What is it about?
Fingerprint verification is a widely used authentication method in commercial applications and most fingerprint verification systems rely on minutiae for com- paring two fingerprints. Typical steps of fingerprint image processing include segmentation, orientation field estimation, image enhancement by contextual filtering and minutiae extraction. Additionally, many systems include nowadays a software-based liveness detection module which can e.g. be based on histograms of invariant gradients as a countermeasure against so-called spoof attacks. In this paper, we focus on the fingerprint image segmentation step and we propose a global three-part decomposition (G3PD) method to achieve an accurate extraction of the foreground region.
Featured Image
Why is it important?
Image segmentation is a fundamental step of image processing. In this article, we build a mathematical model for an inverse problem in imaging, i.e. fingerprint image can be considered as the sum of three parts, namely a cartoon (smoothness), a texture (sparsity in the frequency domain), and small objects (noises). Performance evaluations show that our proposed G3PD method consistently outperforms existing methods in terms of segmenta- tion accuracy.
Perspectives
Read the Original
This page is a summary of: Global variational method for fingerprint segmentation by three-part decomposition , IET Biometrics, June 2016, the Institution of Engineering and Technology (the IET),
DOI: 10.1049/iet-bmt.2015.0010.
You can read the full text:
Contributors
The following have contributed to this page