An accurate fingerprint reference point determination method based on curvature estimation of separated ridges

Abstract

This paper presents an effective method for the detection of a fingerprint’s reference point by analyzing fingerprint ridges’ curvatures. The proposed approach is a multi-stage system. The first step extracts the fingerprint ridges from an image and transforms them into chains of discrete points. In the second step, the obtained chains of points are processed by a dedicated algorithm to detect corners and other points of highest curvature on their planar surface. In a series of experiments we demonstrate that the proposed method based on this algorithm allows effective determination of fingerprint reference points. Furthermore, the proposed method is relatively simple and achieves better results when compared with the approaches known from the literature. The reference point detection experiments were conducted using publicly available fingerprint databases FVC2000, FVC2002, FVC2004 and NIST

Orientation Local Binary Pattern Based Fingerprint Matching

“…Each database contains 800 images from 100 individuals and eight fingerprint impression per individual at different rotations. The performance of the proposed algorithm is measured by means of Genuine Acceptance Rate (GAR) and Equal Error Rate (EER) [30]. GAR is defined as the rate at which true claims are accepted; whereas, EER is the rate at which the frequency of False Acceptance Rate (FAR) is equal to the frequency of False Rejection Rate (FRR) defined as follows:…”

“…A reference point as the point of maximum curvature on the innermost ridge is detected, which can either be a core point or any singular point excluding delta points (the delta point is a pattern of a fingerprint that resembles the Greek letter delta) which are generally located near the boundary of the images and, therefore, do not allow the extraction of ROI. The present work employed the methods described in [29][30][31] for reliable extraction of reference point. This method detects a point with the highest curvature using complex filtering.…”

Integration of Biometrics and Steganography: A Comprehensive Review

“…Fingerprint [11,22], facial recognition [23], and iris [20]/retina detail are all physiological biometric parameters. Other physiological parameters available in the field of biometrics include vein recognition [24], ear recognition [25], palm print recognition [26], hand geometry [27], finger geometry [28], lip furrows [29,30], DNA [31], and odour/scent recognition [32].…”

The Method of Automatic Knuckle Image Acquisition for Continuous Verification Systems

“…Such indications should be treated as outliers. To eliminate outliers, a method described in [23] was used. In this method, for each point p i (x, y), its parameter D i is determined.…”

“…Outliers should not be taken into account when determining the average value. In the proposed method, the point p i (x, y) is removed from the set P, if the value D i determined for this point is greater than 0.39 [23]. All remaining points are put into set O.…”