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

Iris positioning segmentation system based on cavity residual attention structure

A technology of iris positioning and attention, applied in the direction of neural architecture, instruments, biological neural network models, etc., to achieve the effect of realizing the quality of iris area

Active Publication Date: 2019-07-26
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF6 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The technical problem to be solved by the present invention is to provide an iris image segmentation system aimed at solving iris positioning and segmentation under complex conditions

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
  • Iris positioning segmentation system based on cavity residual attention structure
  • Iris positioning segmentation system based on cavity residual attention structure
  • Iris positioning segmentation system based on cavity residual attention structure

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] 1. Structural design

[0028] In order to better perform iris segmentation, this paper proposes a spatial attention mechanism to improve segmentation performance. The specific attention structure is as follows: figure 2 shown. In order to avoid the problems caused by direct channel compression, the attention mechanism proposed in this paper first performs global pooling on the feature map, for a dimension of After the feature tensor of the global pooling is obtained, the dimension is , and then use the two-layer fully connected network to obtain the channel mask vector. The two-layer fully connected network implements the mapping process. After the channel mask is obtained, the channel-weighted feature map is obtained by multiplying the corresponding channels, and then the feature map channel is compressed to 1 using a convolution operation with a convolution kernel of 1×1. Expand the compressed single-channel feature map into a vector form to obtain a spatia...

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 an iris positioning segmentation system based on a cavity residual attention structure. The iris positioning segmentation system comprises a feature extraction structure, a segmentation structure and a scoring structure. The feature extraction structure comprises M feature extraction modules which are connected in series step by step, and output of an Mth feature extractionmodule is connected with input of the scoring structure and input of the segmentation structure. The segmentation structure includes M-1 upsampling modules and M-2 feature fusion modules arranged in cascade. Output of an i-th feature extraction module is input to an M-i-level feature fusion module. The other input of a feature fusion module is connected to output of an upper sampling module. A M-1th upper sampling module outputs a mask image after cutting. The feature extraction structure comprises N cavity attention residual structure DARB and 1 downsampled hole attention residual structure DADRB. The DARB is used for building the neural network. Accurate and rapid iris area division is achieved by designing the multi-branch network, and automatic iris area quality evaluation is achieved.

Description

technical field [0001] The invention belongs to the field of digital image processing and machine learning, and is mainly used for iris image detection and segmentation. Background technique [0002] Iris recognition is a biometric recognition technology that uses the characteristic points formed by the internal fibrous tissue of the biological iris to verify and identify people's identities. Because the iris has high stability, uniqueness, anti-counterfeiting and other excellent properties. Compared with other biometric identification methods, iris verification has higher recognition accuracy and is therefore more widely used. [0003] The iris recognition process generally consists of preprocessing, iris location, feature extraction, and feature matching. The iris location and segmentation algorithm is the most important part of iris recognition, and the inaccurate iris location area will directly affect the accuracy of subsequent feature extraction, resulting in a decli...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/34G06N3/04
CPCG06V40/18G06V10/267G06N3/045
Inventor 解梅赵雷廖炳焱钮孟洋
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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