English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Many approaches have been proposed to enhance the performance of biometric-based recognition when using poor quality biometric samples. It has been shown that reporting average accuracy, to cover a wide range of quality of biometric samples, is not enough to reflect the actual performance. This raises the need to evaluate biometric systems at each quality level separately. Therefore, challenging biometric databases have been recorded with variety of quality conditions and made publically available. This paper highlights the importance of using Adaptive Quality-Based Thresholding (AQBT) when evaluating the performance of biometric systems under different quality conditions. Furthermore, it shows that, in many cases, recognition accuracy evaluation reported in the literature under different quality conditions has two problems. First, the performance is evaluated under hidden assumption, which is AQBT. Second, the reported results cannot be achieved in real-life applications. In order to remedy the two problems, two requirements are to be imposed: 1) the matching criteria should be based on AQBT and 2) at the verification stage the quality level of an input biometric sample should be determined and classified into one of a nonoverlapping predefined quality levels. Finally, we propose the use of adaptive quality-based feature extraction to enhance the accuracy of biometric systems. Although this paper focuses on face biometric as a case study, the discussion can be equally applied to other biometric treats. We illustrate our ideas by experiments conducted on the extended Yale B face benchmark dataset.

Exploiting Samples Quality in Evaluating and Improving Performance of Biometric Systems