Interest in the exploitation of soft biometrics information has continued to develop over the last decade or so. In comparison with traditional biometrics, which focuses principally on person identification, the idea of soft biometrics processing is to study the utilisation of more general information regarding a system user, which is not necessarily unique. There are increasing indications that this type of data will have great value in providing complementary information for user authentication. However, the authors have also seen a growing interest in broadening the predictive capabilities of biometric data, encompassing both easily definable characteristics such as subject age and, most recently, ‘higher level’ characteristics such as emotional or mental states. This study will present a selective review of the predictive capabilities, in the widest sense, of biometric data processing, providing an analysis of the key issues still adequately to be addressed if this concept of predictive biometrics is to be fully exploited in the future.

Ocular biometrics refers to the use of features of the eye for person recognition. For instance, the unique and stable texture of the iris has been recognised as a powerful ocular biometric characteristic. In this study, the authors propose to improve biometric authentication with a multimodal ocular biometric system based on the iris pattern and the three-dimensional shape of the cornea. They show how the cornea can be used as a biometric trait for person recognition and then, they propose an intra-ocular fusion with iris features to improve the overall performance of the system. Feature extraction was done by modelling the shape of the cornea with a Zernike polynomial expansion. Then the best linear combinations of Zernike coefficients were found with linear discriminant analysis and used as biometric identifier. The iris texture was analysed with a typical methodology using Gabor filtering and phase encoding. The fusion was performed at the matching score level using min, max, sum and weighted-sum rule. The experimental results on a new database constructed for this bi-modal study showed impressive performance of the proposed ocular biometric system with equal error rate decreasing to 0% with the weighted-sum rule.

Footsteps, as a main kind of behavioural trait are a universally available signal. However, it remains a challenging problem to construct a robust footstep authentication system. Since, in practise, footsteps are usually accompanied with noisy and environmental sounds, it is difficult to extract stable footstep features from the mixed sounds. This study describes a novel robust footstep identification system. To extract stable features from the footsteps mixed with noisy or environmental sounds, a robust acoustic local feature extraction method is proposed. In the proposed method, the main frequency components of footsteps are determined, and then their local distributions and variations in time–frequency domain are obtained and regarded as the acoustic local features. These local features are robust to white noise, pink noise and invariant to the intensity of the footsteps. However, the conventional pattern recognition methods are not suitable for them due to that these local features are observably different from the frequently used acoustic features, and so the authors introduce a Bayesian decision classifier to implement footstep identification. Theoretical and experimental results demonstrate that this system is relatively robust to white noise, pink noise, environmental sounds and yields a better classification performance compared with the existing methods.

Identification based on signature verification is one of the most popular biometric methods which were used even prior to the advent of computers. In the field of dynamic signature verification, signatures' time functions (e.g. pen tip velocity, acceleration and pressure) are analysed in addition to static appearance of signatures. Dynamic feature stability (DFS) experiment is a process for finding the most stable signature partitions which are difficult to forge. The experiment can most effectively lead to a focus on the signature trajectories. Due to the different angles related to the signature pad which are regarded as major problem in signature verification algorithms, in this study, radon transform is used to transform rotation effect to shift effect. Convolutional neural network disregards the precise location of image features, as well as shift effects in both axes of image that is decline by its nature. According to DFS experiment, three independent recognition paths are structured and their effects on final classification are determined by the experiment. Three Persian datasets are analysed by DFS, which led to a reliable trait of Persian signatures. As a result, the least verification error is attained. Besides, SVC2004, as an international benchmark is evaluated by the proposed algorithm.

This study describes the generation process of a synthetic fingerprint using the Petri nets. The presented Petri net simulates this process, in general. Then, there are presented various initial markings which simulate a particular real-life scenario. The proposed Petri net could be used to make the creation process more understandable and also it gives some space for optimisation and parallelisation.

Person authentication is an important part to protect individual privacy in the informational society. With the development of electroencephalogram (EEG), it gradually becomes feasible using EEG signals to identify person recognition. However, the analysis of EEG signals is complex, unstable and non-linear. With this fact, non-linear analysis such as entropy would be more appropriate. In this study, four types of entropies are used to extract EEG signals features for the purpose of person authentication, and the performance of person authentication based on different entropies is compared. In this study, self-face and non-self-face images are used to induce EEG signals for the authentication process. Eventually, the average accuracy of 16 subjects by jackknife test was 90.7%, which demonstrating its better authentication performance and the proposed method achieving higher performance compared with previous methods of EEG-based person authentication. The results also show that, though the four types of entropies were used as the feature extraction methods, the fuzzy entropy achieved the best performance for person authentication.

A very common case for law enforcement is recognition of suspects from a long distance or in a crowd. This is an important application for low-resolution face recognition (in the authors' case, face region below 40 × 40 pixels in size). Normally, high-resolution images of the suspects are used as references, which will lead to a resolution mismatch of the target and reference images since the target images are usually taken at a long distance and are of low resolution. Most existing methods that are designed to match high-resolution images cannot handle low-resolution probes well. In this study, they propose a novel method especially designed to compare low-resolution images with high-resolution ones, which is based on the log-likelihood ratio (LLR). In addition, they demonstrate the difference in recognition performance between real low-resolution images and images down-sampled from high-resolution ones. Misalignment is one of the most important issues in low-resolution face recognition. Two approaches – matching-score-based registration and extended training of images with various alignments – are introduced to handle the alignment problem. Their experiments on real low-resolution face databases show that their methods outperform the state-of-the-art.

This study addresses automatic prediction of the writer's gender. We propose the use of fuzzy integral (FI) operators to combine support vector machines (SVMs) associated with different local features. Presently, we focus on local histogram-based features that describe different kinds of handwriting traits to ensure SVM complementarity. First, we introduce a new feature based on the histogram of templates that aims to highlight local orientations of the text strokes. As a second feature, we propose the rotation invariant uniform local binary patterns to enhance local textural information, whereas the third feature is the gradient local binary patterns. Various forms of the FI are used for combining these predictors. Experiments are conducted on four standard datasets of English, Arabic and French handwritten text. First, for each language, the prediction task is evaluated by considering text-independent and writer-independent design. Then, a more challenging prediction is tried by adding the language-independency constraint. The results obtained confirm the effectiveness of the proposed features. Also, they highlight the contribution of the combination step to achieve a robust prediction.

This study presents a new method for enhancing fingerprint image. The process of the enhancement is divided into two phases: fingerprint is first enhanced using Gabor filtering and then the enhanced fingerprint can be further enhanced by using sparse representation with the priori information of ridge pattern based on classification dictionaries learning. In the second stage, first, the orientations of fingerprint patches are estimated by the weighted linear projection analysis and the quality of patches are evaluated by the coherence of point orientations. Second, the training patches are classified into eight groups based on their own orientations, and the training samples of each class are selected from candidate patches by their own quality. The corresponding classification dictionaries are learned in frequency domain. Finally, the fingerprint image is enhanced based on spectra diffusion by using classification dictionaries learning. The experiments are carried out using various fingerprint enhancement methods. The experiments show that the proposed method achieves better results in comparison with other methods, and can significantly improve the performance of automatic fingerprint identification system.

Security measures have to be taken to protect the privacy of biometric data. Cancellable biometrics is proposed as an effective mechanism of using and protecting biometrics. The authors propose a new approach of constructing cancellable fingerprint template by mapping real minutiae to randomly generated minutiae in a synthetic template. The synthetic minutiae are selected based on the k-nearest neighbour method. One synthetic neighbour of each real minutia is utilised to construct the verification template (VT). Since the synthetic template is determined by a user-specific PIN and a random salt, the proposed method is in fact a two-factor authentication scheme. Multiple VTs can be generated easily by applying different PINs and salts to a real template. To prove the validity of the scheme, testing is carried out on three databases. A few factors affecting matching are also investigated. The results show that the constructed templates satisfy the requirements of cancellable biometrics. False non-match rate and False matching rate with the transformed templates can be much lower than those with the original templates by properly selecting the size of the synthetic templates and the ordinal number of the nearest neighbours. The proposed approach can also be utilised for multi-generation template transformation.

The increasing demand on biometric authentication systems (BASs) has brought the need of secure and privacy-preserving solutions accepted by a wider community of users. The decision makers pay a great attention to how people react to BASs and their opinions about the features and procedures of the system. In this work, a generic Biometric Technology Acceptance Model (BioTAM) is proposed. BioTAM encounters trust as an objective measure of privacy-security tradeoff, public willingness and user confidence. BioTAM takes into account social and human factors which prominently affect the wider dissemination and easy penetration of BASs. To scrutinise people's behavioural intention to use a BAS, BioTAM melds traditional Technology Acceptance Model constructs and the trust model offered as a new construct. In order to inspire stakeholders on how BioTAM can be used to assess a BAS, a sample case study is investigated.

The authors present a novel face tracking approach where optical flow information is incorporated into a modified version of the Viola–Jones detection algorithm. In the original algorithm, detection is static, as information from previous frames is not considered; in addition, candidate windows have to pass all stages of the classification cascade, otherwise they are discarded as containing no face. In contrast, the proposed tracker preserves information about the number of classification stages passed by each window. Such information is used to build a likelihood map, which represents the probability of having a face located at that position. Tracking capabilities are provided by extrapolating the position of the likelihood map to the next frame by optical flow computation. The proposed algorithm works in real time on a standard laptop. The system is verified on the Boston Head Tracking Database, showing that the proposed algorithm outperforms the standard Viola–Jones detector in terms of detection rate and stability of the output bounding box, as well as including the capability to deal with occlusions. The authors also evaluate two recently published face detectors based on convolutional networks and deformable part models with their algorithm showing a comparable accuracy at a fraction of the computation time.

Facial ageing modelling has been an active research topic in the field of anthropology. Considering the fact that ageing is a non-uniform and a non-linear process for different face types (e.g. origins, gender etc.), dealing with a reliable face-ageing model may considerably help investigators working in some specific fields such as forensics. Unlike numerous studies dealing with forward or predictive face models, in this study, the authors propose a backward model aiming at estimating childhood face images using their corresponding adult face appearance as an input. For the proposed approach, face contour and different components are modified non-linearly, based on an estimated geometrical model. On the other hand, the face texture is estimated by mapping a reference face texture to the estimated geometrical model. This approach will show that it will be possible to ‘digitally’ rejuvenate an adult person's face down to it being 3–4 years old. For evaluation purposes, a database has been created from 112 subjects. Results have been evaluated using both objective (face recognition system) and subjective (human perception) criteria. The most promising and interesting results will be highlighted further ahead.

Few facial image datasets are suitable for forensic research. In this study, the authors present ForenFace, a facial image and video dataset. It contains video sequences and extracted images of 97 subjects recorded with six different surveillance camera of various types. Moreover, it also contains high-resolution images and 3D scans. The novelty of this dataset lies in two aspects: (i) a subset of 435 images (87 subjects, five images per subject) has been manually annotated, yielding a very rich forensically relevant annotation of almost 19.000 facial parts, and (ii) making available a toolset to create, view, and extract the annotation. The authors present protocols and the result of a baseline experiment in which two commercial software packages and an annotated facial feature contained in this dataset are compared. The dataset, the annotation and tools are available under a usage license.

The performances of the bag-of-words approach in biometric applications using electrocardiography (ECG) signals have been analysed according to the influence of specific design parameters. Optimal setup scenarios have been identified combining five encoding procedures, two pooling methods, and three classification strategies. The method does not require waveform segmentation nor fiducial points detection. Comparative results based on extensive experiments conducted on real ECG recordings collected on chest, finger, and hand palm are presented. Sparse representations yield best results, exceeding 99% correct classification rate for a number of 100 subjects, while additionally exhibiting robustness against modifications of the experimental setup.