Unlock instant, AI-driven research and patent intelligence for your innovation.

A kind of 3D palmprint authentication method

A palmprint, 3D technology, applied in the field of palmprint recognition, to achieve good robustness, high accuracy, and simple effects

Active Publication Date: 2018-08-24
HEBEI UNIVERSITY
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a new 3D palmprint authentication method to solve the problem that the current existing methods are difficult to make full use of the useful information in the palmprint image for palmprint authentication and identification

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A kind of 3D palmprint authentication method
  • A kind of 3D palmprint authentication method
  • A kind of 3D palmprint authentication method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] follow below figure 1 The flow chart shown illustrates the invention in detail.

[0027] 1), data preprocessing

[0028] The sample bank used in this embodiment comes from the Hong Kong Polytechnic University. The samples in the sample bank are 4000 3D palmprint images (200 people × 2 hands / person × 10 3D palmprints / hands). The 3D palmprint image of a hand (that is, a user) is 1 category, then the sample database contains 400 categories of 3D palmprint images of 400 users, as follows:

[0029]

[0030] Among them, X represents a 3D palmprint image, its subscript represents the ranking among similar samples, and its superscript represents the category ranking of samples.

[0031] In addition, one of the 10 3D palmprint images of each user is selected and pre-stored in addition.

[0032] 1.1) Realize 3D image planarization by calculating curvature features:

[0033] According to the knowledge of differential geometry, the surface S (that is, the 3D palmprint image,...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a 3D palmprint authentication method. The method is to firstly preprocess the 3D palmprint and turn it into a one-dimensional image, and then calculate the relative coherence coefficient of the intra-class graph and the inter-class graph as a new sample, using The new sample trains the stacked denoising autoencoder, extracts the features of the new sample, and obtains the trained denoising autoencoder; then trains several weak classifiers based on the real adaboost algorithm and the new sample features extracted in the previous step , the combination of these weak classifiers constitutes a strong classifier; then the user inputs a new 3D palmprint image, compares it with the user's 3D palmprint image pre-stored in the sample library, and judges whether they belong to the same class or heterogeneous. The method of the present invention is simple and fast to implement, uses a coherent algorithm to calculate the consistency between images, and judges whether they are data samples of the same user, and solves the problem of small translation of palmprint images and the difficult problem of finding positioning information for registration algorithms. It has better robustness.

Description

technical field [0001] The invention relates to palmprint recognition technology, in particular to a 3D palmprint authentication method. Background technique [0002] Palmprint recognition is an important biometric technology. The palmprint image contains rich line features and texture features, which makes it of high application value in identity authentication. However, there are some problems in the traditional 2D palmprint recognition technology that are difficult to solve properly: (1) 2D palmprint is easily disturbed by factors such as light changes and skin surface smears. (2) Changes in image features caused by finger gesture changes. (3) The existing feature extraction methods are difficult to make full use of the useful information in the image. Recently, Zhang Dapeng of the Hong Kong Polytechnic University and others proposed a three-dimensional palmprint recognition technology, which can effectively overcome the interference of light changes, but the other two ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00
CPCG06V40/1347G06V40/1365
Inventor 刘明苗荣凡刘秀玲娄存广王光磊刘晓光
Owner HEBEI UNIVERSITY