No abstract

Since 2008, intelligence units of six states of the western part of Switzerland have been sharing a common database for the analysis of high volume crimes. On a daily basis, events reported to the police are analysed, filtered and classified to detect crime repetitions and interpret the crime environment. Several forensic outcomes are integrated in the system such as matches of traces with persons, and links between scenes detected by the comparison of forensic case data. Systematic procedures have been settled to integrate links assumed mainly through DNA profiles, shoemarks patterns and images. A statistical outlook on a retrospective dataset of series from 2009 to 2011 of the database informs for instance on the number of repetition detected or confirmed and increased by forensic case data. Time needed to obtain forensic intelligence in regard with the type of marks treated, is seen as a critical issue. Furthermore, the underlying integration process of forensic intelligence into the crime intelligence database raised several difficulties in regards of the acquisition of data and the models used in the forensic databases. Solutions found and adopted operational procedures are described and discussed. This process form the basis to many other researches aimed at developing forensic intelligence models.

No abstract

The MBioID initiative has been set up to address the following germane question: What and how biometric technologies could be deployed in identity documents in the foreseeable future? This research effort proposes to look at current and future practices and systems of establishing and using biometric identity documents (IDs) and evaluate their effectiveness in large-scale developments.The first objective of the MBioID project is to present a review document establishing the current state-of-the-art related to the use of multimodal biometrics in an IDs application. This research report gives the main definitions, properties and the framework of use related to biometrics, an overview of the main standards developed in the biometric industry and standardisation organisations to ensure interoperability, as well as some of the legal framework and the issues associated to biometrics such as privacy and personal data protection. The state-of-the-art in terms of technological development is also summarised for a range of single biometric modalities (2D and 3D face, fingerprint, iris, on-line signature and speech), chosen according to ICAO recommendations and availabilities, and for various multimodal approaches. This paper gives a summary of the main elements of that report.The second objective of the MBioID project is to propose relevant acquisition and evaluation protocols for a large-scale deployment of biometric IDs. Combined with the protocols, a multimodal database will be acquired in a realistic way, in order to be as close as possible to a real biometric IDs deployment. In this paper, the issues and solutions related to the acquisition setup are briefly presented. #

In this article, we compare aural and automatic speaker recognition in the context of forensic analyses, using a Bayesian framework for the interpretation of evidence.We use perceptual tests performed by non-experts and compare their performance with that of an automatic speaker recognition system. These experiments are performed with 90 phonetically untrained subjects. Several forensic cases were simulated, using the Polyphone IPSC-02 database, varying in linguistic content and technical conditions of recording. We estimate the strength of evidence for both humans and the baseline automatic system, calculating likelihood ratios using perceptual scores for humans and log-likelihood scores for the automatic system. A methodology analogous to the Bayesian interpretation in forensic automatic speaker recognition is applied to the perceptual scores given by humans in order to estimate the strength of evidence. The degradation of the accuracy of human recognition in mismatched recording conditions is contrasted with that of the automatic system under similar recording conditions. The conditions considered are fixed telephone, cellular telephone and noisy speech in forensically realistic conditions. The perceptual cues that the human subjects use to perceive differences in voices are studied, along with their importance in different recording conditions. We observe that while automatic speaker recognition shows higher accuracy in matched conditions of training and testing, its performance degrades significantly in mismatched conditions. Aural recognition accuracy is also observed to degrade from matched conditions to mismatched conditions and in mismatched conditions, the baseline automatic systems showed comparable or slightly degraded performance compared to the aural recognition systems. The baseline automatic system with adaptation to noisy conditions showed comparable or better performance than aural recognition. The higher level perceptual cues used by human listeners in order to recognise speakers are discussed. We also discuss the possibility of increasing the accuracy of automatic systems using the perceptual cues that remain robust to mismatched recording conditions.