Ruben Tolosana

Preprocessing and Feature Selection for Improved Sensor Interoperability in Online Biometric Signature Verification

Due to the technological evolution and the increasing popularity of smartphones, people can access an application using authentication based on biometric approaches from many different devices. Device interoperability is a very challenging problem for biometrics, which needs to be further studied. In this paper, we focus on interoperability device compensation for online signature verification since this biometric trait is gaining a significant interest in banking and commercial sector in the last years. The proposed approach is based on two main stages. The first one is a preprocessing stage where data acquired from different devices are processed in order to normalize the signals in similar ranges. The second one is based on feature selection taking into account the device interoperability case, in order to select to select features which are robust in these conditions. This proposed approach has been successfully applied in a similar way to two common system approaches in online signature verification, i.e., a global features-based system and a time functions-based system. Experiments are carried out using Biosecure DS2 (Wacom device) and DS3 (Personal Digital Assistant mobile device) dynamic signature data sets which take into account multisession and two different scenarios emulating real operation conditions. The performance of the proposed global features-based and time functions-based systems applying the two main stages considered in this paper have provided an average relative improvement of performance of 60.3% and 26.5% Equal Error Rate (EER), respectively, for random forgeries cases, compared with baseline systems. Finally, a fusion of the proposed systems has achieved a further significant improvement for the device interoperability problem, especially for skilled forgeries. In this case, the proposed fusion system has achieved an average relative improvement of 27.7% EER compared with the best performance of time functions-based system. These results prove the robustness of the proposed approach and open the door for future works using devices as smartphones or tablets, commonly used nowadays.

E-biosign: stylus- and finger-input multi-device database for dynamic signature recognition

This paper describes the design, acquisition process and a baseline evaluation of e-BioSign, a new database of dynamic signature and handwriting. e-BioSign is comprised of 5 devices in total, three Wacom devices (DTU-500, DTU-530 and STU 1031) specifically designed to capture dynamic signatures and handwriting, and two Samsung general purpose tablets (Samsung Galaxy Note 10.1 and Samsung ATIV). For these two Samsung tablets data is collected using a pen stylus but also the finger to study the performance of signature verification in a mobile scenario. Data was collected in two sessions for 70 subjects, and includes dynamic information of the signature, the full name and number sequences. For signature and the full name skilled forgeries were also performed. A signature baseline evaluation is carried out for a predefined recognition system based on DTW, achieving a benchmark performance for each of the devices. The use of finger for signing achieves good results for the case of random forgeries (less than 1% EER), but the performance is degraded significantly for the case of skilled forgeries compared to the case using the pen stylus.

Feature-based dynamic signature verification under forensic scenarios

Nowadays forensic document examiners (FDE) have to analyse more and more signatures captured by digital devices. While they can still use the static image of the signature, it has been proven that the dynamic information contains very discriminative information. This paper is focused on dynamic signature recognition applied to forensic scenarios. An automatic featured-based or global recognition system is considered as some of the features extracted by these systems could be used by FDE in their work. A system comprised of 117 global features is proposed and evaluated with BioSecure DS2 database. A subset of 40 features is selected by SFFS algorithm as the optimal feature vector in the development phase. Results of 10.6% EER are achieved for skilled forgeries which improve previous results using similar approaches. In addition, a set of selected features have been analysed statistically for genuine and forged signatures in order to obtain useful information that could be used by forensic experts in their reports.

Keystroke Biometrics Ongoing Competition

This paper presents the first Keystroke Biometrics Ongoing Competition (KBOC) organized to establish a reproducible baseline in person authentication using keystroke biometrics. The competition has been developed using the BEAT platform and includes one of the largest keystroke databases publicly available based on a fixed text scenario. The database includes genuine and attacker keystroke sequences from 300 users acquired in four different sessions distributed in a four month time span. The sequences correspond to the users name and surname, and therefore, each user comprises an individual and personal sequence. As baseline for KBOC, we report the results of 31 different algorithms evaluated according to accuracy and robustness. The systems have achieved EERs as low as 5.32% and high robustness to multisession variability with accuracy degradation lower than 1% for probes separated by months. The entire database is publicly available at the competition website.

Signature recognition: establishing human baseline performance via crowdsourcing

This work explores crowdsourcing for the establishment of human baseline performance on signature recognition. We present five experiments according to three different scenarios in which laymen, people without Forensic Document Examiner experience, have to decide about the authenticity of a given signature. The scenarios include single comparisons between one genuine sample and one unlabeled sample based on image, video or time sequences and comparisons with multiple training and test sets. The human performance obtained varies from 7% to 80% depending of the scenario and the results suggest the large potential of these collaborative platforms and encourage to further research on this area.

Optimal feature selection and inter-operability compensation for on-line biometric signature authentication

Due to the technological evolution and the increasing popularity of smartphones, people can access an application with many different devices. This device interoperability is a very challenging problem for biometrics. In this paper we focus on inter-operability device compensation for on-line signature verification. The proposed approach is based on two main stages. The first one is a preprocessing stage where data acquired from different devices are processed in order to normalize the signals in similar ranges. The second one is based on a feature selection of time functions taking into account the inter-operability device comparisons in order to select features which are robust in these conditions. The experimental work has been carried out with Biosecure database using a Wacom tablet (DS2) and a PDA tablet (DS3), and the system developed is based on dynamic time warping (DTW) elastic measure over the selected time functions. The performance of the proposed system is very similar compared to an ideal system. Also, the proposed approach provides average relative improvements for the cases of inter-operability comparisons of 26.5% for random forgeries and, around 14.2% for the case of skilled forgeries comparing the results with the case of having a system specifically tuned for each device, proving the robustness of the proposed approach. These results open the door for future works using devices as smartphones or tablets, commonly used nowadays.

Exploring Recurrent Neural Networks for On-Line Handwritten Signature Biometrics

Systems based on deep neural networks have made a breakthrough in many different pattern recognition tasks. However, the use of these systems with traditional architectures seems not to work properly when the amount of training data is scarce. This is the case of the on-line signature verification task. In this paper, we propose a novel writer-independent on-line signature verification systems based on Recurrent Neural Networks (RNNs) with a Siamese architecture whose goal is to learn a dissimilarity metric from the pairs of signatures. To the best of our knowledge, this is the first time these recurrent Siamese networks are applied to the field of on-line signature verification, which provides our main motivation. We propose both Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) systems with a Siamese architecture. In addition, a bidirectional scheme (which is able to access both past and future context) is considered for both LSTM- and GRU-based systems. An exhaustive analysis of the system performance and also the time consumed during the training process for each recurrent Siamese network is carried out in order to compare the advantages and disadvantages for practical applications. For the experimental work, we use the BiosecurID database comprised of 400 users who contributed a total of 11,200 signatures in four separated acquisition sessions. Results achieved using our proposed recurrent Siamese networks have outperformed the state-of-the-art on-line signature verification systems using the same database.

Benchmarking desktop and mobile handwriting across COTS devices: The e-BioSign biometric database

This paper describes the design, acquisition process and baseline evaluation of the new e-BioSign database, which includes dynamic signature and handwriting information. Data is acquired from 5 different COTS devices: three Wacom devices (STU-500, STU-530 and DTU-1031) specifically designed to capture dynamic signatures and handwriting, and two general purpose tablets (Samsung Galaxy Note 10.1 and Samsung ATIV 7). For the two Samsung tablets, data is collected using both pen stylus and also the finger in order to study the performance of signature verification in a mobile scenario. Data was collected in two sessions for 65 subjects, and includes dynamic information of the signature, the full name and alpha numeric sequences. Skilled forgeries were also performed for signatures and full names. We also report a benchmark evaluation based on e-BioSign for person verification under three different real scenarios: 1) intra-device, 2) inter-device, and 3) mixed writing-tool. We have experimented the proposed benchmark using the main existing approaches for signature verification: feature- and time functions-based. As a result, new insights into the problem of signature biometrics in sensor-interoperable scenarios have been obtained, namely: the importance of specific methods for dealing with device interoperability, and the necessity of a deeper analysis on signatures acquired using the finger as the writing tool. This e-BioSign public database allows the research community to: 1) further analyse and develop signature verification systems in realistic scenarios, and 2) investigate towards a better understanding of the nature of the human handwriting when captured using electronic COTS devices in realistic conditions.

e-BioSign tool: Towards scientific assessment of dynamic signatures under forensic conditions

This paper presents a new tool specifically designed to carry out dynamic signature forensic analysis and give scientific support to forensic handwriting examiners (FHEs). Traditionally FHEs have performed forensic analysis of paper-based signatures for court cases, but with the rapid evolution of the technology, nowadays they are being asked to carry out analysis based on signatures acquired by digitizing tablets more and more often. In some cases, an option followed has been to obtain a paper impression of these signatures and carry out a traditional analysis, but there are many deficiencies in this approach regarding the low spatial resolution of some devices compared to original off-line signatures and also the fact that the dynamic information, which has been proved to be very discriminative by the biometric community, is lost and not taken into account at all. The tool we present in this paper allows the FHEs to carry out a forensic analysis taking into account both the traditional off-line information normally used in paper-based signature analysis, and also the dynamic information of the signatures. Additionally, the tool incorporates two important functionalities, the first is the provision of statistical support to the analysis by including population statistics for genuine and forged signatures for some selected features, and the second is the incorporation of an automatic dynamic signature matcher, from which a likelihood ratio (LR) can be obtained from the matching comparison between the known and questioned signatures under analysis.

Increasing the robustness of biometric templates for dynamic signature biometric systems

Due to the high deployment of devices such as smartphones and tablets and their increasing popularity in our society, the use of biometric traits in commercial and banking applications through these novel devices as an easy, quick and reliable way to perform payments is rapidly increasing. The handwritten signature is one of the most socially accepted biometric traits in these sectors due to the fact that it has been used in financial and legal transitions for centuries. In this paper we focus on dynamic signature verification systems. Nowadays, most of the state-of-the-art systems are based on extracting information contained in the X and Y spatial position coordinates of the signing process, which is stored in the biometric templates. However, it is critical to protect this sensible information of the users signatures against possible external attacks that would allow criminals to perform direct attacks to a biometric system or carry out high quality forgeries of the users signatures. Following this problem, the goal of this work is to study the performance of the system in two cases: first, an optimal time functions-based system taking into account the information related to X and Y coordinates and pressure, which is the common practice (i.e. Standard System). Second, we study an extreme case not considering information related to X, Y coordinates and their derivatives on the biometric system (i.e. Secure System), which would be a much more robust system against attacks, as this critical information would not be stored anywhere. The experimental work is carried out using e-BioSign database which makes use of 5 devices in total. The systems considered in this work are based on Dynamic Time Warping (DTW), an elastic measure over the selected time functions. Sequential Forward Features Selection (SFFS) is applied as a reliable way to obtain an optimal time functions vector over a development subset of users of the database. The results obtained over the evaluation subset of users of the database show a similar performance for both Standard and Secure Systems. Therefore, the use of a Secure System can be useful in some applications such as banking in order to avoid the lost of important user information against possible external attacks.