2014 IEEE 22nd Signal Processing and Communications Applications Conference (SIU 2014)
%ø5%ø7'g1hùh0h7$%$1/, EL DAMARI %øY20(75ø 6ø67(0ø
ONE-BIT TRANSFORM BASED HAND VEIN BIOMETRIC SYSTEM
.EUD&H]DLUOLR÷OX1, Mehtap Çakmak1, Sarp Ertürk1,2
1
.RFDHOLhQLYHUVLWHVLøúDUHWYH*|UQWøúOHPH/DE.8/,6
ElHNWURQLNYH+DEHUOHúPH0K%|OP, Kocaeli Üniversitesi, 41380 Kocaeli
2
Pars Ar-Ge Ltd., Kocaeli Üniversitesi Teknopark, Kocaeli, Türkiye
ELOJLVLQGHQ \ROD oÕNPDNWDGÕU 'DPDUGDNL NDQ DNÕúÕ LVH
VLVWHPLQ DOGDWÕOPDVÕQÕ ]RUODúWÕUPDNWD YH JLULú \DSDQ
NLúLQLQ FDQOÕOÕ÷Õ YH JHUoHNOL÷L NRQXVXQGD ELOJL
oÕNDUWÕOPDVÕQDRODQDNVD÷ODPDNWDGÕU
%L\RPHWULN NLPOLN GR÷UXODPD \|QWHPOHULQ JHQHO
oDOÕúPD SUHQVLEL LNL DúDPDGDQ ROXúPDNWDGÕU %LULQFL
DGÕPGD WDQÕQDFDN NLúLQLQ EL\RPHWULN YHULVL VLVWHPH
WDQÕWÕOPDN ]HUH DOJRULWPDODU LOH DQDOL] VRQXFX
oÕNDUÕOPDNWD YH YHULWDEDQÕQD ND\GHGLOPHNWHGLU H÷LWLP
YH\D GHYUH\H DOPD DúDPDVÕ øNLQFL DúDPDGD VLVWHPH
JLULú\DSDQNLúLQLQEL\RPHWULNYHULVLD\QÕDOJRULWPDODUOD
oÕNDUÕOPDNWD YH YHULWDEDQÕQGDNL ELOJLOHU LOH
NDUúÕODúWÕUÕOÕS HúOHúPH GHUHFHVLQH EDNÕOPDNWDGÕU WHVW
YH\D NXOODQÕP DúDPDVÕ (úOHúPH GXUXPXQGD NLúLQLQ
NLPOL÷LRQD\ODQPDNWDGÕU
>@¶GH DYXFXQ NÕ]ÕO|WHVL J|UQWOHUL NXOODQDQ ELU
biyometrik dR÷UXODPD VLVWHPL |QHULOPLú ROXS, iki
SDUPD÷ÕQ NRQXPX oDNÕúWÕUPD LoLQ NXOODQÕODUDN LOJLOLOLN
E|OJHVL oÕNDUÕOPÕú VRQUDVÕQGD VXE|OP ZDWHUVKHG
G|QúP NXOODQÕODUDN |]QLWHOLN QRNWDVÕ oÕNDUWÕPÕ
\DSÕOPÕúWÕU>@¶GH\DNÕQNÕ]ÕO|WHVLJ|UQWOHPHQLQX]DNNÕ]ÕO|WHVLWHUPRJUDIL\HRUDQODGDPDUJ|UQWOHPHVLLoLQ
daha uygun ROGX÷XJ|VWHULOPLúWLU>@¶GHLVHDYXoLPJHVL
LOHDYXoGDPDULPJHOHULND\QDúWÕUÕODUDNELU/DSODVDYXo
WDQÕPODPDVÕ ROXúWXUXODUDN NLúL WHVSLWL \DSÕOPÕúWÕU $YXo
GDPDUODUÕQÕQ G|QP YH VRQODQPD QRNWDODUÕQÕQ
EL\RPHWULN WDQÕPODPD LoLQ NXOODQÕOGÕ÷Õ ELU úLIUH VLVWemi
>@¶GH |QHULOPLúWLU >@¶GD LVH 6,)7 X\XPODPD WHPHOOL
ELU DYXo GDPDU GR÷UXODPD VLVWHPL VXQXOPXúWXU >@¶GH
NHQDU WHVSLWL WHPHOOL ELU \DNODúÕPOD LNLOL LPJH\H
G|QúWUOHQ DYXo GDPDU J|UQWOHUL EL\RPHWULN
GR÷UXODPDGD NXOODQÕOPÕúWÕU $YXo GDPDU GR÷UXODPD
VLVWHPOHULLoLQJHQHOELULQFHOHPHLVH>@¶GH\DSÕOPÕúWÕU
>@¶GD DWHúE|FH÷L NPHOHPHVL NXOODQDQ ELU DYXo GDPDU
WHVSLW\DNODúÕPÕ|QHULOPLúWLU
Bu bildiride bir-ELW G|QúP WDEDQOÕ |]JQ ELU
GDPDUKDULWDVÕoÕNDUWÕP\DNODúÕPÕ|QHULOPLúWLUgQHULOHQ
\DNODúÕPÕQ DYXo GDPDUODUÕ oÕNDUWÕPÕQGD NXOODQÕPÕ HOH
DOÕQDUDN HO GDPDU L]L WDEDQOÕ ELU EL\RPHWULN NLPOLN
GR÷UXODPDVLVWHPLJHUoHNOHúWLULOPLúWLU
ÖZETÇE
Bu bildiride, eldeki damar verisi kullanÕODUDN NLúLVHO
NLPOLN GR÷UXODPD DPDoOÕ ELU ELW G|QúP WDEDQOÕ
|]JQ ELU \|QWHP |QHULOPLúWLU <DSÕODQ oDOÕúPDGD avuç
damar bilgilerinin oÕNDUÕOGÕ÷ÕNÕ]ÕO|WHVLUHVLP YHULWDEDQÕ
NXOODQÕODUDNLPJHOHU NDUúÕWOÕNL\LOHúWLUPH |QLúOHPinden
geçirildikten sonra HOLQ VW NÕVPÕQGDQ GDPDU WDQÕma
LoLQ NXOODQÕODFDN LOJLOLOLN E|OJHVL HOGH HGLOPLúWLU. Bu
oDOÕúPDGD |]JQ RODUDN ELU ELW G|QúP NÕVDFD %7
(One Bit Transform), NXOODQÕODUDN GDPDUODUÕQ
oÕNDUÕOGÕ÷Õ LNLOL LPJH elde edilmektedir. 'DPDUODUÕQ
EHOLUJLQ úHNLOGH HOGH HGLOPHVL LoLQ VRQ DúDPDGD
PRUIRORMLN LúOHPOHU X\JXODQPDNWDGÕU 7DQÕPD
DúDPDVÕQGD LNLOL ilinti (korelasyon) LúOHPL \DSÕODUDN
WHVSLWJHUoHNOHúWLULOPHNWHGLU.
ABSTRACT
In this paper, a novel scheme for personal
authentication using palm vein information based on
one-bit transform is presented. In this work, infrared
palm images which contain the palm vein information
are used and after contrast enhancement based preprocessing the corresponding region of interest is
extracted. The one-bit transform is used to obtain the
binary image containing vein information in a novel
approach. To obtain the vein data perceptibly,
morphological processing is used in the final processing
step. In the recognition step binary correlation based
identification is accomplished.
1. *ø5øù
%L\RPHWULN NLPOLN GR÷UXODPD VLVWHPOHUL JQP]GH
JYHQOLN VD÷OÕN EDQNDFÕOÕN JLEL ELUoRN DODQGD
kullanÕOPD\D EDúODQPÕúWÕU 3DUPDN L]L LOH EDúOD\DQ
EL\RPHWULNNLPOLNGR÷UXODPDVLVWHPOHULGDKDVRQUDVÕQGD
\D\JÕQODúDUDN \] WDQÕPD, iris WDQÕPD UHWLQD WDUDPD
JLEL IDUNOÕ EL\RPHWULN |]HOOLNOHUGHQ ID\GDODQPD\D
EDúODPÕúWÕU <DNÕQ ]DPDQGD SDUPDN HO YH DYXo LoL
GDPDU WDQÕPD VLVWHPOHULQLQ NXOODQÕPÕ |]HOOLNOH WHPDV
JHUHNWLUPHPHVL YH VLVWHPLQ DOGDWÕOPDVÕQÕQ GD J|UHFHOL
RODUDN]RUROPDVÕQHGHQL\OHGLNNDWoHNPH\HYHNXOODQÕP
DODQODUÕEXOPD\DEDúODPÕúWÕU>1].
'DPDU WDQÕPD VLVWHPOHUL LQVDQODUÕQ damar
KDULWDVÕQÕQ NHQGLVLQH |]J YH EHQ]HUVL] ROGX÷X
2. ø/*ø/ø/ø. BÖLGESød,.$57,0,
(O GDPDU L]L WDEDQOÕ EL\RPHWULN WDQÕPD VLVWHPLQLQ LON
DúDPDVÕ NXOODQÕODFDN DYXo E|Ogesinin tespit edilerek
ilgililik bölgesinin (region of interest ROI)
1094
2014 IEEE 22nd Signal Processing and Communications Applications Conference (SIU 2014)
oÕNDUWÕOPDVÕGÕU Bu amaçla ilk önce yakalanan el
görüntüsünün çevriti VÕQÕU KDWODUÕ oÕNDUÕOPDNWDGÕU
Bunun için Sobel temelli kenar tespiti uygulanmakta ve
HQ GÕú oHYULW WDNLEL JHUoHNOHúWLULOHUHN HO oHYULWL HOGH
HGLOPHNWHGLU%XLúOHPùHNLOD¶GHJ|VWHULOHQ|UQHNELU
HO LPJHVL LoLQ ùHNLO E¶GH J|VWHULOHQ HO oHYULWLQL
vermektedir.
øNLOL oHYULW LPJHVL ELU VRQUDNL DúDPDGD GDPDU
WHVSLWLQLQ \DSÕODFD÷Õ DYXo VW E|OJHVLQLQ oÕNDUWÕOPDVÕ
LoLQNXOODQÕOPDNWDGÕU%XDPDoODoHYULW imgesi üzerinde
\DWD\ WDUDPD \DSÕODUDN -0 JHoLú VD\ÕVÕ NRQWURO
edilmektedir. 3DUPDNODUÕQ ROGX÷X E|OPGH -0
JHoLúOHULQLQVD\ÕVÕID]ODDYXobölgesinde ise daha D]GÕU
%X \]GHQ EX GH÷LúLP VD\ÕVÕ EHOOL ELU GH÷HULQ DOWÕQD
GúW÷ SLNVHOLQ NRRUGLQDWODUÕ NXOODQÕODUDN LOJLOLOLN
bölgesinin \DWD\ VÕQÕUÕ ROXúWXUXOPDNWDGÕU. 'úH\ VÕQÕU
için ise, ELOH÷LQ EDúODQJÕo konumu temel DOÕQPDNWD YH
VRODOWNÕVÕPGDHOLQJHQLúOHPH|]HOOL÷LLOHGúH\ referans
QRNWDVÕ HOGH HGLOPHktedir. Bu sayede oOXúturulan
EDúODQJÕo (referans) noktasÕndan itibaren 65 x 60
J|UQW HOHPDQÕ SLNVHO ER\XWXQGD ELU 52, elde
HGLOPHNWHGLU%XLúOHPLQVRQXFXoHYULWLPJHVL]HULQGHQ
ùHNLO F¶GH oÕNDUÕODQ 52, LVH ùHNLO G¶GH
gösterilmektedir.
(a)
L\LOHúWLUPH LúOHPLQL JHUoHNOHúWLUPHNWH YH EX VD\HGH
\HUHO GHWD\ODUÕQ GDKD L\L EHOLUJLQOHúWLULOPHVLQL
VD÷ODPDNWDGÕU. Bu amaçla ilgililik bölgesi öncelikle
dikdörtgensel bölgelere bölünür ve her bir bölgeye
NDUúÕWOÕNVÕQÕUODPDOÕKLVWRJUDPHúLWOHPHLúOHPLX\JXODQÕU
BX VD\HGH SHNLúWLULOPLú ELU J|UQW HOGH HGLOPHNWHGLU
&/$+( GDPDU L]OHULQLQ EHOLUJLQOHúWLULOPHVL LoLQ
NXOODQÕODELOHFHN HWNLOL ELU \DNODúÕPGÕU >10,12]. ùHNLO
¶GH|UQHNJ|UQWLoLQRULMLQDO52,YH&/$+(VRQUDVÕ
elde edilen ROI gösterilmektedir.
(a)
(b)
ùHNLO (a) Orijinal ROI E&/$+(X\JXODQPÕú52,
3.2. %LU%LW'|QúP
%LUELWG|QúP1 BT: One-bit transform), gri seviyeli
imgeleri HWNLOL ELU úHNLOGH LNLOL imgelere G|QúWUPHk
LoLQNXOODQÕOPDNWDGÕU>12]. 1BT’de imge öncelikli olarak
çoklu bant-geçiren bir süzgeçten geçirilmektHGLU ùHNLO
¶GH YHULOHQ V]JHo oHNLUGH÷L NXOODQÕODUDN EX
V]JHoOHPH
LúOHPL
GúN
LúOHP
\N\OH
JHUoHNOHúWLULOHELOPHNWHGLU
(b)
(c)
(d)
ùHNLO (a) Örnek el imgesi (b) Elin çevriti
F52,oÕNDUWÕPÕ(d) ROI
ùHNLO dDOÕúPDGDNXOODQÕODQ%7süzgeç
oHNLUGH÷L
3. DAMAR ÇIKARTIMI
3.1. øPJH3HNLúWLUPH
I LPJHVLQLQV]JHoOHQPLúKDOL
(O GDPDUODUÕQÕQ GDKD EHOLUJLQ ROPDVÕ LoLQ |QFHOLNOL
RODUDN LOJLOLOLN E|OJHVLQH NDUúÕWOÕN SHNLúWLUPH
X\JXODQPDNWDGÕU %X DPDoOD NDUúÕWOÕN VÕQÕUODPDOÕ
X\DUODQÕU histogram GHQNOHúWLUPH (CLAHE: ContrastLimited
Adaptive
Histogram
Equalization)
NXOODQÕOPÕúWÕU
6ÕUDGDQ KLVWRJUDP GHQNOHúWLUPH +(: Histogram
Equalization) yöntemi, imgeyi bir bütün olarak
LúOHPHNWHGLU. CLAHE ise yerel bölgeler üzerinden
‫ܭ כ ܫ = ܨܫ‬
(1)
úeklinde elde edilebilmekte olup, burada *, konvolüsyon
LúOHYLQL J|VWHUPHNWHGLU
%7 G|QúP LoLQ
V]JHoOHQPLú
VRQXo
LOH
RULMLQDO
LPJH
NDUúÕODúWÕUÕOPDNWDGÕU
‫ܫ > ܨܫ = ܤ‬
1095
(2)
2014 IEEE 22nd Signal Processing and Communications Applications Conference (SIU 2014)
Bu sayede ikili bir imge elde edilmektedir. Bu imgede
GHWD\ODU RUWD\D oÕNDUÕOPDNWDGÕU ùHNLO E¶GH YHULOHQ
&/$+( VRQUDVÕ 52, LoLQ V]JHoOHPH VRQXFX ùHNLO
4(a)’GD%7VRQXFXLVHùHNLO(b)’de gösterilmektedir.
*|UOG÷ JLELELOGLULGH|QHULOHQ|]JQ%7\DNODúÕPÕ
WHPHOOL \|QWHP VD\HVLQGH HO GDPDUODUÕ EDúDUÕ\OD
oÕNDUÕOPDNWDGÕU
(a)
benzerlik ölçütü sonucuna göre karar verilecek uygun
ELUHúLNGH÷HULQLQEHOLUOHQPHVLJHUHNPHNWHGLU
4.2. Uygun (úLN'H÷HULQLQ%HOLUOHQPHVL
%L\RPHWULNWDQÕPDVLVWHPLLoLQX\JXQELUHúLNGH÷HULQLQ
belirlenmHVLDGÕPÕGHQHWLPOLúHNLOGHJHUoHNOHúWLULOPLúWLU
%X DPDoOD YHULWDEDQÕQGD D\QÕ HOLQ IDUNOÕ ]DPDQODUÕGD
DOÕQPÕú J|UQWOHUL NXOODQÕODUDN D\QÕ HOH DLW GDPDU
KDULWDODUÕQÕQ+DPPLQJPHVDIHKLVWRJUDPÕoÕNDUWÕOPÕúWÕU
6RQUDVÕQGD ise IDUNOÕHOlere ait tüm imgeler NXOODQÕODUDN
IDUNOÕ NLúLOHULQ GDPDU KDULWDODUÕQÕQ PHVDIH KLVWRJUDPÕ
HOGH HGLOPLúWLU øNL KLVWRJUDPÕQ NHVLúLP QRNWDVÕ LVH
X\JXQELUHúLNGH÷HURODUDNEHOLUOHQPLúWLU
ùHNLO ¶GD D\QÕ HOH DLW IDUNOÕ ]DPDQODUGD DOÕQPÕú
J|UQWOHUGHQoÕNDUWÕODQGDPDUKDULWDVÕJLELIDUNOÕHOOHUH
DLW LNL GDPDU KDULWDVÕQÕQ GD J|UQP YH +DPPLQJ
mesafesi gösterilmektedir. G|UOG÷ JLEL D\QÕ HOH DLW
GDPDU KDULWDODUÕ EHQ]HU ROXS QRUPDOL]H +DPPLQJ
mesafesi 0.0832 RODUDNøNLIDUNOÕHOHDLWGDPDUKDULWDODUÕ
içinse normalize Hamming mesafesi daha yüksek ve
0.3872 olarak EXOXQPXúWXU
(b)
ùHNLO (a) Süzgeçleme sonucu (b) 1BT sonucu
3.3. Morfolojik 6RQøúOHP
Son elde edilen ikili imgedeki gürültülerin giderilmesi
için VRQ DúDPDGD temel morfolojik (biçimbilimsel)
LúOHPOHU uygulanmaktadÕU Bunun için damar \DSÕODUÕQD
uygun bir \DSÕ HOHPDQÕ VWUXFWXUH HOHPHQW ile önce
aúÕQGÕUPD HURVLRQ X\JXODQDUDN GDPDUODUÕQ temel
LVNHOHWL oÕNDUWÕOmakta, sRQUDVÕQGD ise NDSDPD LúOHPL LOH
VRQ GDPDU KDULWDVÕ HOGH HGLOPHNWHGLU. Morfolojik son
LúOHPVRQXFXùHNLO¶GHJ|VWHULOPHNWHGLU
(a)
ùHNLO 9HULWDEDQÕQGDQVHoLOHQD\QÕYHIDUNOÕHOOHULoLQ
mesafe ölçütü
(b)
ùHNLO (a) øNLOL imge; (b) Morfolojik VRQLúOHPOHHOGH
HGLOHQGDPDUKDULWDVÕ
5. DENEYSEL SONUÇLAR
'HQH\VHO VRQXoODU LoLQ %R÷D]LoL hQLYHUVLWHVLQLQ el
damar görünWVYHULWDEDQÕNXOODQÕOPÕúWÕU [13],[14]. Bu
YHUL WDEDQÕQGD HO GDPDU J|UQWOHUL \DNÕQ NÕ]ÕO-ötesi
1,5 J|UQWOHPH WHNQRORMLVL\OH \DNDODQPÕúWÕU
.XOODQÕODQ NDPHUDQÕQ \DNODúÕN 0 nm dalga boyuna
NDGDUJ|UQW\DNDOD\DELOPH|]HOOL÷LROXSYHULWDEDQÕQD
IDUNOÕ NLúL\H DLW ¶GHQ ID]OD HO J|rüntüsü
EXOXQPDNWDGÕU *|UQWOHU, SLNVHO J|UQW HOHPDQÕ
EDúÕQD ELW GHULQOL÷H VDKLS YH î 13 piksel
ER\XWXQGDGÕU
%X YHULWDEDQÕQGDn EX oDOÕúPD
NDSVDPÕQGD her elden IDUNOÕ ]DPDQlarGD DOÕQPÕú 2 adet
J|UQW NXOODQÕOPÕúWÕU. Bu sayede NLúLnin elini sisteme
okuturken ROXúDELOHFHN IDUNOÕOÕNODU, zaman veya çevre
NRúXOODUÕQÕQ IDUNOÕ HWNLOHUL VÕFDN RUWDP VR÷XN RUWDP
vb.) GH÷HUOHQGLUPHNDSVDPÕQDDOÕQPÕúWÕU.
<DSÕODQ oDOÕúPD VRQXFXQGD YHULWDEDQÕQGDNL D\QÕ
yönelime sahip resimler DOW NPHVL NXOODQÕOPÕú YH
NXOODQÕODQ WP UHVLPOHU LoLQ GR÷UX WDQÕPD
JHUoHNOHúWLULOPLúWLUøOJLOLHOLQNDUúÕOÕ÷ÕRODQJ|UQWOHULQ
NDSVDP GÕúÕQD DOÕQPDVÕ GXUXPXQGD LVH VLVWHP aranan
4. '$0$5(ù/(ù7ø50(
4.1. 1BT verilerinde benzerlik ölçütü
øNL D\UÕ GDPDU KDULWDVÕQÕQ NDUúÕODúWÕUÕOPDVÕ LoLQ
NRUHODV\RQ WDEDQOÕ EDVLW ELU \DNODúÕP NXOODQÕOPÕúWÕU Bu
DPDoODLNLIDUNÕJ|UQWGHQoÕNDUÕOD%7WDEDQOÕGDPDU
KDULWDODUÕQÕQ LNLOL NRUHODV\RQX DOÕQDUDN ELU EHQ]HUOLN
ölçütü elde edilebilmektedir. øúOHPVHONROD\OÕNDoÕVÕQGDQ
EX LúOHY ELU EHQ]HPH]OLN |OoW úHNOLQGH
JHUoHNOHúWLULOPHNWHGLU 'DPDU KDULWDODUÕQD PDQWÕNVDO
EX-25 X\JXODQDUDN EHQ]HUOLN GXUXPXQGD GúN
EHQ]HPH]OLN GXUXPXQGD LVH \NVHN ELU GH÷HU HOGH
edilmektedir. %X \DNODúÕP HVDVÕQGD damar haritalarÕQÕQ
+DPPLQJ
PHVDIHVLQLQ
EXOXQPDVÕQD
NDUúÕOÕN
gelmektedir. 1RUPDOL]DV\RQ \DSÕOGÕ÷ÕQGD Hamming
PHVDIHVLLOHDUDVÕQGDGH÷LúHFHNROXSD\QÕNLúL\HDLW
GDPDU KDULWDODUÕ LoLQ VÕIÕUD \DNÕQ GúN GH÷HUOHU IDUNOÕ
NLúLOHUH DLW KDULWDODU LoLQVH \NVHN GH÷HUOHU elde
edilecektir. Pratik sistemde biyometrik WDQÕPD için
1096
2014 IEEE 22nd Signal Processing and Communications Applications Conference (SIU 2014)
elin veritabaQÕQGD EXOXQPDGÕ÷Õ úHNOLQGH G|Qú
\DSPDNWDGÕU $\QÕ HOH DLW IDUNOÕ ]DPDQODUGD DOÕQPÕú LNL
J|UQWGHQ oÕNDUWÕODQ GDPDU KDULWDODUÕQD ELU |UQHN
ùHNLO 7’de verilmektedir.
[5] B. Prasanalakshmi, A. Kannammal, “A secure
cryptosystem from palm vein biometrics”,
Proceedings of the 2nd International Conference on
Interaction Sciences: Information Technology,
Culture and Human, ICIS '09, pp. 1401-1405, 2009.
[6] P.-O. Ladoux, C. Rosenberger, B. Dorizzi, “Palm
Vein Verification System Based on SIFT
Matching”, Advances in Biometrics, Lecture Notes
in Computer Science, Vol. 5558, pp 1290-1298,
2009.
[7] P. Ghosh, R. Dutta, “A new approach towards
Biometric Authentication System in Palm Vein
Domain”, International Journal of Advance
Innovations, IJAITI, Vol. 1, No.2, pp. 1-10, 2012.
[8] I. Sarkar, F. Alisherov, T.-H. Kim, D.
Bhattacharyya, “Palm Vein Authentication System:
A Review”, International Journal of Control and
$XWRPDWLRQ9Rú1RSS-33, May 2010.
[9] Z. Honarpisheh, K. Faez, “An Efficient Dorsal
Hand Vein Recognition Based on Firefly
Algorithm”, International Journal of Electrical and
Computer Engineering, IJECE, Vol.3, No.1 , pp.
30-41, Feb. 2013.
[10] M. Yakno, J. Mohamad Saleh, B. Affendi Rosdi,
“Low Contrast Hand Vein Image Enhancement”,
2011 IEEE International Conference on Signal and
Image Processing Applications (ICSIPA2011), pp.
390 – 392, 2011.
[11] S.D. Raut, V.T. Humbe, “Analysis of Multispectral
Palm Vein Image Using Enhancement Operations”,
International Journal of Computer Science
Engineering (IJCSE), pp.103-106, 2013.
[12] S.Erturk, “Region of Interest Extraction in Infrared
6. SONUÇ
Bu bildiride bir-ELWG|QúPNXOODQDQ|]JQELUGDPDU
L]L WDEDQOÕ EL\RPHWULN WDQÕPD VLVWHPL |QHULOPLúWLU
<DSÕODQ GHQH\OHUGH |QHULOHQ \DNODúÕPÕQ GDPDU KDULWDVÕ
oÕNDUÕPÕ LoLQ HIHNWLI YH EDúDUÕOÕ ELU VRQXo YHUGL÷L WHVSLW
HGLOPLúWLU
(a)
(b)
(c)
(d)
ùHNLO(a) øONJ|UQW\HDLW52,E$\QÕHle DLWIDUNOÕ
]DPDQGDDOÕQPÕú52, (c) LONJ|UQWQQGDPDUKDULWDVÕ
(d) LNLQFLJ|UQWQQGDPDUKDULWDVÕ
7. 7(ù(..h5
Images Using One-Bit Transform”, IEEE Signal
Processing Letters, Vol. 20, No. 10, pp. 952-955, 2013.
[13] http://bosphorus.ee.boun.edu.tr/hand/Home.aspx
[14] A. Yuksel, L. Akarun, B. Sankur, “Hand vein biometry
based on geometry and appearance methods”, IET
Computer Vision, Vol. 5, No. 6, 2011
<D]DUODU %R÷D]LoL hQLYHUVLWHVL HO GDPDU J|UQWV
YHULWDEDQÕQÕ KD]ÕUOD\DQ YH SD\ODúDQ 3URI 'U %OHQW
6DQNXU¶DWHúHNNUHGHUOHU
8. KAYNAKÇA
[1] B.Ergen, $dDOÕúNDQ ³%L\RPHWULN 6LVWHPOHU YH (O
7DEDQOÕ %L\RPHWULN 7DQÕPD .DUDNWHULVWLNOHUL´ 6th
International Advanced Technologies Symposium
(IATS’11), pp. 455- 460, (OD]Õ÷7XUNH\ , 2011.
[2] C.-L. Lin, K.-C. Fan, “Biometric verification using
thermal images of palm-dorsa vein patterns”, IEEE
Trans. on Circuits and Systems for Video
Technology, Vol. 14, No.2, pp. 199-213, Feb. 2004.
[3] L. Wang, G. Leedham, “Near- and Far- Infrared
Imaging for Vein Pattern Biometrics” , IEEE Int.
Conf. on Video and Signal Based Surveillance,
AVSS '06, pp. 52, Nov. 2006.
[4] J.-G. Wang, W.-Y. Yau, A. Suwandy, E. Sung,
“Person recognition by fusing palmprint and palm
vein
images
based
on
“Laplacianpalm”
representation”, Pattern Recognition, Vol. 41, No.
5, pp. 1514-1527, May 2008.
1097
Download

BİR BİT DÖNÜŞÜMÜ TABANLI EL DAMARI BİYOMETRİ