Nowadays, several biometric databases already contain millions of entries of individuals. With an increasing number of enrolled individuals, the response time of queries grows and can become critical. Fingerprint indexing offers a set of techniques to reduce the workload of entries, which have to be compared thoroughly. This work surveys research on such techniques. It focuses on the features of fingerprints, which are used as input. This survey also provides an assessment of the quality of the body of research in this field. Deficiencies herein are identified, e.g. there is a lack of common datasets and metrics used for testing.

Gait recognition has emerged as an attractive biometric technology for the identification of people by analysing the way they walk. However, one of the main challenges of the technology is to address the effects of inherent various intra-class variations caused by covariate factors such as clothing, carrying conditions, and view angle that adversely affect the recognition performance. The main aim of this survey is to provide a comprehensive overview of existing robust gait recognition methods. This is intended to provide researchers with state of the art approaches in order to help advance the research topic through an understanding of basic taxonomies, comparisons, and summaries of the state-of-the-art performances on several widely used gait recognition datasets.

The historical NEXUS iris kiosks log dataset collected by the Canada Border Services Agency from 2003 to 2014 has become the focus of scientific attention due to its involvement in the iris ageing debate between the National Institute of Standard and Technology and the University of Notre Dame researchers. To facilitate this debate, this study provides additional details on how this dataset was collected, its various properties and irregularities, and presents new results related to the effect of ageing, age, and other factors on the system performance obtained using the portions of the dataset that have not been previously analysed. In doing that, the importance of conducting subject-based performance analysis, as opposed to the traditionally done transaction-based analysis, is emphasised. The significance of factor effects is examined. Recommendations on further improvement of the technology are made.

Finger texture (FT) is currently attracting significant attention in the area of human recognition. FT covers the area between the lower knuckle of the finger and the upper phalanx before the fingerprint. It involves rich features which can be efficiently used as a biometric characteristic. In this study, the authors contribute to this growing area by proposing a new verification approach, i.e. deep FT learning. To the best of the authors’ knowledge, this is the first time that deep learning is employed for recognising people by using the FT characteristic. Four databases have been used to evaluate the proposed method: the Hong Kong Polytechnic University Contact-free 3D/2D (PolyU2D), Indian Institute of Technology Delhi (IITD), CASIA Blue spectral (CASIA-BLU) corresponding to spectral 460 nm and CASIA White spectral (CASIA-WHT) from the CASIA Multi-Spectral images database. The obtained results have shown superior performance compared with recent literature. The verification accuracies have attained 100, 98.65, 100 and 98% for the four databases of PolyU2D, IITD, CASIA-BLU and CASIA-WHT, respectively.

Sparse representation classification (SRC) is one of the popular methods of classification in biometrics, in which the decision of class for the test sample was based on the class with minimum reconstruction error. As SRC is based on the sparsity of the images, a decision based on reconstruction error is not ideal. In this study, an efficient classification methodology for finger vein recognition called mutual SRC (MSRC) is proposed. MSRC classifies the test sample by a new decision rule which significantly improves the recognition rate of the conventional SRC. By this new decision rule, the classification of the test sample is not only based on the nearest sparse neighbour but also on determining the training sample which considers the test sample as its nearest neighbour (NN). In this method, the training set is selected based on reconstruction error for the test sample, then which training sample considers the test sample as its NN is identified by sparse representation. Increases of 4.67, 10.59, 26.82, and 3.44% in the recognition rates are observed for the proposed MSRC method when compared with conventional SRC using the four public finger vein database.

Dimensionality reduction techniques are powerful tools for face recognition, because they obtain important information from a dataset. Several dimensionality reduction methods proposed in literature have been improved thanks to pre-processing approaches. However, they also require post-processing to rectify and increase the quality of projected data. This study presents a simple and new discriminative post-processing framework to make the dimensionality reduction methods robust to outliers. In detail, the proposed approach separates features according to their scale using multidimensional ensemble empirical mode decomposition (MEEMD) and then the spatial and frequency domain processing methods are employed to preserve crucial features. The performance of the proposed method is evaluated on ORL, Extended Yale B, AR, and LFW datasets by several dimensionality reduction techniques. The experimental results demonstrate that the proposed algorithm can perform very well in face recognition.

The use of low-resolution images adopting more relaxed acquisition conditions such as mobile phones and surveillance videos is becoming increasingly common in iris recognition nowadays. Concurrently, a great variety of single image super-resolution techniques are emerging, especially with the use of convolutional neural networks (CNNs). The main objective of these methods is to try to recover finer texture details generating more photo-realistic images based on the optimisation of an objective function depending basically on the CNN architecture and training approach. In this work, the authors explore single image super-resolution using CNNs for iris recognition. For this, they test different CNN architectures and use different training databases, validating their approach on a database of 1.872 near infrared iris images and on a mobile phone image database. They also use quality assessment, visual results and recognition experiments to verify if the photo-realism provided by the CNNs which have already proven to be effective for natural images can reflect in a better recognition rate for iris recognition. The results show that using deeper architectures trained with texture databases that provide a balance between edge preservation and the smoothness of the method can lead to good results in the iris recognition process.

Biometric-based authentication systems offer undeniable benefits to users. However, biometric features are vulnerable to attacks, especially those happening over transmission network or at the stored biometric templates. In this work, we propose a novel biometric-based remote authentication framework to deal with malicious attacks over the transmission channel and at the untrusted server. More concretely, the proposed framework is not only resistant against attacks on the network but also protects biometric templates stored in the untrusted server's database, thanks to the combination of fuzzy commitment protocol and non-invertible transformation techniques. The notable feature as compared to previous biometric based remote authentication framework is its ability to defend the sensitive data against different kinds of insider attacks. The server's administrator is incapable of utilizing information saved in its database to impersonate the clients and deceive the whole system because secure computing in the server is guaranteed by employing a secure coprocessor embedded in the server. In addition, the system performance is maintained with the support of random orthonormal project, which reduces computational complexity while preserving its accuracy.

An analysis of a large set of biometric data obtained during the enrolment and the verification phase in an experimental biometric system installed in bank branches is presented. Subjective opinions of bank clients and of bank tellers were also surveyed concerning the studied biometric methods in order to discover and to explore relations emerging from the obtained multimodal dataset. First, data acquisition and identity verification methods are described in this study. Then, relationships between ratios of successful and failed verifications between pairs, triplets, and quartets of biometric modalities are studied. An analysis of the sentiment of clients and of banking tellers related to each identity verification attempt was performed based on linguistic methods. The data mining process is described, based on the rough sets methodology, aimed at deriving rules pertaining to consecutive identity verification attempts.

This article studies user classification into children and adults according to their interaction with touchscreen devices. The authors analyse the performance of two sets of features derived from the sigma-lognormal theory of rapid human movements and global characterisation of touchscreen interaction. The authors propose an active detection approach aimed to continuously monitor the user patterns. The experimentation is conducted on a publicly available database with samples obtained from 89 children between 3 and 6 years old and 30 adults. The authors have used support vector machines algorithm to classify the resulting features into age groups. The sets of features are fused at the score level using data from smartphones and tablets. The results, with correct classification rates over 96%, show the discriminative ability of the proposed neuromotor-inspired features to classify age groups according to the interaction with touch devices. In the active detection set-up, the authors’ method is able to identify a child using only four gestures in average.