Likelihood Ratio-Based Biometric Score Fusion

By Karthik Nandakumar, Yi Chen, Sarat Chandra Dass, Anil K. Jain. 2008

552 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Geometry and Topology, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Geometry and Topology

In Signal Processing

Citations in the corpus

Security Evaluation of Pattern Classifiers under Attack (2014, methodology): “As in [1], [2], we consider the widely used likelihood ratio (LLR) score fusion rule [46], and the matching scores as its features.”
Statistical Meta-Analysis of Presentation Attacks for Secure Multibiometric Systems (2016, methodology): “In [13], [37] spoofing-aware score-level fusion rules were proposed, as variants of the well-known LLR rule [38]:”
Heterogeneous information fusion: A novel fusion paradigm for biometric systems (2011, unknown): “Fusion Mechanisms Incorporating Biometric Sample Quality By exploiting biometric sample quality derived at the signal level, a growing body of literature [36, 5, 8, 18, 11, 13, 22, 14] has shown that this approach can enhance multimodal biometric fusion.”
Robustness of multimodal biometric fusion methods against spoof attacks (2009, methodology): “The likelihood ratio (LLR) between the genuine and impostor distribution is known to be the optimal fusion method, in the sense that it minimizes the probability of error [8] [9].”
Query-adaptive late fusion for image search and person re-identification (2015, methodology): “On the other hand, late fusion refers to fusion at score or decision levels [20, 8, 29, 9].”
Human Identification Using Finger Images (2012, methodology): “The estimation of score densities can be efficiently achieved using Gaussian mixture models (GMMs), and such an approach has shown [40] to offer superior performance than several fixed combination approaches.”
Evaluation of biometric spoofing in a multimodal system (2010, methodology): “We compare three different fusion schemes: weighted sum, likelihood ratio (LLR)[10] and Bayesian LLR [13].”
Increase the Security of Multibiometric Systems by Incorporating a Spoofing Detection Algorithm in the Fusion Mechanism (2011, methodology): “in [12], is the framework based on the Likelihood Ratio test, where the scores are modeled as mixture of Gaussians and a statistical test Ψ(s) is performed to discriminate between genuine and impostor classes.”
Comparison and combination of iris matchers for reliable personal authentication (2010, background): “The score-level combination offers best trade-off in terms of information content and ease in fusion [32].”
Fusing Multi-Stream Deep Networks for Video Classification (2015, methodology): “performed score fusion using a method called likelihood ratio test [30].”