Have a personal or library account? Click to login
Research on Iris Feature Extraction and Recognition Technology Based on Deep Learning Cover

Research on Iris Feature Extraction and Recognition Technology Based on Deep Learning

By: Yufei Chen,  Yiyang Zhao,  Bing Zhao and  Hao Wei  
Open Access
|Mar 2024

Figures & Tables

Figure. 1.

The human eye
The human eye

Figure. 2.

Flowchart of iris recognition
Flowchart of iris recognition

Figure. 3.

Example of two iris images
Example of two iris images

Figure. 4.

Flow of LMD improvement algorithm
Flow of LMD improvement algorithm

Figure. 5.

Basic flow of Faster R-CNN model
Basic flow of Faster R-CNN model

Figure. 6.

U-net network structure
U-net network structure

Figure. 7.

CNN recognition result rate
CNN recognition result rate

Figure. 8.

Code Run Diagram
Code Run Diagram

Figure. 9.

Front-end page display
Front-end page display

Figure. 10.

Selecting the iris image to be matched against the image in the database
Selecting the iris image to be matched against the image in the database

Figure. 11.

Image matching demonstration
Image matching demonstration

Figure. 12.

Interface display
Interface display

Figure. 13.

Edge extraction to obtain iris
Edge extraction to obtain iris

Figure. 14.

Image after normalization and feature extraction
Image after normalization and feature extraction

Figure. 15.

Matching successful image
Matching successful image

Analysis of experimental data

Test MethodsTraining SetTest SetNumber Of Correct IdentificationsRecognition Rate % (Crr)
CNN400605692
LMD400604778

Comparison of the content of the two datasets

Training SetTest SetTraining SetTest Set
Number of categories508456
Number of images/classes20201010
Total number of images3425834258
Language: English
Page range: 35 - 45
Published on: Mar 15, 2024
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2024 Yufei Chen, Yiyang Zhao, Bing Zhao, Hao Wei, published by Xi’an Technological University
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.