Minutia Tensor Matrix: A New Strategy for Fingerprint Matching

Fu, Xiang; Feng, Jufu

doi:10.1371/journal.pone.0118910

Cited by 9 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Minutiae based approaches usually fit a thin-plate spline model with paired minutiae as landmark points. The correspondences between minutiae are established by considering both similarity of minutiae descriptors and compatibility between minutiae pairs [27,28]. Cao et al [7] introduced the compatibility between minutiae triplets to further eliminate false correspondences.…”

Section: A Global Registrationmentioning

confidence: 99%

Latent Fingerprint Registration via Matching Densely Sampled Points

Feng

et al. 2021

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

Latent fingerprint matching is a very important but unsolved problem. As a key step of fingerprint matching, fingerprint registration has a great impact on the recognition performance. Existing latent fingerprint registration approaches are mainly based on establishing correspondences between minutiae, and hence will certainly fail when there are no sufficient number of extracted minutiae due to small fingerprint area or poor image quality. Minutiae extraction has become the bottleneck of latent fingerprint registration. In this paper, we propose a non-minutia latent fingerprint registration method which estimates the spatial transformation between a pair of fingerprints through a dense fingerprint patch alignment and matching procedure. Given a pair of fingerprints to match, we bypass the minutiae extraction step and take uniformly sampled points as key points. Then the proposed patch alignment and matching algorithm compares all pairs of sampling points and produces their similarities along with alignment parameters. Finally, a set of consistent correspondences are found by spectral clustering. Extensive experiments on NIST27 database and MOLF database show that the proposed method achieves the state-of-the-art registration performance, especially under challenging conditions.

show abstract

Section: A Global Registrationmentioning

confidence: 99%

Latent Fingerprint Registration via Matching Densely Sampled Points

Feng

et al. 2021

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

show abstract

“…(2) Feature point extraction : first, the average direction field difference is calculated, 9 and the calculation formula is:

P false(i, j false) = \frac{\int_{0}^{2 π} \frac{\partial}{\partial} O^{'} false(i + ε \cos θ, j + ε \sin θ false) italicd θ}{2 π},

where P ( i , j ) is the average direction field difference of pixels, when it is 0.5, the pixel is the center point, and when it is −0.5, the pixel is the triangle point; O ′ (•) is the direction field, ε is the radius of the circle around, and θ is the angle of the circle around. The endpoint and cross point of fingerprint feature points are determined by calculating the sum of the absolute value of the gray difference between the two pixels around the point.…”

Section: Traditional Point Matching Fingerprint Recognitionmentioning

confidence: 99%

Feature extraction, recognition, and matching of damaged fingerprint: Application of deep learning network

2020

Concurrency and Computation

View full text Add to dashboard Cite

With the improvement of informatization and the development of computer technology, identity recognition with fingerprint has been the growing trend, but the stained fingerprint will make recognition difficult. To solve the above problem, this paper briefly introduced the traditional point matching fingerprint recognition algorithm and the damaged fingerprint recognition method based on the convolution neural network (CNN). Then in MATLAB software, the damaged fingerprint recognition method based on CNN was simulated and compared with the traditional point matching recognition method and traditional CNN recognition method. The results showed that the improved CNN recognition method iterated fewer times and achieved smaller training error during the training; the improved CNN method had lower false recognition rate and rejection rate in identifying the unknown fingerprint; the traditional point matching method spent the most time in identifying the unknown fingerprint, followed by the traditional CNN method and the improved CNN recognition method. This study used CNN for fingerprint recognition and improved CNN to improve its recognition accuracy for the damaged fingerprint. The improved CNN can effectively enhance the recognition accuracy of the damaged fingerprint, which provides a useful reference for the improvement of the accuracy and applicability of the fingerprint recognition system.

show abstract

“…This combined approach is defined on image in threedimensional space and its solution degenerated to minimum weighting of bipartite graph. Also, a graph-based approach was studied using Minutia Tensor Matrix (MTM) for fingerprint matching [12]. This approach is designed to address both local and global matching of fingerprint minutiae based on similarities and compatibilities.…”

Section: Literature Reviewmentioning

confidence: 99%

Optimisation of Point-Set Matching Model for Robust Fingerprint Verification in Changing Weather Conditions

Udo

Akhigbe

Afolabi

2019

AJRCoS

View full text Add to dashboard Cite

Aims: To provide a baseline for the configuration of Automated Fingerprint Verification System (AFVS) in the face of changing weather and environmental conditions in order to ensure performance accuracy. Study Design: Statistical and theoretical research approaches. Place and Duration of Study: Department of Computer Science and Engineering, Obafemi Awolowo University, Ile-Ife, Nigeria, between July 2017 and July 2018. Methodology: Data set were collected in the South-South geopolitical zone of Nigeria. We use 10,000 minutiae points defined by location and orientation features extracted from fingerprint samples obtained at 9 various physical and environmental conditions over 12 months period. These data were used to formulate linear regression models that were used as constraints to the verification objective function derived as constrained linear least squares. The effects of the changing weather and environmental conditions were incorporated into the optimised point-set matching model in order to minimise the total relative error on location and orientation differences between pairs of minutiae. The model was implemented using interior-point convex quadratic programming was implemented in Matlab. Results: The results obtained from the optimisation function by adjusting the thresholds of the effects of weather and environmental conditions to 0.0, 0.0 for location and orientation properties of minutiae, respectively, showed minimal total relative errors on the corresponding pairs of matched minutiae, when compared with using the default threshold values of the selected conditions. Conclusion: The optimisation of point-set based model could provide a computational basis for accurate fingerprint verification for low and high-security AFVS in unfavourable conditions if they are incorporated into the matching model. However, further validation and evaluation of the model with data sets from regions with similar weather and environmental conditions is needed to further validate its robustness in terms of performance accuracy

show abstract

Minutia Tensor Matrix: A New Strategy for Fingerprint Matching

Cited by 9 publications

References 26 publications

Latent Fingerprint Registration via Matching Densely Sampled Points

Latent Fingerprint Registration via Matching Densely Sampled Points

Feature extraction, recognition, and matching of damaged fingerprint: Application of deep learning network

Optimisation of Point-Set Matching Model for Robust Fingerprint Verification in Changing Weather Conditions

Contact Info

Product

Resources

About