Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Finger vein recognition method based on residual attention mechanism

A recognition method, finger vein technology, applied in neural learning methods, character and pattern recognition, computer parts and other directions, can solve the complex extraction process, high image quality requirements, model design image data volume does not match model parameters, etc. problems, to achieve the effect of accelerating network training, improving accuracy, and enhancing feature extraction and representation capabilities

Active Publication Date: 2021-06-11
CHANGCHUN UNIV OF TECH
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional recognition generally includes preprocessing, feature enhancement, feature extraction, and feature matching. Vein features are extracted through human intervention or manual design. Such methods require more processing work on the image in the early stage, and the extraction process is often complicated. Higher; the method based on deep learning can effectively reduce the steps of image processing, and only needs simple preprocessing, but the model design and the mismatch between the amount of image data and the model parameters are the difficulties of this type of research

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
  • Finger vein recognition method based on residual attention mechanism
  • Finger vein recognition method based on residual attention mechanism
  • Finger vein recognition method based on residual attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] The data set used in this example is FV-USM, which was created by the Bakhtiar team of University Sains Malaysia. The data set contains a total of 492 fingers of different types, and each finger is divided into two stages to collect 6 images. The original image size is 640×480, and the “jpg” format is as follows: figure 1 shown.

[0035] This embodiment proposes a finger vein recognition method based on a residual attention mechanism, including the following method steps:

[0036] Training a Finger Vein Image Classification Network Based on Residual Attention Mechanism

[0037] S1, for example image 3 The original finger vein image shown in (a) performs edge detection, determines the upper and lower edges of the finger, and obtains a finger edge line with a single pixel width;

[0038] The specific acquisition process of the single-pixel-wide finger edge line is as follows:

[0039] S11, such as image 3 As shown in (b), the Gaussian filter is used to denoise the ...

Embodiment 2

[0075] Based on the same inventive concept as the method described in Embodiment 1, this embodiment proposes a system based on the finger vein recognition method based on the residual attention mechanism, which includes:

[0076] Model training module for training finger vein image classification network based on residual attention mechanism

[0077] The finger vein recognition module is used to input the obtained original finger vein image into the trained finger vein image classification network based on the residual attention mechanism, and then output the recognition result.

Embodiment 3

[0079] Based on the same inventive concept as the method described in Embodiment 1, this embodiment proposes a terminal, including a processor, a memory, and a program of a finger vein recognition method based on a residual attention mechanism stored in the memory. When the program of the finger vein recognition method based on the residual attention mechanism is run by the processor, the steps of the finger vein recognition method based on the residual attention mechanism as described in Embodiment 1 are realized.

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 provides a finger vein recognition method based on a residual attention mechanism, and the method comprises the following steps: carrying out the edge detection of an original finger vein image, obtaining a single-pixel wide finger edge line, and carrying out the rotation correction of the original finger vein image; obtaining a finger vein foreground image which does not contain background information; determining an ROI of the finger vein; repeating the above steps on the finger vein image samples in each category of the data set to obtain the ROI of each finger vein image; performing training set and test set division on the processed finger vein data set; carrying out fusion feature extraction by using an improved residual attention network; training the improved residual attention network to obtain a finger vein image classification network; and inputting the obtained original finger vein image into the trained finger vein image classification network based on the residual attention mechanism to output a recognition result.

Description

technical field [0001] The invention belongs to the technical field of image processing and biological recognition, and in particular relates to a finger vein recognition method based on a residual attention mechanism. Background technique [0002] In the era of integrated development of digitalization and informationization, people's awareness of personal information protection is gradually strengthening. Traditional identity authentication technology, such as physical objects such as magnetic cards, certificates, and identity tags, has problems such as easy loss, easy theft, and easy copying, and can no longer meet the high security and high confidentiality needs of individuals for identity information. Not only that, traditional identity authentication items store limited information, and in an information environment based on big data, such limited data means high risk. [0003] As a product of the information age, biometric recognition technology uses inherent physical...

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
IPC IPC(8): G06K9/00G06N3/04G06N3/08
CPCG06N3/084G06V40/1347G06V40/1365G06N3/047G06N3/045
Inventor 鲁慧民刘伟业王一凡李玉鹏马宁
Owner CHANGCHUN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com