Hand vein-based biometric authentication with limited training samples
| dc.contributor.advisor | Coetzer, Johannes | en_ZA |
| dc.contributor.advisor | Swanepoel, J. | en_ZA |
| dc.contributor.author | Beukes, Emile | en_ZA |
| dc.contributor.other | Stellenbosch University. Faculty of Science. Dept. of Mathematical Sciences (Applied Mathematics) | en_ZA |
| dc.date.accessioned | 2018-02-16T11:11:47Z | |
| dc.date.accessioned | 2018-04-09T06:54:41Z | |
| dc.date.available | 2018-02-16T11:11:47Z | |
| dc.date.available | 2018-04-09T06:54:41Z | |
| dc.date.issued | 2018-03 | |
| dc.description | Thesis (MSc)--Stellenbosch University, 2018. | en_ZA |
| dc.description.abstract | ENGLISH ABSTRACT : A number of novel hand vein-based biometric authentication systems are proposed. Said systems are non-intrusive and may for example assist with user authentication at automated teller machines. An infrared image of either the dorsal or ventral surface of an individual's hand is acquired through specialised equipment, after which the geometrical properties of the hand are used to extract a suitable region of interest (ROI). A novel protocol, which is based on morphological reconstruction, is employed for the purpose of isolating the veins within the ROI. Feature vectors are extracted from the isolated veins through the calculation of the discrete Radon transform. The feature vectors are appropriately normalised in order to ensure rotational, translational and scale invariance. The dissimilarity between the corresponding feature vectors extracted from a questioned image and a reference image belonging to the claimed client are represented by an average Euclidean or dynamic time warping-based distance. A score-based or rank-based classi er is subsequently employed for authentication purposes. It is demonstrated that, when only one training sample (of arbitrary quality) is available per client, and the client is granted six opportunities for authentication, an average error rate (AER) of 2.85% is achievable for a data set that contains dorsal hand vein patterns from 100 individuals. In a scenario where the single training sample is guaranteed to be of very high quality and the client is granted only three opportunities for authentication, the AER may be reduced to 0.77%. | en_ZA |
| dc.description.abstract | AFRIKAANSE OPSOMMING : 'n Aantal nuwe handbloedvatgebaseerde biometriese veri kasiestelsels word voorgestel. Hierdie stelsels is nie-indringend en kan byvoorbeeld met gebruikersverifikasie by outomatiese tellermasjiene help. 'n Infrarooi-beeld van die dorsale of ventrale oppervlak van 'n individu se hand word met behulp van gespesialiseerde toerusting verkry, waarna die geometriese eienskappe van die hand gebruik word om 'n geskikte fokusgebied (FG) te onttrek. 'n Nuwe protokol, wat op morfologiese rekonstruksie gebaseer is, word gebruik om die bloedvate in die FG te isoleer. Kenmerkvektore word vanuit die geisoleerde bloedvate met behulp van die berekening van die diskrete Radon-transform onttrek. Die kenmerkvektore word vervolgens sinvol genormaliseer ten einde rotasie-, translasie- en skaal-invariansie te verseker. Die verskil tussen die ooreenstemmende kenmerkvektore wat vanuit 'n bevraagtekende beeld en 'n verwysingsbeeld wat aan die beweerde kliënt behoort, onttrek is, word deur 'n gemiddelde Euklidiese of dinamiese tydsverbuigingsafstand voorgestel. 'n Telling- of ranggebaseerde klassi seerder word vervolgens vir veri kasiedoeleindes aangewend. Daar word aangetoon dat, wanneer slegs een afrigvoorbeeld (van lukrake kwaliteit) per kliënt beskikbaar is, en die kliënt ses geleenthede vir veri kasie gebied word, 'n gemiddelde foutkoers (GFK) van 2.85% haalbaar is vir 'n datastel wat dorsale handbloedvatpattrone van 100 individue bevat. In 'n scenario waar die enkele afrigvoorbeeld gewaarborg is om van 'n baie hoë kwaliteit te wees en die kliënt slegs drie geleenthede vir veri kasie gebied word, kan die GFK na 0.77% verminder word. | af_ZA |
| dc.format.extent | xvi, 92 pages : illustrations (some colour) | en_ZA |
| dc.identifier.uri | http://hdl.handle.net/10019.1/103387 | |
| dc.language.iso | en_ZA | en_ZA |
| dc.publisher | Stellenbosch : Stellenbosch University | en_ZA |
| dc.rights.holder | Stellenbosch University | en_ZA |
| dc.subject | Hand veins authentication | en_ZA |
| dc.subject | Biometry -- Research -- South Africa | en_ZA |
| dc.subject | Non-intrusive authentication | en_ZA |
| dc.subject | Identification systems | en_ZA |
| dc.subject | UCTD | en_ZA |
| dc.subject | Biometric identification | en_ZA |
| dc.title | Hand vein-based biometric authentication with limited training samples | en_ZA |
| dc.type | Thesis | en_ZA |