The objective of the EU project „FAST and efficient international disaster victim IDentification“ (FASTID) is to establish a new international database to improve the identification of victims after natural disasters with respect to efficiency and speed.
It is spearheaded by INTERPOL and involves the project partners Bundeskriminalamt, Crabbe Consulting, Fraunhofer, Plass Data and the University Dundee.
Within the work package Image Retrieval, the Fraunhofer Institute IOSB develops an image based disaster victim identification module which supports the identification of persons based on images of body modifications (e.g. tattoos, scars, etc). Previous procedures are based upon a manual classification into a fixed set of categories which is often subjective, time-consuming and sometimes error-prone.
By using state-of-the-art computer-based image retrieval techniques, this search can be carried out faster, more accurate and more detailed as images are no longer regarded as match if they are assigned to the same category (e.g. “dragon”) but only if they show the same dragon tattoo.
Fig.1: Comparing the visual details of tattoos can assist the identification of persons after a disaster.
Images:© Poulsons Photography - Fotolia.com