GAN-Based Blur Restoration for Finger Wrinkle Biometrics System

Fingerprint recognition is one of the most well-known biometrics and the most widely utilized biometric option for computerized system authentication. The traditional secret key and token-based verification methods are gradually being phased out in favor of biometric solutions. The two most important aspects to consider when designing a biometric system are security and accuracy of recognition. In this work, a comprehensive assessment is presented to throw light on the most recent advances in the study of unique fingerprint-based biometrics from these two perspectives, with the objective of enhancing system security and recognition accuracy. Also, researchers suggest utilizing a normalized fingerprint model to combine fingerprints from various layouts. Constrictions of previous exploration work are highlighted, and ideas for future work are made, based on an in-depth investigation and discussion. According to the report, scientists continue to have issues dealing with the two most fundamental attacks on biometric systems, namely, attacks on the User Interface (UI) and attacks on format databases. Building genuine remedies to thwart these attacks, while ensuring strong security while maintaining high recognition precision, is a hot research topic right now, and will be in the near future. Furthermore, recognition accuracy in less-than-ideal circumstances is sure to be unsatisfactory, necessitating special attention in the design of a biometric system. This research also depicts related obstacles and patterns of momentum exploration.

show abstract

“…TV screens are shaded by picture shows. Computer systems drive screens with the output of picture and design show cards [15,16,17].…”

Section: Image Generationmentioning

confidence: 99%

A Paper on Finger Print Recognition

Bhalla

Gupta

2021

JPRI

View full text Add to dashboard Cite

show abstract

“…Be that as it may, existing methods for deblurring of biometric images are mostly focused on other biometric traits such as iris [12], face [1], and hand-based biometric traits such as palm prints [23] or finger wrinkles [5]. Cho et al [5] proposed a GAN-based model for deblurring finger wrinkles for authentication. Having mentioned these techniques, fingerprint deblurring has not been explored enough.…”

Section: Related Workmentioning

confidence: 99%

FDeblur-GAN: Fingerprint Deblurring using Generative Adversarial Network

Joshi¹,

Dabouei²,

Dawson³

et al. 2021

Preprint

View full text Add to dashboard Cite

While working with fingerprint images acquired from crime scenes, mobile cameras, or low-quality sensors, it becomes difficult for automated identification systems to verify the identity due to image blur and distortion. We propose a fingerprint deblurring model FDeblur-GAN, based on the conditional Generative Adversarial Networks (cGANs) and multi-stage framework of the stack GAN. Additionally, we integrate two auxiliary sub-networks into the model for the deblurring task. The first sub-network is a ridge extractor model. It is added to generate ridge maps to ensure that fingerprint information and minutiae are preserved in the deblurring process and prevent the model from generating erroneous minutiae. The second sub-network is a verifier that helps the generator to preserve the ID information during the generation process. Using a database of blurred fingerprints and corresponding ridge maps, the deep network learns to deblur from the input blurry samples. We evaluate the proposed method in combination with two different fingerprint matching algorithms. We achieved an accuracy of 95.18% on our fingerprint database for the task of matching deblurred and ground truth fingerprints.

show abstract