Screw operator-based fingerprint detail point characteristic obtaining method

Abstract

The invention discloses a screw operator-based fingerprint detail point characteristic obtaining method, and relates to the technologies of fingerprint preprocessing and the like. The method comprises three steps of fingerprint preprocessing, phase field demodulation and detail point characteristic obtaining; a fingerprint image is regarded as a two-dimensional variable-cycle stripe graph signal; detail points in the fingerprint image are expressed as screw singular points in a phase field; and a fingerprint detail point detection problem is converted into a function solving problem, wherein phase demodulation is obtained by solving a single base signal through an analysis function, and the screw singular points are obtained by calculating extremum points obtained by Poincare index integral of the phase field. According to the method, the problem of information loss in operations of binarization, detailing and the like is avoided; very good noise resistance is achieved; the post-processing of removing fake detail points does not need to be carried out; the system time cost is effectively reduced; meanwhile, the influence caused by a fingerprint pressure, translation, rotation and a dry-wet environment during fingerprint image collection is avoided; and the method has the advantages of low complexity, high processing speed, high comparison precision and the like.

Description

technical field [0001] The method for acquiring fingerprint minutiae feature based on spiral operator of the present invention relates to fingerprint preprocessing technology, stripe pattern signal phase information acquisition technology and fingerprint minutiae feature extraction technology. Background technique [0002] Fingerprint, due to its stability, collectability, universality and uniqueness, has become synonymous with biological characteristics. Fingerprints refer to the uneven lines on the skin on the front of the end of human fingers. These lines are regularly arranged to form different patterns. Fingerprint image recognition system is a typical pattern recognition system, including fingerprint image acquisition, preprocessing, feature extraction and matching modules. Among them, the accuracy of fingerprint feature extraction has an important impact on the accuracy of fingerprint matching. In the fingerprint image, the minutiae points generated by the discontin...

AI Technical Summary

Problems solved by technology

Among them, the binarization process will cause the loss of part of the original information, affecting the accuracy of the comparison results

However, noises such as burrs, short lines, broken lines, false bridges, small holes, and islands in the fingerprint image will generate false minutiae points in the thinning process, and the post-processing operation of removing the false minutiae points will increase to a considerable extent. Identify system time costs

More importantly, the fingerprint image acquisition process is affected by factors such as finger pressure, translation, and rotation, which will cause nonlinear changes in the coordinates and directions of the final extracted minutiae points, so that the minutiae information extracted in the time domain cannot guarantee high reliability. The degree affects the identification efficiency of the fingerprint system

Images