Matching images captured from unmanned aerial vehicle

Police database cannot have images of first-time offenders; hence, apprehending them becomes a very challenging task. In this paper, we propose a novel technique to apprehend first-time offenders using composite sketches and images captured by unmanned aerial vehicles. The key contribution of this paper is we have developed a new technology to match composite sketches with images captured by unmanned aerial vehicle to apprehend first-time criminals in a very short time period. The unmanned aerial vehicle is sent in the area where the first-time offender is likely to be present. The image captured by unmanned aerial vehicle is passed to face detection module so that only human faces are obtained. Feature extraction is performed using multi-resolution uniform local binary pattern, and classification is performed using dictionary matching. This proposed method is validated by composite sketches generated using SketchCop FACETTE face design system software and images captured by Phantom 3 professional unmanned aerial vehicle.