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

DAISY-characteristic-based deformable iris matching method

A technology of local features and texture images, applied in computer parts, instruments, characters and pattern recognition, etc., can solve problems such as easy matching errors, reduce system performance, etc., to achieve high robustness, improve system performance, and high efficiency. Effect

Active Publication Date: 2013-05-01
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When the iris deformation is very serious, the eigenvalue pairs of the corresponding points are shifted, so the fixed point-to-point matching method is easy to match wrong eigenvalues, thereby reducing system performance

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
  • DAISY-characteristic-based deformable iris matching method
  • DAISY-characteristic-based deformable iris matching method
  • DAISY-characteristic-based deformable iris matching method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment example 1

[0068] Implementation Case 1: The deformable iris matching method based on local features of texture images is suitable for handheld iris recognition devices.

[0069] The invention can be widely applied to hand-held iris recognition systems. Some handheld devices have their own light source. When the user uses the device, the distance between the device and the human eye changes the light that hits the eye. Therefore, the iris of the person will also deform to varying degrees with the light. When the iris is deformed, the false rejection rate will be greatly increased, that is, the user who has already registered will be rejected, which will seriously affect the recognition rate and convenience of the user using the device. Using the existing recognition algorithm, when using the device, the user needs to manually adjust the distance between the device and the eyes within the focal length range, so that the lighting conditions of each recognition are similar, so as to ensure ...

Embodiment example 2

[0070] Implementation case 2: The application of the deformed iris matching method based on the local features of the texture image in the long-distance iris recognition system.

[0071] The invention can be widely applied to long-distance iris recognition systems. In order to make the user more comfortable to use, the iris recognition system has gradually evolved into a form where the machine cooperates with the human. The human does not need to deliberately cooperate with the machine during the recognition, but the machine can adaptively recognize the height of the person and automatically obtain the iris image. However, when the remote device acquires the iris image, it is greatly disturbed by the external ambient light, and it is inconvenient to adjust the light intensity, so the elastic deformation of the iris will also interfere with the system. Existing methods cannot identify people well, and often reject registered people, which greatly reduces user experience. The p...

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 DAISY-characteristic-based deformable iris matching method. The method comprises two parts, namely characteristic extraction and key point matching, wherein the characteristic extraction part comprises the following steps: calculating a direction gradient diagram of each normalized iris image according to a plurality of set directions, and taking weighted arrays of different pixel point direction gradient diagrams as DAISY characteristics of the image; and the key point matching part comprises the following steps: selecting a key point according to the DAISY characteristics of the pixel points, and finding out points on another image closest to characteristic Euclidean distance of each key point through local dynamic searching, wherein the average of matching values of each key point is the matching result of two images. By the method, the process of characteristic extraction is quick, and the characteristics are robust to light and noise. The key point matching can dynamically find the matching points in a local area, is particularly suitable for deformable iris matching, and can improve the accuracy of low-quality iris image matching and the adaptabilityof an iris identification system to a complex application environment.

Description

technical field [0001] The invention relates to the technical fields of computer vision, digital image processing, pattern recognition, etc., in particular to a deformable iris matching method (Deformable DAISYMatcher, DDM for short) based on local features of texture images. Background technique [0002] The iris is an externally visible internal organ of the human eye, which can present rich textures under near-infrared light. Iris recognition is to identify people's identities by analyzing the texture of the iris of the human eye. Because the iris has many advantages such as uniqueness, stability and non-invasiveness, it has been successfully applied to identity verification in coal mines, airports, banks and other places. [0003] However, under different lighting environments, the pupil size of the human eye will scale with the light. According to statistics, when the external light is strong, the diameter of the pupil of the human eye will shrink to about 1.5 mm, and ...

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): G06K9/62G06K9/46
Inventor 谭铁牛孙哲南张曼
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
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