Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Bücher
Zeitschriften
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
MTZ-Fachkongress

Antriebe und Energiesysteme von morgen


Austausch von Automobil- und Energiewirtschaft

Am 14./15. Mai geht es in Chemnitz um die Mobilitätswende durch Elektro- und Wasserstofffahrzeuge.
Erweiterte Suche
Anmelden
nach oben

2009 | Buch

Statistical Evaluation of Biometric Evidence in Forensic Automatic Speaker Recognition Kapitel lesen Erstes Kapitel lesen

Computational Forensics

Third International Workshop, IWCF 2009, The Hague, The Netherlands, August 13-14, 2009. Proceedings

herausgegeben von: Zeno J. M. H. Geradts, Katrin Y. Franke, Cor J. Veenman

Verlag: Springer Berlin Heidelberg

Buchreihe : Lecture Notes in Computer Science

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

This book constitutes the refereed proceedings of the Third International Workshop, IWCF 2009, held in The Hague, The Netherlands, August 13-14, 2009. The 16 revised full papers presented were carefully reviewed and are organized in topical sections on speech and linguistics, fingerprints, handwriting, documents, printers, multimedia and visualization. This volume is interesting to researchers and professionals who deal with forensic problems using computational methods. Its primary goal is the discovery and advancement of forensic knowledge involving modeling, computer simulation, and computer-based analysis and recognition in studying and solving forensic problems.

Inhaltsverzeichnis

Frontmatter

Speech and Linguistics

Statistical Evaluation of Biometric Evidence in Forensic Automatic Speaker Recognition
Abstract
Forensic speaker recognition is the process of determining if a specific individual (suspected speaker) is the source of a questioned voice recording (trace). This paper aims at presenting forensic automatic speaker recognition (FASR) methods that provide a coherent way of quantifying and presenting recorded voice as biometric evidence. In such methods, the biometric evidence consists of the quantified degree of similarity between speaker-dependent features extracted from the trace and speaker-dependent features extracted from recorded speech of a suspect. The interpretation of recorded voice as evidence in the forensic context presents particular challenges, including within-speaker (within-source) variability and between-speakers (between-sources) variability. Consequently, FASR methods must provide a statistical evaluation which gives the court an indication of the strength of the evidence given the estimated within-source and between-sources variabilities. This paper reports on the first ENFSI evaluation campaign through a fake case, organized by the Netherlands Forensic Institute (NFI), as an example, where an automatic method using the Gaussian mixture models (GMMs) and the Bayesian interpretation (BI) framework were implemented for the forensic speaker recognition task.
Andrzej Drygajlo
Forensic Authorship Attribution Using Compression Distances to Prototypes
Abstract
In several situations authors prefer to hide their identity. In forensic applications, one can think of extortion and threats in emails and forum messages. These types of messages can easily be adjusted, such that meta data referring to names and addresses is at least unreliable. In this paper, we propose a method to identify authors of short informal messages solely based on the text content. The method uses compression distances between texts as features. Using these features a supervised classifier is learned on a training set of known authors. For the experiments, we prepared a dataset from Dutch newsgroup texts. We compared several state-of-the-art methods to our proposed method for the identification of messages from up to 50 authors. Our method clearly outperformed the other methods. In 65% of the cases the author could be correctly identified, while in 88% of the cases the true author was in the top 5 of the produced ranked list.
Maarten Lambers, Cor J. Veenman

Printers

Estimation of Inkjet Printer Spur Gear Teeth Number from Pitch Data String of Limited Length
Abstract
In this paper, we investigate the feasibility of estimating the number of inkjet printer spur gear teeth from pitch data strings of limited length by maximum entropy spectral analysis. The purpose of this study is to improve the efficiency of inkjet printer model identification based on spur mark comparison in the field of forensic document analysis. Experiments were performed using two spur gears in different color inkjet printer models, and eight different lengths of pitch data strings—ranging from three to 10 times the number of spur gear teeth. The result for a data string longer than five times the number of teeth showed a proper estimation within a deviation of one tooth. However, the estimation failed for shorter data strings because the order in maximum entropy analysis was determined inappropriately. The presented results provide information on the number of spur gear teeth from shorter data strings than in a previous study.
Yoshinori Akao, Atsushi Yamamoto, Yoshiyasu Higashikawa
Detecting the Spur Marks of Ink-Jet Printed Documents Using a Multiband Scanner in NIR Mode and Image Restoration
Abstract
Ink-jet printers are frequently used in crime such as counterfeiting bank notes, driving licenses, and identification cards. Police investigators required us to identify makers or brands of ink-jet printers from counterfeits. In such demands, classifying ink-jet printers according to spur marks which were made by spur gears located in front of print heads for paper feed has been addressed by document examiners. However, spur marks are significantly faint so that it is difficult to detect them. In this study, we propose the new method for detecting spur marks using a multiband scanner in near infrared (NIR) mode and estimations of point spread function (PSF). As estimating PSF we used cepstrum which is inverse Fourier transform of logarithm spectrum. The proposed method provided the clear image of the spur marks.
Takeshi Furukawa

Fingerprints

A Computational Discriminability Analysis on Twin Fingerprints
Abstract
Sharing similar genetic traits makes the investigation of twins an important study in forensics and biometrics. Fingerprints are one of the most commonly found types of forensic evidence. The similarity between twins’ prints is critical establish to the reliability of fingerprint identification. We present a quantitative analysis of the discriminability of twin fingerprints on a new data set (227 pairs of identical twins and fraternal twins) recently collected from a twin population using both level 1 and level 2 features. Although the patterns of minutiae among twins are more similar than in the general population, the similarity of fingerprints of twins is significantly different from that between genuine prints of the same finger. Twins fingerprints are discriminable with a 1.5%~1.7% higher EER than non-twins. And identical twins can be distinguished by examine fingerprint with a slightly higher error rate than fraternal twins.
Yu Liu, Sargur N. Srihari
Probability of Random Correspondence for Fingerprints
Abstract
Individuality of fingerprints can be quantified by computing the probabilistic metrics for measuring the degree of fingerprint individuality. In this paper, we present a novel individuality evaluation approach to estimate the probability of random correspondence (PRC). Three generative models are developed respectively to represent the distribution of fingerprint features: ridge flow, minutiae and minutiae together with ridge points. A mathematical model that computes the PRCs are derived based on the generative models. Three metrics are discussed in this paper: (i) PRC of two samples, (ii) PRC among a random set of n samples (nPRC) and (iii) PRC between a specific sample among n others (specific nPRC). Experimental results show that the theoretical estimates of fingerprint individuality using our model consistently follow the empirical values based on the NIST4 database.
Chang Su, Sargur N. Srihari

Visualisation

A New Computational Methodology for the Construction of Forensic, Facial Composites
Abstract
A facial composite generated from an eyewitness’s memory often constitutes the first and only means available for police forces to identify a criminal suspect. To date, commercial computerised systems for constructing facial composites have relied almost exclusively on a feature-based, ‘cut-andpaste’ method whose effectiveness has been fundamentally limited by both the witness’s limited ability to recall and verbalise facial features and by the large dimensionality of the search space. We outline a radically new approach to composite generation which combines a parametric, statistical model of facial appearance with a computational search algorithm based on interactive, evolutionary principles. We describe the fundamental principles on which the new system has been constructed, outline recent innovations in the computational search procedure and also report on the real-world experience of UK police forces who have been using a commercial version of the system.
Christopher Solomon, Stuart Gibson, Matthew Maylin
Geovisualization Approaches for Spatio-temporal Crime Scene Analysis – Towards 4D Crime Mapping
Abstract
This paper presents a set of methods and techniques for analysis and multidimensional visualisation of crime scenes in a German city. As a first step the approach implies spatio-temporal analysis of crime scenes. Against this background a GIS-based application is developed that facilitates discovering initial trends in spatio-temporal crime scene distributions even for a GIS untrained user. Based on these results further spatio-temporal analysis is conducted to detect variations of certain hotspots in space and time. In a next step these findings of crime scene analysis are integrated into a geovirtual environment. Behind this background the concept of the space-time cube is adopted to allow for visual analysis of repeat burglary victimisation. Since these procedures require incorporating temporal elements into virtual 3D environments, basic methods for 4D crime scene visualisation are outlined in this paper.
Markus Wolff, Hartmut Asche

Multimedia

Multimedia Forensics Is Not Computer Forensics
Abstract
The recent popularity of research on topics of multimedia forensics justifies reflections on the definition of the field. This paper devises an ontology that structures forensic disciplines by their primary domain of evidence. In this sense, both multimedia forensics and computer forensics belong to the class of digital forensics, but they differ notably in the underlying observer model that defines the forensic investigator’s view on (parts of) reality, which itself is not fully cognizable. Important consequences on the reliability of probative facts emerge with regard to available counter-forensic techniques: while perfect concealment of traces is possible for computer forensics, this level of certainty cannot be expected for manipulations of sensor data. We cite concrete examples and refer to established techniques to support our arguments.
Rainer Böhme, Felix C. Freiling, Thomas Gloe, Matthias Kirchner
Using Sensor Noise to Identify Low Resolution Compressed Videos from YouTube
Abstract
The Photo Response Non-Uniformity acts as a digital fingerprint that can be used to identify image sensors. This characteristic has been used in previous research to identify scanners, digital photo cameras and digital video cameras. In this paper we use a wavelet filter from Lukáš et al [1] to extract the PRNU patterns from multiply compressed low resolution video files originating from webcameras after they have been uploaded to YouTube. The video files were recorded with various resolutions, and the resulting video files were encoded with different codecs. Depending on video characteristics (e.g. codec quality settings, recording resolution), it is possible to correctly identify cameras based on these videos.
Wiger van Houten, Zeno Geradts
Using the ENF Criterion for Determining the Time of Recording of Short Digital Audio Recordings
Abstract
The Electric Network Frequency (ENF) Criterion is a recently developed forensic technique for determining the time of recording of digital audio recordings, by matching the ENF pattern from a questioned recording with an ENF pattern database. In this paper we discuss its inherent limitations in the case of short – i.e., less than 10 minutes in duration – digital audio recordings. We also present a matching procedure based on the correlation coefficient, as a more robust alternative to squared error matching.
Maarten Huijbregtse, Zeno Geradts

Handwriting

A Machine Learning Approach to Off-Line Signature Verification Using Bayesian Inference
Abstract
A machine learning approach to off-line signature verification is presented. The prior distributions are determined from genuine and forged signatures of several individuals. The task of signature verification is a problem of determining genuine-class membership of a questioned (test) signature. We take a 3-step, writer independent approach: 1) Determine the prior parameter distributions for means of both “genuine vs. genuine” and “forgery vs. known” classes using a distance metric. 2) Enroll n genuine and m forgery signatures for a particular writer and calculate both the posterior class probabilities for both classes. 3) When evaluating a questioned signature, determine the probabilities for each class and choose the class with bigger probability. By using this approach, performance over other approaches to the same problem is dramatically improved, especially when the number of available signatures for enrollment is small. On the NISDCC dataset, when enrolling 4 genuine signatures, the new method yielded a 12.1% average error rate, a significant improvement over a previously described Bayesian method.
Danjun Pu, Gregory R. Ball, Sargur N. Srihari
Computer-Assisted Handwriting Analysis: Interaction with Legal Issues in U.S. Courts
Abstract
Advances in the development of computer-assisted handwriting analysis have led to the consideration of a computational system by courts in the United States. Computer-assisted handwriting analysis has been introduced in the context of Frye or Daubert hearings conducted to determine the admissibility of handwriting testimony by questioned document examiners, as expert witnesses, in civil and criminal proceedings. This paper provides a comparison of scientific and judicial methods, and examines concerns over reliability of handwriting analysis expressed in judicial decisions. Recently, the National Research Council assessed that “the scientific basis for handwriting comparisons needs to be strengthened”. Recent studies involving computer-assisted handwriting analysis are reviewed in light of the concerns expressed by the judiciary and National Research Council. A future potential role for computer-assisted handwriting analysis in the courts is identified.
Kenneth A. Manning, Sargur N. Srihari
Analysis of Authentic Signatures and Forgeries
Abstract
The paper presents empirical studies of kinematic and kinetic signature characteristics. In contrast to previous studies a more in-depth analysis is performed which reveals insides on differences and similarities of authentic and mimicked signing movements. It is shown that the signing behavior of genuine writers and impostors is only likely to differ in terms of local characteristics. Global characteristics can easily be imitated by skilled forgers. Moreover, it is shown that authentic writing characteristics cover a broad value range that might interfere with value ranges of unsophisticated forgeries. In our experiments signing behavior of 55 authentic writers and of 32 writers mimicking signature samples with three different levels of graphical complexity is studied. We discuss implications for ink-trace characteristics on paper and provide recommendations for implementing computer-based analysis methods.
Katrin Franke

Documents

Automatic Line Orientation Measurement for Questioned Document Examination
Abstract
In questioned document examination many different problems arise: documents can be forged or altered, signatures can be counterfeited, etc. When experts attempt to identify such forgeries manually, they use among others line orientation as a feature. This paper describes an automatic mean for measuring the line justification and helping the specialist to find suspicious lines. The goal is to use this method as one of several screening tools for scanning large document collections for the potential presence of forgeries. This method extracts the text-lines, measures their orientation angle and decides the validity of these measured angles based on previously trained parameters.
Joost van Beusekom, Faisal Shafait, Thomas Breuel
Color Deconvolution and Support Vector Machines
Abstract
Methods for machine learning (support vector machines) and image processing (color deconvolution) are combined in this paper for the purpose of separating colors in images of documents. After determining the background color, samples from the image that are representative of the colors to be separated are mapped to a feature space. Given the clusters of samples of either color the support vector machine (SVM) method is used to find an optimal separating line between the clusters in feature space. Deconvolution image processing parameters are determined from the separating line. A number of examples of applications in forensic casework are presented.
Charles E. H. Berger, Cor J. Veenman
Backmatter
Metadaten
Titel
Computational Forensics
herausgegeben von
Zeno J. M. H. Geradts
Katrin Y. Franke
Cor J. Veenman
Copyright-Jahr
2009
Verlag
Springer Berlin Heidelberg
Electronic ISBN
978-3-642-03521-0
Print ISBN
978-3-642-03520-3
DOI
https://doi.org/10.1007/978-3-642-03521-0

Premium Partner

    Bildnachweise