Dr Richard Guest
Richard Guest is Reader of Biometric Systems Engineering and Deputy Head of School.
He received a BEng (Hons) in Computer Science from the University of York in 1995. This was followed by a PhD in Electronic Engineering from the University of Kent in 2000. His research work is in the area of biometric technologies, examining aspects of systems deployment and algorithm development, usability, standardisation, sample quality and conformance. His work has also examined the use of human identification/verification mechanisms within automated processes.
He is a member of the UK Government’s Biometric and Forensic Ethics Committee and a core member of the Kent Interdisciplinary Research Centre in Cyber Security. He is also the Chair of the Training and Education Committee of the European Association of Biometrics (EAB) and a Fellow of the British Computer Society. He has had significant involvement with biometrics standards development as UK Principal Expert to ISO/IEC JTC1 SC37.
He has attracted over £4M over the past 10 years from external sources of funding including EPSRC, EU, ESRC, Leverhulme and industry as PI of 20 separate research grants. He has published over 125 peer reviewed articles and his research featured in the Government Select Committee Report into the Future of Biometrics and also in the Government Chief Scientist’s Annual Report.
He currently the Project Coordinator for the AMBER EU Marie Skłodowska-Curie ITN in Mobile Biometrics (2017-2020) and is Kent PI on the EPSRC Hummingbird Project (2018-2019). He is also a member of the Kent academic team for the PriMa EU Marie Skłodowska-Curie ITN (2020-2023).
Richard Guest has extensive research experience in the areas of image processing and pattern recognition specialising in biometric and forensic applications, His current research work is exploring the relationship between human and machine decision making processes, and issue concerning the use of mobile platforms for biometric authentication.
Fellow of the British Computer Society
Showing 50 of 142 total publications in the Kent Academic Repository. View all publications.
Blanco-Gonzalo, R., Miguel-Hurtado, O., Lunerti, C., Guest, R., Corsetti, B., Ellavarason, E. and Sanchez-Reillo, R. (2019). Biometric Systems Interaction Assessment: The State of the Art. IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS [Online]. Available at: https://doi.org/10.1109/THMS.2019.2913672.The design and implementation of effective and efficient biometric systems presents a series of challenges to information technology (IT) designers to ensure robust performance. One of the most important factors across biometric systems, aside from algorithmic matching ability, is the human interaction influence on performance. Changes in biometric system paradigms have motivated further testing methods, especially within mobile environments, where the interaction with the device has fewer environmental constraints, whichmay severely affect system performance. Testing methods involve the need for reflecting on the influence of user-system interaction on the overall system performance in order to provide information for design and testing. This paper reflects on the state of the art of biometric systems interaction assessment, leading to a comprehensive document of the relevant research and standards in this area. Furthermore, the current challenges are discussed and thus we provide a roadmap for the future of biometrics systems interaction research.
Yanushkevich, S., Sundberg, K., Twyman, N., Guest, R. and Shmerko, V. (2019). Cognitive Checkpoint: Emerging Technologies for Biometric-Enabled Watchlist Screening. Computers & Security [Online]. Available at: https:/dx./doi.org/10.1016/j.cose.2019.05.002.This paper revisits the problem of individual risk assessment in the layered security model. It contributes to the concept of balancing security and privacy via cognitive-centric machine called an ’e-interviewer’. Cognitive checkpoint is a cyber-physical security frontier in mass-transit hubs that provides an automated screening using all types of identity (attributed, biometric, and biographical) from both physical and virtual worlds. We investigate how the development of the next generation of watchlist for rapid screening impacts a sensitive balancing mechanism between security and privacy. We identify directions of such an impact, trends in watchlist technologies, and propose ways to mitigate the potential risks.
Blanco-Gonzalo, R., Lunerti, C., Sanchez- Reillo, R. and Guest, R. (2018). Biometrics: Accessibility challenge or opportunity?. PlosOne [Online] 13:e0194111. Available at: https://doi.org/10.1371/journal.pone.0194111.Biometric recognition is currently implemented in several authentication contexts, most
recently in mobile devices where it is expected to complement or even replace traditional
authentication modalities such as PIN (Personal Identification Number) or passwords. The
assumed convenience characteristics of biometrics are transparency, reliability and easeof-
use, however, the question of whether biometric recognition is as intuitive and straightforward
to use is open to debate. Can biometric systems make some tasks easier for people
with accessibility concerns? To investigate this question, an accessibility evaluation of a
mobile app was conducted where test subjects withdraw money from a fictitious ATM (Automated
Teller Machine) scenario. The biometric authentication mechanisms used include
face, voice, and fingerprint. Furthermore, we employed traditional modalities of PIN and pattern
in order to check if biometric recognition is indeed a real improvement. The trial test subjects
within this work were people with real-life accessibility concerns. A group of people
without accessibility concerns also participated, providing a baseline performance. Experimental
results are presented concerning performance, HCI (Human-Computer Interaction)
and accessibility, grouped according to category of accessibility concern. Our results reveal
links between individual modalities and user category establishing guidelines for future
accessible biometric products.
Guest, R., Miguel-Hurtado, O., Stevenage, S. and Black, S. (2017). Exploring the relationship between stride, stature and hand size for forensic assessment. Journal of Forensic and Legal Medicine [Online] 52:46-55. Available at: http://dx.doi.org/10.1016/j.jflm.2017.08.006.Forensic evidence often relies on a combination of accurately recorded measurements, estimated measurements from landmark data such as a subject's stature given a known measurement within an image, and inferred data. In this study a novel dataset is used to explore linkages between hand measurements, stature, leg length and stride. These three measurements replicate the type of evidence found in surveillance videos with stride being extracted from an automated gait analysis system. Through correlations and regression modelling, it is possible to generate accurate predictions of stature from hand size, leg length and stride length (and vice versa), and to predict leg and stride length from hand size with, or without, stature as an intermediary variable. The study also shows improved accuracy when a subject's sex is known a-priori. Our method and models indicate the possibility of calculating or checking relationships between a suspect's physical measurements, particularly when only one component is captured as an accurately recorded measurement.
Riggs, C., Cornes, K., Godwin, H., Liversedge, S., Guest, R. and Donnelly, N. (2017). The importance of search strategy for finding targets in open terrain. Cognitive Research: Principles and Implications [Online] 2. Available at: https://doi.org/10.1186/s41235-017-0049-4.A number of real-world search tasks (i.e. police search, detection of improvised explosive devices (IEDs)) require searchers to search exhaustively across open ground. In the present study, we simulated this problem by asking individuals (Experiments 1a and 1b) and dyads (Experiment 2) to search for coin targets pseudo-randomly located in a bounded area of open grassland terrain. In Experiment 1a, accuracy, search time, and the route used to search an area were measured. Participants tended to use an ‘S’-shaped pattern with a common width of search lane. Increased accuracy was associated with slower, but also variable, search speed, though only when participants moved along the length (as opposed to across the width) of the search area. Experiment 1b varied the number of targets available within the bounded search area and in doing so varied target prevalence and density. The results confirmed that the route taken in Experiment 1a generalizes across variations in target prevalence/density. In Experiment 2, accuracy, search time, and the search strategy used by dyads was measured. While dyads were more accurate than individuals, dyads that opted to conduct two independent searches were more accurate than those who opted to split the search space. The implications of these results for individuals and dyads when searching for targets in open space are discussed.
Miguel-Hurtado, O., Guest, R., Stevenage, S., Neil, G. and Black, S. (2016). Comparing Machine Learning Classifiers and Linear/Logistic Regression to Explore the Relationship between Hand Dimensions and Demographic Characteristics. PLOS ONE [Online] 11:e0165521. Available at: http://doi.org/10.1371/journal.pone.0165521.Understanding the relationship between physiological measurements from human subjects and their demographic data is important within both the biometric and forensic domains. In this paper we explore the relationship between measurements of the human hand and a range of demographic features. We assess the ability of linear regression and machine learning classifiers to predict demographics from hand features, thereby providing evidence on both the strength of relationship and the key features underpinning this relationship. Our results show that we are able to predict sex, height, weight and foot size accurately within various data-range bin sizes, with machine learning classification algorithms out-performing linear regression in most situations. In addition, we identify the features used to provide these relationships applicable across multiple applications.
Robertson, J., Guest, R., Elliott, S. and OConnor, K. (2016). A Framework for Biometric and Interaction Performance Assessment of Automated Border Control Processes. IEEE Transactions on Human-Machine Systems [Online] 47:983-993. Available at: http://doi.org/10.1109/THMS.2016.2611822.Automated Border Control (ABC) in airports and land crossings utilize automated technology to verify passenger identity claims. Accuracy, interaction stability, user error, and the need for a harmonized approach to implementation are required. Two models proposed in this paper establish a global path through ABC processes. The first, the generic model, maps separately the enrolment and verification phases of an ABC scenario. This allows a standardization of the process and an exploration of variances and similarities between configurations across implementations. The second, the identity claim process, decomposes the verification phase of the generic model to an enhanced resolution of ABC implementations. Harnessing a human-biometric sensor interaction framework allows the identification and quantification of errors within the system's use, attributing these errors to either system performance or human interaction. Data from a live operational scenario are used to analyze behaviors, which aid in establishing what effect these have on system performance. Utilizing the proposed method will aid already established methods in improving the performance assessment of a system. Through analyzing interactions and possible behavioral scenarios from the live trial, it was observed that 30.96% of interactions included some major user error. Future development using our proposed framework will see technological advances for biometric systems that are able to categorize interaction errors and feedback appropriately.
Guest, R., Brockly, M., Elliott, S. and Scott, J. (2016). An assessment of the usability of biometric signature systems using the human-biometric sensor interaction model’. International Journal of Computer Applications in Technology [Online] 53:336-347. Available at: http://dx.doi.org/10.1504/IJCAT.2016.076810.Signature biometrics is a widely used form of user authentication. As a behavioural biometric, samples have inherent inconsistencies which must be accounted for within an automated system. Performance deterioration of a tuned biometric software system may be caused by an interaction error with a biometric capture device, however, using conventional error metrics, system and user interaction errors are combined, thereby masking the contribution by each element. In this paper we explore the application of the Human-Biometric Sensor Interaction (HBSI) model to signature as an exemplar of a behavioural biometric. Using observational data collected from a range of subjects, our study shows that usability issues can be identified specific to individual capture device technologies. While most interactions are successful, a range of common interaction errors need to be mitigated by design to reduce overall error rates.
Miguel-Hurtado, O., Stevenage, S., Bevan, C. and Guest, R. (2016). Predicting sex as a soft-biometrics from device interaction swipe gestures. Pattern Recognition Letters [Online] 79:44-51. Available at: http://www.dx.doi.org/10.1016/j.patrec.2016.04.024.Touch and multi-touch gestures are becoming the most common way to interact with technology such as smart phones, tablets and other mobile devices. The latest touch-screen input capacities have tremendously increased the quantity and quality of available gesture data, which has led to the exploration of its use in multiple disciplines from psychology to biometrics. Following research studies undertaken in similar modalities such as keystroke and mouse usage biometrics, the present work proposes the use of swipe gesture data for the prediction of soft-biometrics, specifically the user's sex. This paper details the software and protocol used for the data collection, the feature set extracted and subsequent machine learning analysis. Within this analysis, the BestFirst feature selection technique and classification algorithms (naïve Bayes, logistic regression, support vector machine and decision tree) have been tested. The results of this exploratory analysis have confirmed the possibility of sex prediction from the swipe gesture data, obtaining an encouraging 78% accuracy rate using swipe gesture data from two different directions. These results will hopefully encourage further research in this area, where the prediction of soft-biometrics traits from swipe gesture data can play an important role in enhancing the authentication processes based on touch-screen devices.
Stevenage, S. and Guest, R. (2016). Combining Forces: Data fusion across man and machine for biometric analysis. Image and Vision Computing [Online]. Available at: http://doi.org/10.1016/j.imavis.2016.03.012.Through the HUMMINGBIRD framework outlined here,we seek to encourage a novel multidisciplinary approach to biometric analysis with the goal of enhancing both understanding and accuracy of identification.
Liang, Y., Fairhurst, M., Guest, R. and Erbilek, M. (2016). Automatic Handwriting Feature Extraction, Analysis and Visualization in the Context of Digital Palaeography. International Journal of Pattern Recognition and Artificial Intelligence [Online] 30:1653001. Available at: http://doi.org/10.1142/S0218001416530013.Digital palaeography is an emerging research area which aims to introduce digital image processing techniques into palaeographic analysis for the purpose of providing objective quantitative measurements. This paper explores the use of a fully automated handwriting feature extraction, visualization, and analysis system for digital palaeography which bridges the gap between traditional and digital palaeography in terms of the deployment of feature extraction techniques and handwriting metrics. We propose the application of a set of features, more closely related to conventional palaeographic assesment metrics than those commonly adopted in automatic writer identification. These features are emprically tested on two datasets in order to assess their effectiveness for automatic writer identification and aid attribution of individual handwriting characteristics in historical manuscripts. Finally, we introduce tools to support visualization of the extracted features in a comparative way, showing how they can best be exploited in the implementation of a content-based image retrieval (CBIR) system for digital archiving.
Read More: http://www.worldscientific.com/doi/abs/10.1142/S0218001416530013
Robertson, J. and Guest, R. (2015). A feature based comparison of pen and swipe based signature characteristics. Human Movement Science [Online] 43:169-182. Available at: http://doi.org/10.1016/j.humov.2015.06.003.Dynamic Signature Verification (DSV) is a biometric modality that identifies anatomical and behavioral characteristics when an individual signs their name. Conventionally signature data has been captured using pen/tablet apparatus. However, the use of other devices such as the touch-screen tablets has expanded in recent years affording the possibility of assessing biometric interaction on this new technology.
To explore the potential of employing DSV techniques when a user signs or swipes with their finger, we report a study to correlate pen and finger generated features. Investigating the stability and correlation between a set of characteristic features recorded in participant’s signatures and touch-based swipe gestures, a statistical analysis was conducted to assess consistency between capture scenarios.
The results indicate that there is a range of static and dynamic features such as the rate of jerk, size, duration and the distance the pen traveled that can lead to interoperability between these two systems for input methods for use within a potential biometric context. It can be concluded that this data indicates that a general principle is that the same underlying constructional mechanisms are evident.
Morton, A., Reid, W., Buntin, C., Brockly, M., O’Neill, J., Elliott, S. and Guest, R. (2015). Signature forgery and the forger – an assessment of influence on handwritten signature production. IT in Industry 3:54-58.Signatures are widely used as a form of personal authentication. Despite ubiquity in deployment, individual signatures are relatively easy to forge, especially when only the static ‘pictorial’ outcome of the signature is considered at verification time. In this study, we explore opinions on signature usage for verification purposes, and how individuals rate a particular third-party signature in terms of ease of forgeability and their own ability to forge. We examine responses with respect to an individual’s experience of the forgeability/complexity of their own signature. Our study shows that past experience does not generally have an effect on perceived signature complexity nor the perceived effectiveness of an individual to themselves forge a signature. In assessing forgeability, most subjects cite the overall signature complexity and distinguishing features in reaching this decision. Furthermore, our research indicates that individuals typically vary their signature according to the scenario but generally little effort into the production of the signature.
Riggs, C., Cornes, K., Godwin, H., Guest, R. and Donnelly, N. (2015). The Importance of Slow Consistent Movement when Searching for Hard-to-Find Targets in Real-World Visual Search. Journal of Vision [Online] 15:1355. Available at: http://doi.org/10.1167/15.12.1355.Various real-world tasks require careful and exhaustive visual search. For example, searching for forensic evidence or signs of hidden threats (what we call hard-to-find targets). Here, we examine how search accuracy for hard-to-find targets is influenced by search behaviour. Participants searched for coins set amongst a 5m x 15m (defined as x and y axes respectively) piece of grassland. The grassland contained natural distractors of leaves and flowers and was not manicured. Coins were visually detectable from standing height. There was no time limit to the task and participants were instructed to search until they were confident they had completed their search. On average, participants detected 45% (SD=23%) of the targets and took 7:23 (SD=4:44) minutes to complete their search. Participants' movement over space and time was recorded as a series of time-stamped x, y coordinates using a Total Station theodolite. To quantify their search behaviour, the x- and y-coordinates of participants' physical locations as they searched the grassland were converted into the frequency domain using a Fourier transform. Decreases in dominant frequencies, a measure of the time before turning during search, resulted in increased response accuracy as well as increased search times. Furthermore, decreases in the number of iterations, defined by the total search time divided by the dominant frequency, also resulted in increased accuracy and search times. Comparing distance between the two most dominant frequency peaks provided a measure of consistency of movement over time. This measure showed that more variable search was associated with slower search times but no improvement in accuracy. Throughout our analyses, these results were true for the y-axis but not the x-axis. At least with respect to the present task, accurate search for hard-to-find targets is dependent on conducting search at a slow consistent speed where changes in direction are minimised. Meeting abstract presented at VSS 2015.
Tabatabaey-Mashadi, N., Sudirman, R., Guest, R. and Khalid, P. (2014). Analyses of pupils’ polygonal shape drawing strategy with respect to handwriting performance. Pattern Analysis and Applications [Online]:1-16. Available at: http://doi.org/10.1007/s10044-014-0423-5.Polygonal shape drawing tasks are commonly used in psychological, clinical and standard handwriting tests to evaluate children’s development. Early detection of physical/mental disorders within subjects therefore requires objective analysis of the drawing tasks. This analysis would help to identify specific rehabilitation needs and accurate detection of disorders. Herein, the aim is to determine the correlation between the performance of polygonal shape drawing and levels in handwriting performance. In the reported experimentation two groups of participants aged between 6 and 7 were studied. The first group was identified by educational experts as being below-average writers within their age group whilst the second group was age-matched controls of average and above. Subjects were required to draw an isosceles triangle within a novel computer-based framework founded on a pen-based graphic tablet capture device. Subsequently, a sequential feature vector containing performance values relating to the order in which they drew the triangle was extracted from tablet data and compared against one another when presented in constructional strategy models. Statistical analyses and automated classification were applied to sequences to infer handwriting level based on the triangle drawing strategy. From our experiments drawing strategies showed significant differences in drawing end-point position, number of strokes used, and the frequency of particular drawing strategies amongst average and below-average handwriting groups. Additionally, a support vector machine classifier was used to detect group membership based on the triangle drawing strategy. From this exemplar polygonal shape drawing study it is revealed that there are details in children’s drawing strategy which considerably differs in grouping based on handwriting performance.
Guest, R., Hurtado, O. and Henniger, O. (2014). Assessment of methods for image recreation for signature time-series data. IET Biometrics [Online] 3:159-166. Available at: http://dx.doi.org/10.1049/iet-bmt.2013.0022.Human signatures are widely used for biometric authentication. For automatic online signature verification, rather than storing an image of the completed signature, data are represented in the form of a time series of pen position and status information allowing the extraction of temporal-based features. For visualisation purposes, signature images need to be recreated from time-series data. In this study, the authors investigate the accuracy and verification performance of a series of interpolation methods for recreating a signature image from the time-series data contained in two ISO/IEC data storage formats. The authors experiments investigate dynamic data stored at various sample rates and signature images recreated at differing resolutions. Their study indicates possible best practice in terms of image recreation method, recreated image resolution and temporal sample rate and assesses the effect on the accuracy of reconstructed signature data.
Liang, Y., Guest, R., Fairhurst, M., Heutte, L., Nicolas, S., Burnett, A. and Palfray, T. (2014). EMMEL: a framework for historical manuscript analysis and presentation. Universal Access in the Information Society [online] 13:147-160. Available at: http://dx.doi.org/10.1007/s10209-013-0298-z.In this paper, a generic framework for historical manuscript image and data processing, visualisation and analysis is introduced with a focus on the modelling of manuscript metadata underpinning the interaction. The goal of such a framework is to capture the requirements from three types of activities involving historical manuscripts: presentation, management and analysis. In addition to an overall text-based description of an historical manuscript, a central requirement of such a framework is to associate rich media information (e.g. video, flash component, etc.) to the manuscript or a specific region of the manuscript. A second requirement is to enable interchange of the manuscript data as well as the attached information between users. As a result of an extensive analysis of requirements collected across a wide range of target user groups, an XML-based metadata language derived from a relational database model is proposed to form an historical document data model, and a prototype system is developed to demonstrate some of the advanced functionalities enabled by this data model. Thus, the proposed framework provides an important tool in promoting access to historical documents on a wide and diverse basis, embracing the fundamental principles of universal access to a shared cultural heritage.
Eastwood, S., Lai, K., Yanushkevich, S., Guest, R. and Shmerko, V. (2018). Technology Gap Navigator: Emerging Design of Biometric-Enabled Risk Assessment Machines. In: BROSIG 2018: Proceedings of the 17th International Conference of the Biometrics Special Interest Group. IEEE. Available at: http://dx.doi.org/10.23919/BIOSIG.2018.8553056.This paper reports the Technology Gap (TG) navigator, a novel tool for individual risk assessment in the layered security infrastructure. It is motivated by the practical need of the biometricenabled security systems design. The tool helps specify the conditions for bridging the identified TGs. The input data for the TG navigator includes 1) a causal description of the TG, 2) statistics regarding the available resources and performances, and 3) the required performance. The output includes generated probabilistic conditions, and the corresponding technology requirements for bridging the targeted TG.
Brockly, M., Elliott, S., Guest, R. and Blanco Gonzalo, R. (2015). Human Biometric Sensor Interaction. In: Encyclopedia of Biometrics, 2nd Edition. Springer. Available at: http://doi.org/10.1007/978-3-642-27733-7_2261-3.
Henniger, O., Guest, R., Miguel-Hurtado, O. and Kaplan, C. (2015). Signature/Sign Time Series Data:Standardization. In: Encyclopedia of Biometrics. Springer US, pp. 1-9. Available at: http://doi.org/10.1007/978-3-642-27733-7_9125-2.
Conference or workshop item
Lunerti, C., Guest, R., Baker, J., Fernandez-Lopez, P. and Sanchez-Reillo, R. (2018). Sensing Movement on Smartphone Devices to Assess User Interaction for Face Verification. In: IEEE ICCST 2018, Montreal, Canada. IEEE. Available at: http://dx.doi.org/10.1109/CCST.2018.8585547.Unlocking and protecting smartphone devices has
become easier with the introduction of biometric face verification,
as it has the promise of a secure and quick authentication solution
to prevent unauthorised access. However, there are still many
challenges for this biometric modality in a mobile context, where
the user’s posture and capture device are not constrained. This
research proposes a method to assess user interaction by analysing
sensor data collected in the background of smartphone devices
during verification sample capture. From accelerometer data, we
have extracted magnitude variations and angular acceleration for
pitch, roll, and yaw (angles around the x-axis, y-axis, and z-axis of
the smartphone respectively) as features to describe the amplitude
and number of movements during a facial image capture process.
Results obtained from this experiment demonstrate that it can be
possible to ensure good sample quality and high biometric
performance by applying an appropriate threshold that will
regulate the amplitude on variations of the smartphone
movements during facial image capture. Moreover, the results
suggest that better quality images are obtained when users spend
more time positioning the smartphone before taking an image.
Blanco Gonzalo, R., Corsetti, B., Goicoechea-Telleria, I., Husseis, A., Liu-Jimenez, J., Sanchez-Reillo, R., Eglitis, T., Ellavarason, E., Guest, R., Lunerti, C., Azimi, M., Khiarak, J., Ezennaya-Gomez, S., Whiskerd, N., Kuzu, R. and Okoh, E. (2018). Attacking a smartphone biometric fingerprint system:a novice’s approach. In: IEEE ICCST 2018, Montreal, Canada. IEEE. Available at: http://dx.doi/10.1109/CCST.2018.8585726.Biometric systems on mobile devices are an
increasingly ubiquitous method for identity verification. The
majority of contemporary devices have an embedded fingerprint
sensor which may be used for a variety of transactions including
unlock a device or sanction a payment. In this study we explore
how easy it is to successfully attack a fingerprint system using a
fake finger manufactured from commonly available materials.
Importantly our attackers were novices to producing the fingers
and were also constrained by time. Our study shows the relative
ease that modern devices can be attacked and the material
combinations that lead to these attacks.
Tolosana, R., Vera-Rodriguez, R., Guest, R., Fierrez, J. and Ortega-Garcia, J. (2018). Complexity-based Biometric Signature Verification. In: 14th IAPR International Conference on Document Analysis and Recognition. IEEE. Available at: https://doi.org/10.1109/ICDAR.2017.40.On-line signature verification systems are mainly based on two approaches: feature- or time functions-based systems (a.k.a. global and local systems). However, new sources of information can be also considered in order to complement these traditional approaches, reduce the intra-class variability and achieve more robust signature verification systems against forgers. In this paper we focus on the use of the concept of complexity in on-line signature verification systems. The main contributions of the present work are: 1) classification of users according to the complexity level of their signatures using features extracted from the Sigma LogNormal writing generation model, and 2) a new architecture for signature verification exploiting signature complexity that results in highly improved performance. Our proposed approach is tested considering the BiosecurID on-line signature database with a total of 400 users. Results of 5.8% FRR for a FAR = 5.0% have been achieved against skilled forgeries outperforming recent related works. In addition, an analysis of the optimal time functions for each complexity level is performed providing practical insights for the application of signature verification in real scenarios.
Lunerti, C., Guest, R., Blanco-Gonzalo, R., Sanchez-Reillo, R. and Baker, J. (2017). Environmental Effects on Face Recognition in Smartphones. In: 51st IEEE International Carnahan Conference on Security Technology. Institute of Electrical and Electronics Engineers.Face recognition is convenient for user authentication on smartphones as it offers several advantages suitable for mobile environments. There is no need to remember a numeric code or password or carry tokens. Face verification allows the unlocking of the smartphone, pay bills or check emails through looking at the smartphone. However, devices mobility also introduces a lot of factors that may influence the biometric performance mainly regarding interaction and environment. Scenarios can vary significantly as there is no control of the surroundings. Noise can be caused by other people appearing on the background, by different illumination conditions, by different users’ poses and through many other reasons. User-interaction with biometric systems is fundamental: bad experiences may derive to unwillingness to use the technology. But how does the environment influence the quality of facial images? And does it influence the user experience with face recognition? In order to answer these questions, our research investigates the user-biometric system interaction from a non-traditional point of view: we recreate reallife scenarios to test which factors influence the image quality in face recognition and, quantifiably, to what extent. Results indicate the variability in face recognition performance when varying environmental conditions using smartphones.
Alsedais, R. and Guest, R. (2017). Person re-identification from CCTV silhouettes using Generic Fourier Descriptors. In: 51st IEEE International Carnahan Conference on Security Technology. Institute of Electrical and Electronics Engineers.Person re-identification in public areas (such as airports, train stations and shopping malls) has recently received increased attention from computer vision researchers due, in part, to the demand for enhanced levels of security. Reidentifying subjects within non-overlapped camera networks can be considered as a challenging task. Illumination changes in different scenes, variations in camera resolutions, field of view and human natural motion are the key obstacles to accurate implementation. This study assesses the use of Generic Fourier Shape Descriptor (GFD) on person silhouettes for reidentification and further established which sections of a subject’s silhouette is able to deliver optimum performance. Human silhouettes of 90 subjects from the CASIA dataset walking 0° and 90° to a fixed CCTV camera were used for the purpose of re-identification. Each subject’s video sequence comprised between 10 and 50 frames. For both views, silhouettes were segmented into eight algorithmically defined areas: head and neck, shoulders, upper 50%, lower 50%, upper 15%, middle 35%, lower 40% and whole body. A GFD was used independently on each segment at each angle. After extracting the GFD feature for each frame, a linear discriminant analysis (LDA) classifier was used to investigate re-identification accuracy rate, where 50% of each subject’s frames were training and the other 50% were testing. The results show that 97% identification accuracy rate at the 10th rank is achieved by using GFD on the upper 50% segment of the human silhouette front (0°) side. From 90° images, using GFD on the upper 15% silhouette segment was almost 98% accuracy rate at the 10th rank. This study illustrates which segments
Miguel-Hurtado, O., Guest, R. and Lunerti, C. (2017). Voice and face interaction evaluation of a mobile authentication platform. In: 51st IEEE International Carnahan Conference on Security Technology. Institute of Electrical and Electronics Engineers.Biometric authentication in mobile devices has become a key aspect of application security. However, the use of dedicated sensors such as fingerprint/iris sensors may not always be feasible. As an alternative, the use of face and voice biometrics using the generic sensors integrated in smartphones is gaining momentum. This work applied the HBSI framework to analyise the user’s interaction with the mobile PIDaaS platform that integrates voice and face authentication. Our analysis enables a thorough comparison between the user’s interaction for these two modalities with the same population.
Miguel-Hurtado, O., Guest, R., Blanco-Gonzolo, R. and Lunerti, C. (2017). Interaction evaluation of a mobile voice authentication system. In: IEEE International Carnahan Conference on Security Technology. IEEE. Available at: https://doi.org/10.1109/CCST.2016.7815697.Biometric recognition is nowadays widely used in smartphones, making the users' authentication easier and more transparent than PIN codes or patterns. Starting from this idea, the EU project PIDaaS aims to create a secure authentication system through mobile devices based on voice and face recognition as two of the most reliable and user-accepted modalities. This work introduces the project and the first PIDaaS usability evaluation carried out by means of the well-known HBSI model In this experiment, participants interact with a mobile device using the PIDaaS system under laboratory conditions: video recorded and assisted by an operator. Our findings suggest variability among sessions in terms of usability and feed the next PIDaaS HCI design.
Guest, R., Miguel-Hurtado, O. and Chatzisterkotis, T. (2017). A New Approach to Automatic Signature Complexity Assessment. In: IEEE International Carnahan Conference on Security Technology. IEEE. Available at: https://doi.org/10.1109/CCST.2016.7815678.Understanding signature complexity has been shown to be a crucial facet for both forensic and biometric appbcations. The signature complexity can be defined as the difficulty that forgers have when imitating the dynamics (constructional aspects) of other users signatures. Knowledge of complexity along with others facets such stability and signature length can lead to more robust and secure automatic signature verification systems. The work presented in this paper investigates the creation of a novel mathematical model for the automatic assessment of the signature complexity, analysing a wider set of dynamic signature features and also incorporating a new layer of detail, investigating the complexity of individual signature strokes. To demonstrate the effectiveness of the model this work will attempt to reproduce the signature complexity assessment made by experienced FDEs on a dataset of 150 signature samples.
Brockly, M., Elliott, S., Guest, R. and Proctor, R. (2017). The development of a test harness for biometric data collection and validation. In: IEEE International Carnahan Conference on Security Technology. IEEE. Available at: https://doi.org/10.1109/CCST.2016.7815696.Biometric test reports are an important tool in the evaluation of biometric systems, and therefore the data entered into the system needs to be of the highest integrity. Data collection, especially across multiple modalities, can be a challenging experience for test administrators. They have to ensure that the data are collected properly, the test subjects are treated appropriately, and the test plan is followed. Tests become more complex as the number of sensors are increased, and therefore it becomes increasingly important that a test harness be developed to improve the accuracy of the data collection. This paper describes the development of a test harness for a complex multi-sensor, multi-visit data collection, and explains the processes for the development of such a harness. The applicability of such a software package for the broader biometric community is also considered.
Guest, R. and Miguel-Hurtado, O. (2016). User-Interaction Evaluation of a Mobile Authentication System. In: EAB Research Projects Conference (EAB-RPC) 2016.
Miguel-Hurtado, O. and Guest, R. (2016). Users-Centric Design: introducing remote usability evaluation in mobile implementations. In: International Biometric Performance Testing Conference 2016,.
Elliot, S. and Guest, R. (2016). Human Biometric Sensor Interaction, latest research and Process HBSI. In: International Biometric Performance Testing Conference 2016,.
Guest, R. and Hurtado, O. (2015). Mobile Biometric Usability Assessment within PIDaaS. In: EAB Research Projects Conference (EAB-RPC) 2015.
Morton, A., Reid, W., Buntin, C., Brockly, M., O’Neill, J., Elliot, S. and Guest, R. (2015). Signature forgery and the forger – an assessment of influence on handwritten signature production. In: Proceedings ICITA 2015. Available at: http://www.icita.org/2015/abstracts/us-morton.htm.Signatures are widely used as a form of personal authentication. Despite ubiquity in deployment, individual signatures are relatively easy to forge, especially when only the static ‘pictorial’ outcome of the signature is considered at verification time. In this study we explore opinions on signature usage for verification purposes, and how individuals rate a particular third-party signature in terms of ease of forgeability and their own ability to forge. We examine responses with respect to an individual’s experience of the forgeability/complexity of their own signature. Our study shows that past experience does not generally have an effect on perceived signature complexity nor the perceived effectiveness of an individual to themselves forge a signature. In assessing forgeability, most subjects cite overall signature complexity and distinguishing features in reaching this decision. Furthermore, our research indicates that individuals typically vary their signature according to the scenario but generally little effort into the production of the signature.
Felipe, L., De Oliveira, B. and Guest, R. (2015). An Assessment of Dynamic Signature Forgery Creation Methodology and Accuracy. In: 17th International Graphonomics Society Conference.
Riggs, C., Cornes, K., Godwin, H., Guest, R. and Donnelly, N. (2015). The Importance of Slow Consistent Movement when Searching for Hard-to-Find Targets in Real-World Visual Search. In: Proc: Vision Sciences Society, 15th Annual Meeting.
Elliott, S., O’Connor, K., Bartlow, E., Robertson, J. and Guest, R. (2015). Expanding the human-biometric sensor interaction model to identify claim scenarios. In: International Conference of Identity, Security and Behaviour Analysis.
Elliott, S., O’Connor, K., Bartlow, E., Robertson, J. and Guest, R. (2015). Expanding the human-biometric sensor interaction model to identity claim scenarios. In: 2015 IEEE International Conference on Identity, Security and Behavior Analysis (ISBA). pp. 1-6. Available at: http://doi.org/10.1109/ISBA.2015.7126362.Biometric technologies represent a significant component of comprehensive digital identity solutions, and play an important role in crucial security tasks. These technologies support identification and authentication of individuals based on their physiological and behavioral characteristics. This has led many governmental agencies to choose biometrics as a supplement to existing identification schemes, most prominently ID cards and passports. Studies have shown that the success of biometric systems relies, in part, on how humans interact and accept such systems. In this paper, the authors build on previous work related to the Human-Biometric Sensor Interaction (HBSI) model and examine it with respect to the introduction of a token (e.g. an electronic passport or identity card) into the biometric system. The role of the imposter within an Identity Claim scenario has been integrated to expand the HBSI model into a full version, which is able to categorise potential False Claims and Attack Presentations.
Guest, R., Hurtado, O., Stevenage, S., Neil, G. and Black, S. (2014). Biometrics within the superidentity project: a new approach to spanning multiple identity domains. In: IEEE International Carnahan Conference on Security Technology (ICCST).
Brockly, M., Elliott, S., Burdine, J., Frost, M., Riedle, M. and Guest, R. (2014). An investigation into biometric signature capture device performance and user acceptance. In: International Carnahan Conference OnSecurity Technology (ICCST), 2014. IEEE, pp. 1-5. Available at: http://dx.doi.org/10.1109/CCST.2014.6986970.The human signature provides a natural and publically-accepted legally-admissible method for providing authentication to a process. Automatic biometric signature systems assess both the drawn image and the temporal aspects of signature construction, providing enhanced verification rates over and above conventional outcome assessment. To enable the capture of these constructional data requires the use of specialist `tablet' devices. In this paper we explore the enrolment performance using a range of common signature capture devices and investigate the reasons behind user preference. The results show that writing feedback and familiarity with conventional `paper and pen' donation configurations are the primary motivation for user preference. These results inform the choice of signature device from both technical performance and user acceptance viewpoints.
Hurtado, O., Guest, R., Stevenage, S. and Neil, G. (2014). The relationship between handwritten signature production and personality traits. In: International Joint Conference on Biometrics.
Datasets / databases
Miguel-Hurtado, O., Guest, R., Stevenage, S., Neil, G. and Black, S. (2016). Hand images and lengths dataset: Comparing Machine Learning Classifiers and Linear/Logistic Regression to Explore the Relationship between Hand Dimensions and Demographic Characteristics. [Website]. Available at: http://dx.doi.org/10.5281/zenodo.17487.The zip contains right and left hand geometry images from 112 participants. The images were captured using a Nikon D200 SLR camera (format: jpg, size: 3504x2336 pixels), with both the palm of the hand and camera facing downwards. Participants placed each hand on an acetate sheet with a series of positioning pegs.
The excel contains a series of length measurements (based on the underlying skeleton of the hand) manually extracted (see Figure 1 for details) along with demographic information from the participants: sex (male or female), height (in cm), weight (in kg) and foot size (in UK sizes).
Guest, R. (2016). ISO/IEC JTC1 SC37 30106-3:OO BioAPI Part 3:2016. International Standards Organization.
Guest, R. (2014). Information Technology -- Biometric Data Interchange Formats -- 19794-Part 7: Signature/Sign Time Series Data - Second Edition. ISO/IEC.ISO/IEC 19794-7:2014 specifies data interchange formats for signature/sign behavioural data captured in the form of a multi-dimensional time series using devices such as digitizing tablets or advanced pen systems. The data interchange formats are generic, in that they may be applied and used in a wide range of application areas where handwritten signs or signatures are involved. No application-specific requirements or features are addressed in ISO/IEC 19794-7:2014.
Guest, R. (2014). 19794-7/PDAM 1 Information technology -- Biometric data interchange formats -- Part 7: Signature/sign time series data -- Amendment 1: XML encoding. [International Standard].
Research report (external)
House of Commons, S. (2015). Current and Future Uses of Biometric Data and Technologies. Sixth Report of Session 2014-15: Report, Together With Formal Minutes Relating to the Report. [Online]. House of Commons London: The Stationery Office Limited. Available at: http://www.publications.parliament.uk/pa/cm201415/cmselect/cmsctech/734/734.pdf.
Robertson, J. (2017). The Application of the Human-Biometric Sensor Interaction Method to Automated Border Control Systems.Biometrics components are used in many different systems and technologies to verify that the user is whom they say they are. In Automated Border Control systems, biometrics components used in conjunction with a traveller's documents to make sure the user is whom they say they are so that they can cross into a countries borders. The systems are expected to verify the identity with a higher degree than officers who manually check travellers.
Each year the number of travellers crossing through a country borders increases and so systems are expected to handle bigger demands; through improving the user experience to ensuring accuracy and performance standards increase.
While the system does bring its benefits through increased speed and higher security, there are drawbacks. One of the main issues with the systems is a lack of standardisation across implementations. Passing through an automated process at Heathrow may be different to Hong Kong. The infrastructure, information, environment and guidance given during the transaction will all greatly differ for the user. Furthermore, the individual components and subsequent processing will be evaluated using a different methodology too.
This thesis reports on the contrasts between implementations, looking at solutions which utilise different biometric modalities and travel documents. Several models are devised to establish a process map which can be applied to all systems. Investigating further, a framework is described for a novel assessment method to evaluate the performance of a system. An RGB-D sensor is implemented, to track and locate the user within an interactive environment. By doing so, the user's interaction is assessed in real-time. Studies then report on the effectiveness of the solution within a replicated border control scenario. Several relationships are studied to improve the technologies used within the scenario. Successful implementation of the automated assessment method may improve the user's experience with systems, improving information and guidance, increasing the likelihood of successful interaction while maintaining a high level of security and quicker processing times.
Chatzisterkotis, T. (2015). An Examination of Quantitative Methods for Forensic Signature Analysis and the Admissibility of Signature Verification System As Legal Evidence.The experiments described in this thesis deal with handwriting characteristics which are involved in the production of forged and genuine signatures and complexity of signatures. The objectives of this study were (1) to provide su?cient details on which of the signature characteristics are easier to forge, (2) to investigate the capabilities of the signature complexity formula given by Found et al. based on a different signature database provided by University of Kent. This database includes the writing movements of 10 writers producing their genuine signature and of 140 writers forging these sample signatures. Using the 150 genuine signatures without constrictions of the Kent’s database an evaluation of the complexity formula suggested in Found et al took place divided the signature in three categories low, medium and high graphical complexity. The results of the formula implementation were compared with the opinions of three leading professional forensic document examiners employed by Key Forensics in the UK.
The analysis of data for Study I reveals that there is not ample evidence that high quality forgeries are possible after training. In addition, a closer view of the kinematics of the forging writers is responsible for our main conclusion, that forged signatures are widely different from genuine especially in the kinematic domain. From all the parameters used in this study 11 out of 15 experienced significant changes when the comparison of the two groups (genuine versus forged signature) took place and gave a clear picture of which parameters can assist forensic document examiners and can be used by them to examine the signatures forgeries. The movements of the majority of forgers are signi?cantly slower than those of authentic writers. It is also clearly recognizable that the majority of forgers perform higher levels of pressure when trying to forge the genuine signature. The results of Study II although limited and not entirely consistent with the study of Found that proposed this model, indicate that the model can provide valuable objective evidence (regarding complex signatures) in the forensic environment and justify its further investigation but more work is need to be done in order to use this type of models in the court of law. The model was able to predict correctly only 53% of the FDEs opinion regarding the complexity of the signatures.
Apart from the above investigations in this study there will be also a reference at the debate which has started in recent years that is challenging the validity of forensic handwriting experts’ skills and at the effort which has begun by interested parties of this sector to validate and standardise the field of forensic handwriting examination and a discussion started. This effort reveals that forensic document analysis field meets all factors which were set by Daubert ruling in terms of theory proven, education, training, certification, falsifiability, error rate, peer review and publication, general acceptance. However innovative methods are needed for the development of forensic document analysis discipline. Most modern and effective solution in order to prevent observational and emotional bias would be the development of an automated handwriting or signature analysis system. This system will have many advantages in real cases scenario. In addition the significant role of computer-assisted handwriting analysis in the daily work of forensic document examiners (FDE) or the judicial system is in agreement with the assessment of the National Research Council of United States that “the scientific basis for handwriting comparison needs to be strengthened”, however it seems that further research is required in order to be able these systems to reach the accomplishment point of this objective and overcome legal obstacles presented in this study.
Diaz, M., Ferrer, M., Ramalingam, S. and Guest, R. (2019). Investigating the Common Authorship of Signatures by Off-line Automatic Signature Verification without the Use of Reference Signatures. IEEE Transactions on Information Forensics & Security [Online]. Available at: https://doi.org/10.1109/TIFS.2019.2924195.
Diaz, M., Ferrer, M., Guest, R. and Pal, U. (2019). Graphomotor Evolution in the Handwriting of Bengali Children Through Sigma-Lognormal Based-Parameters: A Preliminary Study. In: 19th International Graphonomics Conference.