Design and development of dorsal hand vein recognition biometric system using image processing on fpga

By:
Raghavendra D C
Nischitha B M
Navya L R

1DA10TE038
1DA10TE034
1DA10TE033

Under the Guidance of: Dr. B. Sivakumar

Prof. & Head, Dept. of T.C.E.
Dr. A.I.T.

9 March 2014

1

Outline of Presentation
 Objective
 Introduction

 Existing System
 Proposed System
 Literature Survey
 Block Diagram
 Methodology
 Implementation
 Expected Outcomes
 Advantages
 Disadvantages

 Applications
 Conclusion & Discussion
 References
9 March 2014

Design and Development of Dorsal Hand Vein Recognition
Biometric System Using Image Processing on FPGA

2

Objective
 With an increasing emphasis on security, automated personal

identification using hand Vein biometric feature is becoming a very
active topic in both research and practical applications.

 The objective of this work is to present a biometric authentication

system for high security physical access control based on dorsal
hand vein pattern.

9 March 2014


3

Introduction
 Biometric is an automated method of recognizing an individual

based on measurable biological (anatomical and physiological)
and behavioural characteristics.
 We need biometric system because, now we are living in a global
society of increasingly desperate and dangerous people who
cannot be trusted based on identification documents, identity
thieves steal PIN (date of birth) to withdraw money from
accounts.
 Dorsal hand vein recognition system is the identification of the
person by recognizing the vein pattern presented in the back of
hand.

9 March 2014


4

Existing System
 Existing Biometrics system includes fingerprint, face, iris,
voice, signature and geometry.
Fingerprint recognition:.
• Iris recognition:
Face recognition:
Voice recognition:

9 March 2014


5

Proposed system
 To over come the existing above mentioned biometric system
demerit we are developing a vein recognition system.
 Dorsal hand vein recognition works by scanning the veins inside of
back of hand and creating a digital template that represents vein
pattern. It has the following advantages
 Live Body Identification
 Internal Features
 Non-contact
 High Security

9 March 2014


6

Literature Survey
 The term “biometrics” is derived from the Greek words “bio” (life)
and “metrics” (to measure).

 The first biometric identification system based on vascular
structure called “VeinCheck” was presented by Mac Gregor and
Welford in 1991 .
 In 2000 research paper describing the use of vascular patterns
for recognition is published by Sang-Kyun and Soo-Won Kim. This
paper describes the technology that was to become the first
commercially available vascular pattern recognition system in
2000. The technology uses the subcutaneous blood vessel
pattern in the back of the hands to achieve recognition.
9 March 2014


7

Block Diagram

It consists of following components:




9 March 2014

Person Showing Hand
Camera
Spartan 3e FPGA
LEDs

8

Algorithmic Level Block Diagram:
Image Acquisition

Pre-processing

Segmentation

Post processing

Feature
Extraction

Matching

It consists of following blocks:
 Image Acquisition
 Pre-processing
 Segmentation
 Post Processing
 Features Extraction
 Matching
9 March 2014


9

Methodology
 Base Paper: Dorsal Hand Vein Biometry by Independent Component
Analysis by V.H. Yadav Department of Electronics & Telecommunication.
International Journal on Computer Science and Engineering (IJCSE). ISSN :
0975-3397 Vol. 4 No. 07 July 2012. (Pg No: 1338-1344).
o In this study, they have used the BOSPHORUS hand vein database
which has been taken under a source of near infrared radiation. For
feature extraction they applied appearance based method Independent
Component Analysis (ICA) which produces independent components. To
control over the number of independent component, pre-processed the
data by PCA before applying ICA, and gives good experimental results.

 ICA is very complicated method to extract the features hence we are using a
different method to realise this system.
9 March 2014


10

 Near Infrared Imaging has been used to capture the image since this

produces a more clear vein pattern than visible light.
 First pre-processing the image using median filters for smoothing and
noise removal and histogram stretching for contrast enhancement.
 Next image segmentation is performed using a filters that emphasizes

the directional nature of vein along with local thresholding.
 The image is then post processed to remove noise, using morphological
operations and blob removal.
 Afterwards the entire pattern is thinned to a 1-pixel thick skeleton which
is used as a feature for matching and recognition.
 Images are matched using a Modified Hausdorff Distance which
produces an average distance between two thinned vein patterns.
 A threshold is used to decide if two vein patterns are similar or not.
9 March 2014


11

 Image processing algorithms will be designed and developed for

identification of the back of hand and vein pattern.
 This will be implemented in MATLAB/SIMULINK.
 Verilog implementation of the same will be performed and
implementation of the same is done on FPGA.

 Interfacing of coloured lights to FPGA.
The hardware will be using are
o 3Mpixel Web Camera (more details will be given after
purchase).
o Spartan 3e FPGA (Digilent).
o LEDs.
9 March 2014


12

The software will be using are
o MATLAB/SIMULINK -software model for testing.
o MODELSIM- Simulation tool for Verilog/VHDL.
o Xilinx System Generator- Interface camera to FPGA.

9 March 2014


13

Expected Outcomes
 Designing system has two modes of operation - training mode
and authentication mode.
 In training mode it will update the database of vein pattern of
user.
 In authentication mode it will access the database to identify
the user.
 If authorized person shows the dorsal hand then blue light will
glow otherwise red light will glow.

9 March 2014


14

 The vein patterns are unique to each individual and apart from size,
the pattern does not change over time.
 A contact free technology involving no physical contact with the
hardware.
 Extremely difficult to steal/misuse as veins are not visible to naked
eye.
 Non-contact models are more hygienic than all forms of contact
biometrics.
 Someone might take a picture of authorised user and show it to the

camera but system will not accept.
9 March 2014


15

Disadvantages
 If the person lost his hand then this recognition won’t work.

Applications
 Biowatches and Biokeys.

 Intelligent door handles, door locks.
 Computer Logins.
 POS(Point Of Sale) Authentication.

 Time Attendance Applications.
 Physical Access Control.
9 March 2014


16

Conclusion & Discussion
 We need biometric system because, now we are living in a global society
of increasingly desperate and dangerous people who cannot be trusted
based on identification documents, identity thieves steal PIN (date of
birth) to withdraw money from accounts.
 Hence in this work, we are developing dorsal hand vein biometric system.
 Dorsal hand vein biometric system uses the subcutaneous blood vessel
pattern in the back of the hands to achieve recognition. The length of the
main vein and the angle at the bifurcation points were used as the key
features for this system.

 It offers higher security and reliable for identification. Also it is more
advantageous than other biometric system
9 March 2014


17

 Now we are developing a prototype of dorsal hand vein biometric
recognition system in which it will detect is authenticated user is

showing his dorsal hand or not, by glowing the blue or red LED.
 Future enhancement can be done for dorsal hand vein recognition to
send message to the predefined number whether “authorised(with
predefined name or ID)” or “unauthorised” person is entering into the

company. By doing this one can know who is entering into company.

9 March 2014


18

References
[1]

“Dorsal Hand Vein Biometry by Independent Component Analysis” by V.H.
Yadav Department of Electronics & Telecommunication.

International

Journal on Computer Science and Engineering (IJCSE). ISSN : 0975-3397
Vol. 4 No. 07 July 2012. (Pg No: 1338-1344).
[2]

“An Efficient Hand Dorsal Vein Recognition Based on Neural Networks” by
TUTUMLU H Erdinc KOCER and Novruz ALLAHVERDIc, Selcuk University,

Bozkır Vocational High School, Turkey. Journal of Selçuk University Natural
and Applied Science Online ISSN: 2147-3781. (Pg No:28-39).
[3]

“Biometric Authentication by Dorsal Hand Vein Pattern” by C. Nandini.
International Journal of Engineering and Technology Volume 2 No. 5, May,

2012. ISSN: 2049-3444 © 2012 – IJET Publications UK. (Pg No:-837-840).
[4]

http://www.biometricupdate.com/201211/history-of-biometrics.

9 March 2014


19

THANK YOU…!

9 March 2014


20

Design and development of dorsal hand vein recognition biometric system using image processing on fpga

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to Design and development of dorsal hand vein recognition biometric system using image processing on fpga (20)

Design and development of dorsal hand vein recognition biometric system using image processing on fpga