protection inside and around the Forward Operating Bases, and to augment the analyses of social networks and relationships between and within insurgent organizations. For Homeland Security, this technology could also play an essential role in border security where the area to be covered far exceeds the available manpower and resources to patrol it. For commercial use, this technology could help to augment security at a large industrial facility, such as a set of buildings and factories on a large compound.