Automating the processes involved in facial composite production

Abstract

Bringing a criminal to justice is a labour intensive process. In the current paper, we explored ways of reducing police time when constructing and identifying facial composites. In the former, we designed and evaluated a standalone version of the EvoFIT composite system. This was found to perform similarly to the full system that normally requires several hours of a police
officer’s time. In the latter, we built a small database of composites that could be used to search for matching identities. It was found that pixel intensity (texture) information was valuable for composites produced from a traditional feature-based system, but feature shape information for composites produced from the recognition-based EvoFIT. The results show promise for the automated construction and identification of facial composites.