User Identification System Based On Finger-Vein Patterns Using Convolutional Neural Network
Finger-vein biometric identification has gained attention recently due to its several advantages over fingerprint biometric traits. Finger-vein recognition is a method of biometric authentication that applies pattern recognition techniques based on the image of human finger-vein patterns. This pape...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
Asian Research Publishing Network (ARPN)
2016
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/17278/2/User%20identification%20system%20based%20on%20finger-vein%20pattern%20using%20convolutional%20neural%20network.pdf http://eprints.utem.edu.my/id/eprint/17278/ http://www.arpnjournals.org/jeas/research_papers/rp_2016/jeas_0316_3805.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Finger-vein biometric identification has gained attention recently due to its several advantages over fingerprint
biometric traits. Finger-vein recognition is a method of biometric authentication that applies pattern recognition techniques based on the image of human finger-vein patterns. This paper is focused on developing a MATLAB-based finger-vein recognition system using Convolutional Neural Network (CNN) with Graphical User Interface (GUI) as the user input. Two layers of CNN out of the proposed four-layer CNN have been used to retrain the network for new incoming subjects. The pre-processing steps for finger-vein images and CNN design have been conducted pm different platforms. Therefore, this paper discusses the method of linking both parts from different platforms using MEX-files in MATLAB. Evaluation is carried out using images of 50 subjects that are developed in-house. An accuracy of an average of 96% is obtained to recognize 1 to 10 new subjects within less than 10 seconds. |
|---|