Fingerprint Distortion Rectification Using Deep Convolutional Neural Networks

@article{Dabouei2018FingerprintDR,
  title={Fingerprint Distortion Rectification Using Deep Convolutional Neural Networks},
  author={Ali Dabouei and Hadi Kazemi and Seyed Mehdi Iranmanesh and Jeremy M. Dawson and Nasser M. Nasrabadi},
  journal={2018 International Conference on Biometrics (ICB)},
  year={2018},
  pages={1-8}
}
Elastic distortion of fingerprints has a negative effect on the performance of fingerprint recognition systems. [...] Key Method In this paper, we develop a rectification model based on a Deep Convolutional Neural Network (DCNN) to accurately estimate distortion parameters from the input image. Using a comprehensive database of synthetic distorted samples, the DCNN learns to accurately estimate distortion bases ten times faster than the dictionary search methods used in the previous approaches. Evaluating the…Expand
ID Preserving Generative Adversarial Network for Partial Latent Fingerprint Reconstruction
TLDR
Experimental results of matching reconstructed samples in both latent-to-sensor and latent- to-latent frameworks indicate that the proposed method significantly increases the matching accuracy of the fingerprint recognition systems for the latent samples. Expand
Deep Contactless Fingerprint Unwarping
TLDR
A deep model to rectify the perspective distortion of contactless fingerprints by combining a rectification and a ridge enhancement network is proposed and achieves the equal error rate and Rank-1 accuracy on the challenging dataset of ‘2D/3D’ fingerprints. Expand
Q-Learning Fingerprint Representations Method for Matching Low Quality Fingerprint Features
TLDR
A new machine learning method can be utilized for minutiae detection in low quality fingerprint images using multi-layer perceptions to improve image quality an most advantageous policy. Expand
Dense Fingerprint Registration via Displacement Regression Network
TLDR
This work proposes a dense fingerprint registration algorithm through convolutional neural network that can regress pixel-wise displacement field directly from coarsely aligned fingerprint images and proposes a multi-scale matching score fusion method to show the application of the proposed registration algorithm in improving fingerprint matching accuracy. Expand
FDeblur-GAN: Fingerprint Deblurring using Generative Adversarial Network
TLDR
A fingerprint deblurring model FDeblur-GAN, based on the conditional Generative Adversarial Networks (cGANs) and multi-stage framework of the stack GAN, which achieves an accuracy of 95.18% on the fingerprint database for the task of matching deblurred and ground truth fingerprints. Expand
Fingerprint Pre-Alignment based on Deep Learning
TLDR
A deep learning approach for absolute prealignment of fingerprints is presented and is applied as a preprocessing step in a Fuzzy Vault scheme, achieving higher accuracy w. Expand
White-Box Evaluation of Fingerprint Matchers: Robustness to Minutiae Perturbations
TLDR
Results of a controlled, white-box evaluation of one open-source and two commercial-off-the-shelf minutiae-based matchers in terms of their robustness against controlled perturbations introduced into the inputMinutiae feature sets reveal that the performance of fingerprintminutiae matchers are more susceptible to non-linear distortion and missing minutAE than spurious minutae and small positional displacements of the minutia locations. Expand
Macthing berbasis Average Euclidean Distance
The fingerprint-based biometric system is one of the safety standard in office and used to check employee attendancee. The method that commonly used in fingerprints detection, is image extractionExpand
Fingerprint Feature Extractors Using Multiple Machine Learning: A Comparative In The Study
Using biometrics information of personal identification has recognized more and more attention in the last few years, due to the inevitability to improve the biometric information security and accessExpand
Quality Guided Sketch-to-Photo Image Synthesis
TLDR
This work proposes a novel approach that adopts a generative adversarial network that synthesizes a single sketch into multiple synthetic images with unique attributes like hair color, sex, etc and incorporates multiple attribute assignment to the generator without compromising the identity of the synthesised image. Expand
...
1
2
3
...

References

SHOWING 1-10 OF 42 REFERENCES
Detection and Rectification of Distorted Fingerprints
TLDR
Novel algorithms to detect and rectify skin distortion based on a single fingerprint image are proposed and results have been obtained on three databases containing many distorted fingerprints, namely FVC2004 DB1, Tsinghua Distorted fingerprint database, and the NIST SD27 latent fingerprint database. Expand
Detection and rectification of distorted fingerprints
TLDR
A dictionary based orientation field estimation approach is used to recognize latent fingerprints that is captured using ancient fingerprint sensing techniques to take advantage of stronger prior knowledge of fingerprints so as to further improve the performance. Expand
Generative convolutional networks for latent fingerprint reconstruction
TLDR
Generative convolutional networks are applied to denoise visible minutiae and predict the missing parts of the ridge pattern of the fingerprint ridge pattern in order to reduce false positive and negative detection rates. Expand
Detecting fingerprint distortion from a single image
  • Xuanbin Si, J. Feng, Jie Zhou
  • Computer Science
  • 2012 IEEE International Workshop on Information Forensics and Security (WIFS)
  • 2012
TLDR
A study on fingerprint distortion is conducted and an algorithm to detect fingerprint distortion from a single image which is captured using traditional fingerprint sensing techniques is developed, based on analyzing ridge period and orientation information. Expand
Fingerprint enhancement using STFT analysis
TLDR
A new approach for fingerprint enhancement based on short time Fourier transform (STFT) Analysis is introduced and the algorithm simultaneously estimates all the intrinsic properties of the fingerprints such as the foreground region mask, local ridge orientation and local ridge frequency. Expand
A Comparative Study of Fingerprint Image-Quality Estimation Methods
TLDR
In this work, existing approaches for fingerprint image-quality estimation are reviewed, including the rationale behind the published measures and visual examples showing their behavior under different quality conditions, and a selection offinger image- quality estimation algorithms are tested. Expand
Fingerprint warping using ridge curve correspondences
TLDR
Results based on experimental data consisting of 1,600 fingerprints corresponding to 50 different fingers collected over a period of two weeks show that incorporating the proposed deformation model results in an improvement in the matching performance. Expand
Dynamic behavior analysis in compressed fingerprint videos
TLDR
This work proposes the use of fingerprint video sequences to investigate detecting two aspects of the dynamic behavior of fingerprints, and describes a new concept called the "resultant biometrics", a new type of biometric which has both a physiological, physical, and temporal component, added by a subject to an existing biometric. Expand
Improved Fingerprint Matching by Distortion Removal
TLDR
The novel approach presented here corrects distortions in fingerprints that have already been acquired, and results are presented demonstrating significant improvements in matching accuracy through the application of the technique. Expand
Orientation Field Estimation for Latent Fingerprint Enhancement
TLDR
Experimental results on the challenging NIST SD27 latent fingerprint database and an overlapped latent fingerprints database demonstrate the advantages of the proposed orientation field estimation algorithm over conventional algorithms. Expand
...
1
2
3
4
5
...