Fei Su

Fingerprint Matching With Rotation-Descriptor Texture Features

A novel texture correlation matching method for fingerprint verification using Fourier-Mellin descriptor and phase-only correlation function is proposed in this paper. Fourier-Mellin descriptor correlation is used to align the template and query fingerprint images and a matching score is obtained. Matching takes about 1 second in Celeron 2.0 GHz processor, and the experimental results show that EER is 3.8%; fusion with minutia matching gets a better result

Ridge-based fingerprint recognition

A new fingerprint, matching method is proposed in this paper, with which two fingerprint skeleton images are matched directly. In this method, an associate table is introduced to describe the relation of a ridge with its neighbor ridges, so the whole ridge pattern can be easily handed. In addition, two unique similarity measures, one for ridge curves, another for ridge patterns, are defined with the elastic distortion taken into account. Experiment results on several databases demonstrate the effectiveness and robustness of the proposed method.

A robust fingerprint minutiae matching algorithm based on the support model

A novel method to match two minutiae point-sets based on the support model we proposed is presented in this paper. In this method, coarse matching is first performed from a number of seeds, and the results are then fused to obtain a constrained corresponding relationship of minutiae in two point-sets. By using the support degree of the elements in the constrained relations, the one-to-one correspondence is finally determined by comparison of similarity of local structures. Experiments show that this algorithm is robust and can deal with the translation, rotation, distortion and outlier problems well.

Fingerprint Registration Using Minutia Clusters and Centroid Structure 1

In this paper a novel distortion-tolerant fingerprint registration method based on clustering is proposed. In this method, minutiae features of the query fingerprint are divided into various clusters. Several local structure transformations are estimated by local structure sets. Then the global structures (centroid structures) are constructed according to the local structure transformation. The global transformation is determined by the score of local structure transformation together with the similarity level of the global structure. Experimental results show that this algorithm is robust for aligning fingerprints with a small number of minutia and heavy distortions. Such situations are often encountered in forensic applications

Robust ridge following in fingerprints

In this paper, we presented an improved approach of minutiae detection by following the ridges Our algorithm is based on the two characteristics of the ridge points on the same ridge: connectivity and randomicity Because our algorithm takes full advantage of the characteristics of the ridge points, it is robust to noise Using our algorithm, good quality ridge images can be obtained which can be used to assist minutiae-based fingerprint matching, or directly used for fingerprint matching Experimental results are included.

A hierarchical fingerprint matching method based on rotation invariant features

A hierarchical matching algorithm based on rotation invariant features to align two fingerprints is proposed in this paper The translation and rotational offsets are extracted using the maximum overlapped fingerprint images The proposed method can reduce the searching space in alignment, and what is more attractive is that it obviates the need for extracting minutiae points or the core point to align the fingerprint images Experimental results show that the proposed method is more robust than using the reference point or using the minutiae to align the fingerprint images.

An adaptive fingerprint post-processing algorithm based on mathematical morphology

In this paper, an adaptive post-processing method using mathematical morphology combined with analyzing the properties of each candidate minutia based on the gray-level image, binary image, local ridge spacing and local orientation is presented to decide whether the minutia is false or true and to eliminate the false one The experiment results demonstrate the effectiveness to reduce the number of false minutiae encountered and improve the thinning fingerprint images at the same time.