Show simple item record

dc.contributor.authorPrommegger, Bernhard
dc.contributor.authorKauba, Christof
dc.contributor.authorUhl, Andreas
dc.contributor.editorBrömme, Arslan
dc.contributor.editorBusch, Christoph
dc.contributor.editorDantcheva, Antitza
dc.contributor.editorRathgeb, Christian
dc.contributor.editorUhl, Andreas
dc.date.accessioned2019-06-17T10:00:12Z
dc.date.available2019-06-17T10:00:12Z
dc.date.issued2018
dc.identifier.isbn978-3-88579-676-3
dc.identifier.issn1617-5468
dc.identifier.urihttp://dl.gi.de/handle/20.500.12116/23781
dc.description.abstractFinger-vein scanners or vein-based biometrics in general are becoming more and more popular. Commercial off-the-shelf finger-vein scanners usually capture only one finger from the palmar side using transillumination. Most scanners have a contact area and a finger-shaped support where the finger has to be placed onto in order to prevent misplacements of the finger including shifts, planar rotation and tilts. However, this is not able to prevent rotation of the finger along its longitudinal axis (also called non-planar finger rotation). This kind of finger rotation poses a severe problem in finger-vein recognition as the resulting vein image may represent entirely different patterns due to the perspective projection. We evaluated the robustness of several finger-vein recognition schemes against longitudinal finger rotation. Therefore, we established a finger-vein data set exhibiting longitudinal finger rotation in steps of 1° covering a range of 90°. Our experimental results confirm that the performance of most of the simple recognition schemes rapidly decreases for more than 10° of rotation, while more advanced schemes are able to handle up to 30°.en
dc.language.isoen
dc.publisherKöllen Druck+Verlag GmbH
dc.relation.ispartofBIOSIG 2018 - Proceedings of the 17th International Conference of the Biometrics Special Interest Group
dc.relation.ispartofseriesLecture Notes in Informatics (LNI) - Proceedings, Volume P-282
dc.subjectLongitudinal Finger Rotation
dc.subjectFinger-Vein Recognition
dc.subjectMulti-Perspective Finger-Vein Data Set
dc.subjectPerformance Evaluation
dc.subjectFinger Vein Scanner Device
dc.titleLongitudinal Finger Rotation - Problems and Effects in Finger-Vein Recognitionen
dc.typeText/Conference Paper
dc.pubPlaceBonn
mci.conference.locationDarmstadt
mci.conference.date26.-28. September 2018


Files in this item

Thumbnail

Show simple item record