Fingerprint liveness detection based on histograms of invariant gradients

2016

J Image Video Proc.

We consider the task of image decomposition, and we introduce a new model coined directional global three-part decomposition (DG3PD) for solving it. As key ingredients of the DG3PD model, we introduce a discrete multi-directional total variation norm and a discrete multi-directional G-norm. Using these novel norms, the proposed discrete DG3PD model can decompose an image into two or three parts. Existing models for image decomposition Advantages of the DG3PD model over existing ones lie in the properties enforced on the cartoon and texture images. The geometric objects in the cartoon image have a very smooth surface and sharp edges. The texture image yields oscillating patterns on a defined scale which are both smooth and sparse. Moreover, the DG3PD method achieves the goal of perfect reconstruction by summation of all components better than the other considered methods. Relevant applications of DG3PD are a novel way of image compression as well as feature extraction for applications such as latent fingerprint processing and optical character recognition.

Section: Discussionmentioning

confidence: 99%

Directional global three-part image decomposition

Thai

2016

J Image Video Proc.

“…We conclude that orientation fields contain a high amount of information which is useful for detecting altered fingerprints. An advantage of the HIG descriptor [13] is that it can be computed very fast (in a few milliseconds per image) and it contributes to both alteration detection and liveness detection. The three features MDA, MOA and MH use the minutiae template as input.…”

Section: Discussionmentioning

confidence: 99%

“…HIG Histograms of invariant gradients (HIG) [13] also take the orientation field into account and image gradient directions are computed relative to the local orientation. For normal fingerprints, the majority of gradients form an angle of approximately 90 degrees with the local orientation.…”

Section: A Proposed Featuresmentioning

confidence: 99%

Improving fingerprint alteration detection

2015 9th International Symposium on Image and Signal Processing and Analysis (ISPA)

Mikaelyan

Olsen³

et al. 2015

Self Cite

Abstract-Fingerprint alteration is a type of presentation attack in which the attacker strives to avoid identification, e.g. at border control or in forensic investigations. As a countermeasure, fingerprint alteration detection aims to automatically discover the occurrence of such attacks by classifying fingerprint images as 'normal' or 'altered'. In this paper, we propose four new features for improving the performance of fingerprint alteration detection modules. We evaluate the usefulness of these features on a benchmark and compare them to four existing features from the literature.

“…for fingerprint alignment [1], singular point detection, fingerprint classification, image enhancement by contextual filtering [2,3] or descriptor matching. Moreover, OFs are applied for absolute pre-alignment in fingerprint cryptosystems [4], they are used to compute histograms of invariant gradients (HIG) [5] for fingerprint liveness detection and for improving fingerprint recognition performance by score revaluation [6].…”

Section: Introductionmentioning

confidence: 99%

Performance evaluation of fingerprint orientation field reconstruction methods

Oehlmann

Huckemann

3rd International Workshop on Biometrics and Forensics (IWBF 2015)

2015

Self Cite

Orientation fields (OFs) are a key element of fingerprint recognition systems. They are a requirement for important processing steps such as image enhancement by contextual filtering, and typically, they are estimated from fingerprint images. If information about a fingerprint is available only in form of a stored minutiae template, an OF can be reconstructed from this template up to a certain degree of accuracy. The reconstructed OF can then be used e.g. for fingerprint alignment or as a feature for matching, and thus, for improving directly or indirectly the recognition performance of a system. This study compares reconstruction methods from the literature on a benchmark with ground truth orientation fields. The performance of these methods is evaluated using three metrics measuring the amount of reconstruction errors as well as in terms of computational runtime.