Application of Detection and Recognition Algorithms to Persistent Wide Area Surveillance

Abstract

The persistent airborne surveillance of large geographical areas is now a viable proposition. As well as providing cues to moving objects, it presents new opportunities for understanding the behaviours and motivations of people, both individually and collectively. Exploitation of these huge collections of imagery (a facet of the Big Data challenge) requires more effective tools to derive and abstract useful information to cue the analyst. This paper describes a new system which brings together a number of techniques: moving target detection; tracking; recognition and photogrammetry, to address wide area surveillance problems. We provide a first report on the demands this places on component parts and interfaces. Significantly, we adopt international interoperability standards, particularly with regard to video metadata, to constrain the solution space. We also describe new performance improvements to the video moving target indication and photogrammetry algorithms as well as analysing for the first time the performance of our integrated target model matching capability in our automated system.