Equal Error Rate

The ever increasing demand of security has resulted in wide use of Biometric systems. Despite overcoming the traditional verification problems, the unimodal systems suffer from various challenges like intra class variation, noise in the sensor data etc, affecting the system performance. These problems are effectively handled by multimodal systems. In this paper, we present multimodal approach for palm- and fingerprints by feature level and score level fusions (sum and product rules). The proposed multi-modal systems are tested on a developed database consisting of 440 palm- and fingerprints each of 55 individuals. In feature level fusion, directional energy-based feature vectors of palm- and fingerprint identifiers are combined to form joint feature vector that is subsequently used to identify the individual using a distance classifier. In score level fusion, the matching scores of individual classifiers are fused by sum and product rules. Receiver operating characteristics curves are formed for unimodal and multimodal systems. Equal Error Rate (EER) of 0.538% for feature level fusion shows best performance compared to score level fusion of 0.6141 and 0.5482% of sum and product rules, respectively. Multimodal systems, however, significantly outperform unimodal palm- and fingerprints identifiers with EER of 2.822 and 2.553%, respectively.

The majority of computer systems employ a login ID and password as the principal method for access security. In stand-alone situations, this level of security may be adequate, but when computers are connected to the internet, the vulnerability to a security breach is increased. In order to reduce vulnerability to attack, biometric solutions have been employed. In this paper, we investigate the use of a behavioural biometric based on keystroke dynamics. Although there are several implementations of keystroke dynamics ...

This paper presents a new approach to authenticate individuals using triangulation of hand vein images and simultaneous extraction of knuckle shape information. The proposed method is fully automated and employs palm dorsal hand vein images acquired from the low-cost, near infrared, contactless imaging. The knuckle tips are used as key points for the image normalization and extraction of region of

Wavelets are widely used to extract the texture features for pattern recognition applications including biometric authentication. This can be attributed to the discriminating capability of wavelet features and the availability of fast algorithms for implementing discrete wavelet transform (DWT). In most of the wavelet based palmprint applications, distribution of energies in space-frequency domain are treated as features to classify the palmprints. Although wavelet energies are good discriminating features, they fail to characterize the palmprints sufficiently. In order to enhance the discriminating capability and the palmprint recognition accuracy, we extract the intramodal palmprint line and energy features from the same wavelet decomposition of palmprint. The score level (product rule and sum rule) fusion of these features improves the recognition accuracy significantly. We empirically found 39.38% relative improvement (RI) with an overall equal error rate (EER) of 1.37%, on PolyU online palmprint database (left hand palmprints). The computational burden on feature extraction is very less, and the EER is superior to other state of the art approaches. The comparison of the results with the state of the art wavelet and fusion based palmprint recognition approaches demonstrates the effectiveness of the proposed approach in classifying palmprints.

Besides the iris, conjunctival vasculature may also be used for ocular biometric recognition. Conjunctival vessel patterns can be easily observed in the visible spectrum and can compensate for off-angle or otherwise occluded iridial texture. In this paper, classification of conjunctival vasculature using Gray Level Co-occurrence Matrix (GLCM) is studied. Statistical features of GLCM, i.e., contrast, correlation, energy and homogeneity, were used in conjunction with Fisher linear discriminant analysis and regularized neural network classifiers in order to recognize textures arising from conjunctival vessels. Match score level fusion of Fisher LDA and neural networks provided the best results, resulting in a test set equal error rate (EER) and area under receiver operating characteristics curve (ROC AUC) of 13.97% and 0.9333, respectively. These figures improved to 11.9% and 0.9504 after fusion of LDA and neural network match scores.

This paper proposes a novel method of image-based fingerprint matching based on the features extracted by “FingerCode”. The experiments show that our system outperforms the standard “FingerCode” recognition method and other image-based approaches. Combining the matching score generated by the proposed technique with that obtained from a minutiae-based matcher results in an overall improvement in performance of the fingerprint matching.

Three simple speaker verification techniques are developed and tested using the same experimen-tal protocol. The first two methods are text-independent ones. They are based on vector quan-tization (VQ) [2, 3] and sphericity-based models [4]. Text-independent ...

One methodology for evaluating the matching performance of biometric authentication systems is the detection error tradeoff (DET) curve. The DET curve graphically illustrates the relationship between false rejects and false accepts when varying a threshold across a genuine and an imposter match score distributions. This paper makes two contributions to the literature on the matching performance evaluation of biometric identification or bioauthentication systems. First, we create curvewise DET confidence regions using radial sweep methods. Second we use this same methodology to create pointwise confidence intervals for the equal error rate (EER). The EER is the rate at which the false accept rate and the false reject rate are identical. We utilize resampling or bootstrap methods to estimate the variability in both the DET and the EER. Our radial sweep is based on converting the false reject and false accept errors to polar coordinates. Application is made of these methods to data from three different biometric modalities and we discuss the results of these applications.