Palmprint identification method and palmprint identification system

Abstract

The invention provides a palmprint identification method and a palmprint identification system. The palmprint identification method comprises the steps of acquiring at least two palmprint images with the same size through a mobile terminal, respectively acquiring a plurality of feature points in the two palmprint images, placing the two palmprint images in parallel, calculating one or more pairs of features points with the Euclidean distance between the feature points selected from one of the palmprint image and the feature points selected from the other palmprint image being less than a preset value to act as associated feature point pairs, and thus translating the two palmprint images until two feature points in each feature point pair correspond to each other; enabling the two palmprint images to relatively rotate for a preset number of times according to the feature point pairs, wherein the two palmprint images rotate by a preset angle at each time; performing a Gabor wavelet algorithm after each time of rotation so as to carry out comparison on the feature point pairs and output a score; and judging the similarity of the two palmprint images according a maximum value of the scores, thereby overcoming rotation and translation interference so as to apply a palmprint technology to the mobile terminal.

Description

technical field [0001] The invention relates to the technical field of palmprint recognition, in particular to a palmprint recognition method and system. Background technique [0002] Palmprint recognition is an emerging identification technology, which has the advantages of easy collection and rich texture. A complete palmprint recognition system includes several steps such as palmprint image acquisition, preprocessing, feature extraction, and feature classification and matching. The current palmprint recognition algorithm is roughly divided into the following categories according to the method of feature extraction: based on geometric features, based on transform domain features, and based on algebraic statistical features. The corresponding methods include PCA, Gabor wavelet, Fourier transform, etc. . Among them, Gabor wavelet has been widely used in the field of palmprint recognition because of its high discrimination rate and simple matching. [0003] Although the Ga...

AI Technical Summary

Problems solved by technology

[0003] Although the Gabor wavelet has a high degree of discrimination, due to the limitation of the algorithm, it has strict requirements on the position of the palmprint, which greatly limits the application range of the algorithm.

For mobile devices such as mobile phones and ipads, when collecting palmprints, offset and rotation are unavoidable, and the recognition error rate is high

Images