Frequency Map by Structure Tensor in Logarithmic Scale Space and Forensic Fingerprints

Bigün, Josef; Mikaelyan, Anna

doi:10.1109/cvprw.2016.32

Cited by 5 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As future work we plan to investigate additional features, e.g. ridge frequency maps obtained by the structure tensor [12] or by curved regions [20]. Orientation maps obtained by the structure tensor are complex valued with angular information representing orientation and magnitude defining reliability.…”

Section: Discussionmentioning

confidence: 99%

“…The four best performing features so far are OFA [7], DOFTS [12], COH [14] and SPDA [8]. The two best performing features OFA and DOFTS are based on differences between orientation fields.…”

Section: Discussionmentioning

confidence: 99%

“…The parameters σ in and σ out define the inner and the outer scales of the structure tensor. Recently, a frequency map estimation has been suggested [12]. In combination with orientation maps above it is used to enhance the image and then reestimate both the orientation and the frequency maps iteratively.…”

Section: A Proposed Featuresmentioning

confidence: 99%

See 2 more Smart Citations

Improving fingerprint alteration detection

Gottschlich

Mikaelyan

Olsen³

et al. 2015

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

Self Cite

View full text Add to dashboard 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.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Improving fingerprint alteration detection

Gottschlich

Mikaelyan

Olsen³

et al. 2015

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

Self Cite

View full text Add to dashboard Cite

show abstract

“…This can be provided interactively by forensic expert. However, we did not assume that but used instead a recent automatic method of enhancing orientation together with absolute frequency which was available to us, [10]. It utilizes the theory of Structure Tensor (ST) in logarithmic scale space to estimate absolute frequency and orientation independently.Iterative mutual improvement via enhancement …”

Section: Preprocessing: Dense Orientation and Frequency Mapsmentioning

confidence: 99%

“…It was established in [10], that frequency and orientation are connected through the trace of structure tensor I …”

Section: B Frequency Imagementioning

confidence: 99%

SAFE features for matching fingermarks by neighbourhoods of single minutiae

Mikaelyan

Bigün

2014

2014 14th International Symposium on Communications and Information Technologies (ISCIT)

Self Cite

View full text Add to dashboard Cite

Abstract-Symmetry Assessment by Finite Expansion (SAFE) is a novel description of image information by means of Generalized Structure Tensor. It represents orientation data in neighbourhood of key points projected onto the space of harmonic functions creating a geometrically interpretable feature of low dimension. The proposed feature has built in quality metrics reflecting accuracy of the extracted feature and ultimately the quality of the key point. The feature vector is orientation invariant in that it is orientation steerable with low computational cost. We provide experiments on minutia key points of forensic fingerprints to demonstrate its usefulness. Matching is performed based on minutia in regions with high orientation variance, e.g. in proximity of core points. Performance of single matching minutia equals to 20% EER and Rank-20 CMC 69% on the only publicly available annotated forensic fingerprint SD27 database.Further, we complement SAFE descriptors of orientation maps with SAFE descriptors of frequency features in a similar manner. In case of combined features the performance is improved further to 19% EER and 74% Rank-20 CMC.

show abstract

Perfect fingerprint orientation fields by locally adaptive global models

2016

View full text Add to dashboard Cite

Fingerprint recognition is widely used for verification and identification in many commercial, governmental and forensic applications. The orientation field (OF) plays an important role at various processing stages in fingerprint recognition systems. OFs are used for image enhancement, fingerprint alignment, for fingerprint liveness detection, fingerprint alteration detection and fingerprint matching. In this paper, a novel approach is presented to globally model an OF combined with locally adaptive methods. We show that this model adapts perfectly to the 'true OF' in the limit. This perfect OF is described by a small number of parameters with straightforward geometric interpretation. Applications are manifold: Quick expert marking of very poor quality (for instance latent) OFs, high fidelity low parameter OF compression and a direct road to ground truth OFs markings for large databases, say. In this contribution we describe an algorithm to perfectly estimate OF parameters automatically or semi-automatically, depending on image quality, and we establish the main underlying claim of high fidelity low parameter OF compression.

show abstract

Frequency Map by Structure Tensor in Logarithmic Scale Space and Forensic Fingerprints

Cited by 5 publications

References 15 publications

Improving fingerprint alteration detection

Improving fingerprint alteration detection

SAFE features for matching fingermarks by neighbourhoods of single minutiae

Perfect fingerprint orientation fields by locally adaptive global models

Contact Info

Product

Resources

About