Multiple classifiers in biometrics. part 1: Fundamentals and review

Abstract

We provide an introduction to Multiple Classifier Systems (MCS) including basic nomenclature and describing key elements: classifier dependencies, type of classifier outputs, aggregation procedures, architecture, and types of methods. This introduction complements other existing overviews of MCS, as here we also review the most prevalent theoretical framework for MCS and discuss theoretical developments related to MCS.The introduction to MCS is then followed by a review of the application of MCS to the particular field of multimodal biometric person authentication in the last 25 years, as a prototypical area in which MCS has resulted in important achievements. This review includes general descriptions of successful MCS methods and architectures in order to facilitate the export of them to other information fusion problems.Based on the theory and framework introduced here, in the companion paper we then develop in more technical detail recent trends and developments in MCS from multimodal biometrics that incorporate context information in an adaptive way. These new MCS architectures exploit input quality measures and pattern-specific particularities that move apart from general population statistics, resulting in robust multimodal biometric systems. Similarly as in the present paper, methods in the companion paper are introduced in a general way so they can be applied to other information fusion problems as well. Finally, also in the companion paper, we discuss open challenges in biometrics and the role of MCS to advance them.

Characterization of the Handwriting Skills as a Biomarker for Parkinson’s Disease

“…In addition to the tests on individual tasks, all of the tasks are combined to create a generalized model. This combination is performed in late-fusion strategy, i.e., the scores of the classifier (RBF-SVM in all of the cases) are combined according to the mean rule to obtain a new score [27]. The results are reported in Table III. The results obtained with the fusion strategy are better than those obtained with individual models for each task.…”

Active detection of age groups based on touch interaction

“…As future work, we aim at building a reliable and fast classifier of users from all ages based on their touchscreen interaction [30] based on a combination of different expert systems [31]. The main drawback of other methods like using the browsing history or social network profiles is that they need a high amount of data.…”

“…The global features set (see Table 3) is also extracted for swipe tasks. Both sets (lognormal and global) are compared and combined using: fusion at the score level and fusion at the feature level [31].…”

“…The selection of the threshold to calculate when the user is detected as a child is crucial in performance terms: a high threshold could decrease the number of false detections (adults detected as a child), but also it could increase the time delay (time between the child starts to operate the device till he is detected). In order to choose the best threshold, we will present average detection delay (ADD) and probability of false detection (PFD) curves, previously used in [31]. ADD curves show the number of samples necessary to detect a child as a function of the threshold.…”

Reducing the template ageing effect in on‐line signature biometrics

“…A final fusion of the three systems after applying template update is also performed computing the sum of the matching scores. The sum rule fusion algorithm has been considered in this work as it is one of the most successful and easiest approaches used in many related works [33,34]. Before applying the fusion, the scores from each system were normalised to a common range [0, 1] using tanh-estimators [35].…”