Page 123 - Conference Book The 3rd IConVET "Future of TVET Graduates: Developing Talent for Industry 4.0 and The New Normal"

The 3 International Conference
rd
on Vocational Education and Technology (IConVET)
th
Universitas Pendidikan Ganesha - Bali | 7 November, 2020
EVALUATION OF CONTRAST ENHANCEMENT
METHODS ON FINGER VEIN NIR IMAGES

I M D Maysanjaya , M W A Kesiman and I M Putrama 2
1
1

1 Virtual Vision, Image, and Pattern Research Group (VVIP-RG)
2 Data Science Research Group (DS-RG)

1,2 Faculty of Engineering and Vocational, Universitas Pendidikan Ganesha

Corresponding author: dendi.ms@undiksha.ac.id

Abstract. Biometrics is a technology used to identify a person based on
physical characteristics and behavioural characteristics. Biometrics is used
to increase the importance of personal data. However, many biometric
models can be manipulated, such as fingerprints. To cover the fragility, a
biometric pattern based on a blood vein, such as finger vein pattern, was
developed. To obtain a clear image of the finger vein, one of the acquisitions
tools used is called Near-Infrared (NIR). Despite using NIR technology in
the acquisition process, it is not uncommon for the finger vein pattern to be
unclear. To overcome this problem, it is necessary to increase the contrast
quality of the image. This study proposes the use of the BPDFHE method to
improve the contrast quality of finger vein NIR images. As a comparison
material for performance tests, the HE, AHE, and CLAHE methods were
also tested. The test is carried out according to AMBE, PSNR, SSIM, FSIM,
and computation time parameters. Based on the test, the results showed that
the BPDFHE obtains AMBE, PSNR, SSIM, and FSIM up to 0.054, 26.873,
0.840, and 0.906, respectively. It also gains the less computation time up to
10.988 seconds. These results indicate that BPDFHE is an effective and
efficient method in improving the contrast quality of finger vein NIR images.

| IConVET 2020 | 123

https://conference.undiksha.ac.id/iconvet/ | 123 |