Path: EDN Asia >> Design Ideas >> Test & Measurement >> Finger vein identification lowers cost for biometric system
Test & Measurement Share print

Finger vein identification lowers cost for biometric system

17 Sep 2013  | Tan Pin Yang, Lau Wei Cheang, Bakhtiar Affendi bin Rosdi

Share this page with your friends

Although there is an angle correction algorithm that intelligently crops the image, there may still be some vertical or horizontal translation or rotation. Hence, for every enrolment, the system enrols data from five vein images into the database to enhance the accuracy of the system.

System functionality verification
We verified the algorithm using Labview by collecting two sessions of finger vein sample image from the public. For the first collection, the sample was treated as enrolment. For the second collection from the same source, the sample was treated as tester. With the help of Labview, we duplicated our algorithm and divided it into parts to process the finger vein sample images. All the verification algorithms ran automatically so the time taken to get the optimum parameter was reduced tremendously. From the verification process, we obtained the optimum parameter for each algorithm.

The minimum Equal Error Rate (EER), MGF kernel size, BLPOC ratio and threshold score are the three important parameters that contribute to the stability and accuracy of our system. With the optimum parameters obtained, the system is robust, reliable and fast, and operates with 98.93% accuracy.

In image acquisition, we implemented the closed loop feedback system to control the intensity of near-infrared light to ensure that the best quality of image is acquired. Apart from this, we included an automatic finger detection system to fully automate the image acquisition. We optimised the enhancement and matching processes by looking into the error generated at the result. This was done by choosing the best combination of BLPOC ratio and kernel size as the parameter in our system. We obtained an EER of 1.07% at threshold 0.22235.

We are confident that our system is stable and can identify the finger veins accurately. This system is practical and can be implemented in real life as the user can be identified within a short period of time. The total average time taken from cropping to matching is 545.9324ms.

Future plan
We plan to further improve this system to accommodate remote authorisation since the merging of existing and future networking developments with biometric solutions will allow people to have the opportunity to authorise a wide range of transactions, such as voting, purchasing, accessing and decision-making authorisations, via the network from remote locations.

- Lau Wei Cheang
- Tan Pin Yang
- Tan Teck Leong

  Universiti Sains Malaysia

The authors recently graduated as electronics engineers and this work was part of their project at the university. Their supervising teacher was Bakhtiar Affendi bin Rosdi, senior lecturer at the university.

This case study was submitted to the National Instruments' Graphical Systems Design Achievement Awards contest held last year. It is re-produced here after editing, with permission from National Instruments and the authors.

 First Page Previous Page 1 • 2 • 3

Want to more of this to be delivered to you for FREE?

Subscribe to EDN Asia alerts and receive the latest design ideas and product news in your inbox.

Got to make sure you're not a robot. Please enter the code displayed on the right.

Time to activate your subscription - it's easy!

We have sent an activate request to your registerd e-email. Simply click on the link to activate your subscription.

We're doing this to protect your privacy and ensure you successfully receive your e-mail alerts.

Add New Comment
Visitor (To avoid code verification, simply login or register with us. It is fast and free!)
*Verify code:
Tech Impact

Regional Roundup
Control this smart glass with the blink of an eye
K-Glass 2 detects users' eye movements to point the cursor to recognise computer icons or objects in the Internet, and uses winks for commands. The researchers call this interface the "i-Mouse."

GlobalFoundries extends grants to Singapore students
ARM, Tencent Games team up to improve mobile gaming

News | Products | Design Features | Regional Roundup | Tech Impact