Single-sensor hand and footprint-based multimodal biometric ...
Single-sensor hand and footprint-based multimodal biometric ...
Single-sensor hand and footprint-based multimodal biometric ...
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
Abstract<br />
Biometric systems support the task of reliable automatic authentication, which is a key<br />
function for economic transactions in modern society. So far, no universal <strong>biometric</strong><br />
modality suitable for all applications has been found. This thesis examines the <strong>multimodal</strong><br />
fusion of different modalities using a single high-resolution scan of the human <strong>h<strong>and</strong></strong> as input<br />
<strong>and</strong> relates existing techniques to a new <strong>biometric</strong> modality: the human foot. With the<br />
target application of wellness areas <strong>and</strong> spas, this new modality supports privacy interests<br />
<strong>and</strong> still provides satisfying accuracy. After an introduction to basic design principles,<br />
related work in palmprint, fingerprint, <strong>h<strong>and</strong></strong> geometry, <strong>and</strong> <strong>footprint</strong>-<strong>based</strong> recognition is<br />
discussed. System modules for sensing, preprocessing, feature extraction, matching <strong>and</strong><br />
decision for both implemented prototype <strong>footprint</strong> <strong>and</strong> <strong>h<strong>and</strong></strong>-<strong>based</strong> <strong>biometric</strong> systems are<br />
described in detail. Necessary modifications due to anatomical differences are proposed<br />
<strong>and</strong> presented. Finally, a performance evaluation comparing overall accuracy <strong>and</strong> relative<br />
performance of individual features concludes this work.<br />
III