Creation of a public corpus of contact-less acquired latent fingerprints without privacy implications

Abstract

Data sets of biometric or forensic samples are an important basis for evaluations and research. Especially biometric data is considered as personal data, which is protected by privacy regulations. Since the data cannot be altered or revoked, at least in some countries, this poses a challenge because rights must be granted to the data's subject. In particular in Germany and probably in the entire European Union after its reformation of the data protection legislation it is challenging to use such data. Furthermore, with respect to latent fingerprints only very few public data sets exist nowadays. We propose the creation of a public data set without privacy implications consisting of latent fingerprints from artificial fingerprint patterns. On the foundation of a first set of 50 fingerprints on a compact disk surface we report challenges that need to be solved in order to create realistic samples. © IFIP International Federation for Information Processing 2013.