CN108734085A - Iris identification method and iris authentication system - Google Patents
Iris identification method and iris authentication system Download PDFInfo
- Publication number
- CN108734085A CN108734085A CN201810257237.6A CN201810257237A CN108734085A CN 108734085 A CN108734085 A CN 108734085A CN 201810257237 A CN201810257237 A CN 201810257237A CN 108734085 A CN108734085 A CN 108734085A
- Authority
- CN
- China
- Prior art keywords
- iris
- image
- registration
- matrix
- identification method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000011218 segmentation Effects 0.000 claims abstract description 26
- 230000004927 fusion Effects 0.000 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims abstract description 13
- 239000011159 matrix material Substances 0.000 claims description 38
- 230000009466 transformation Effects 0.000 claims description 18
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 238000013459 approach Methods 0.000 claims description 11
- 238000003708 edge detection Methods 0.000 claims description 6
- 238000012804 iterative process Methods 0.000 claims description 6
- 238000013178 mathematical model Methods 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 6
- 238000010845 search algorithm Methods 0.000 claims description 6
- 238000013499 data model Methods 0.000 claims description 5
- 238000007500 overflow downdraw method Methods 0.000 claims description 5
- 238000013519 translation Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 210000000744 eyelid Anatomy 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 210000000720 eyelash Anatomy 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 230000000903 blocking effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 241000385223 Villosa iris Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005021 gait Effects 0.000 description 1
- 210000004209 hair Anatomy 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000012466 permeate Substances 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/197—Matching; Classification
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Abstract
The present invention relates to iris identification method and its iris authentication systems.This method includes:Segmentation step carries out iris region segmentation to each frame image for the iris sequence being entered;Step of registration carries out image registration to the iris region after being divided;Denoising step carries out clear image information extraction to iris image sequence on the basis of carrying out image registration through the step of registration and removes noise;And fusion steps, to going the clear iris image sequence after noise to merge.Iris identification method and iris authentication system according to the present invention, can improve the accuracy of iris identification method.
Description
Technical field
The present invention relates to computerized algorithms, more particularly to a kind of iris identification method and iris authentication system.
Background technology
Living things feature recognition (Biometrics) refer to by the distinctive physiology of the mankind or behavioural characteristic (such as face,
Iris, fingerprint, gait and person's handwriting etc.) analyzing processing is carried out to differentiate the technology of individual identity.With traditional personal identification method
(such as identity card, job card, password and password) is compared, and carrying out individual identity using human body biological characteristics differentiates that more stabilization can
It leans on and is not easy to forge.
Since the effective coverage of iris is relatively small, iris image quality is to the imaging device depth of field, user's fitness and light
It is higher according to the dependence of condition, it is blocked in addition, upper lower eyelid and eyelashes often result in effective iris region again, these unfavorable factors
It will lead to the accuracy of identification degradation of iris authentication system in practical applications.Under less constraint image-forming condition, influence
The reason of iris image quality, has:When Image Acquisition user's eyelid blink cause lower eyelid to the blocking of iris effective coverage,
Eyelashes to the blocking of iris region, light source iris region flare to the destruction of iris region information and beyond acquisition
The image-forming range etc. of the equipment depth of field.
The existing strategy for solving the problems, such as the technology of low quality iris image identification and mostly using image co-registration greatly, by several rainbows
Film image, which normalizes to, carries out the operations such as denoising, fusion in the template of fixed size.But low quality iris image is come
It says, the Accurate Segmentation due to being influenced iris region by various unfavorable factors is a problem.Precision is natively not high
Iris annular section normalizes after carrying out linear change, then the rectangular area (by taking Daugman methods as an example) after normalizing is same
There are problems that precision and noise jamming.However, traditional iris image registration technique generally uses this normalized method
To simplify image registration operation.
Being disclosed in the information of background parts of the present invention, it is only intended to increase understanding of the overall background of the invention, without answering
It has been the prior art well known to persons skilled in the art when being considered as recognizing or imply that the information is constituted in any form.
Invention content
In view of the above problems, the present invention is intended to provide one kind being capable of providing high-resolution iris image and can ensure
The iris identification method and iris authentication system of the high precision int of iris recognition.
The iris identification method of the present invention, which is characterized in that including:
Segmentation step carries out iris region segmentation to each frame image for the iris sequence being entered;
Step of registration carries out image registration to the iris region after being divided;
Denoising step clearly schemes iris image sequence on the basis of carrying out image registration through the step of registration
As information extraction and remove noise;And
Fusion steps, to going the clear iris image sequence after noise to merge.
Optionally, include in the segmentation step:
Canny edge detections are carried out to detect the marginal point in image to each frame iris image of input;And
The inner and outer boundary of iris region is searched out using Hough transform to the marginal point that detected.
Optionally, the inner and outer boundary as the iris region searches out inside and outside radius size and the center of circle of iris region
Position.
Optionally, the step of registration includes:
To low quality iris image registration problems founding mathematical models;And
Iris image registration is carried out using iteration closest approach algorithm.
Optionally, as the data model, including:The overlap ratio of affine transformation T=(A, t) and two iris regions
Rate factor lambda1And λ2, wherein A indicates that affine matrix, t are translation variable,
Include during using the iteration closest approach algorithm:
The overlapping ratio factor lambda of two iris images subject to registration of input is estimated according to golden section search algorithm1With
λ2;
Into the iterative process of algorithm:The correspondence between two pairs of point sets, root are established according to the initial value of transformation relation
The transformation relation of the correspondence update point set of strong point collection, until algorithm meets given stop condition.
Optionally, it in the step of registration, carries out carrying out clear image to iris image sequence using following formula (1) and (2)
Information extraction and remove noise:
Wherein, matrix X and N is respectively low-rank matrix and noise sparse matrix, and Y is the square for inputting iris image sequence composition
Battle array carries out convex relaxation to above formula (1) using following formula (2):
Wherein, rank of matrix and zero norm are replaced with nuclear norm and a norm respectively.
Optionally, in the fusion steps, to going the clear iris image sequence after noise to be merged using image mean value
Method is merged.
The computer-readable medium of the present invention, is stored thereon with computer program, which is characterized in that the computer program quilt
Processor realizes above-mentioned iris identification method when executing.
The computer equipment of the present invention, including memory, processor and storage are on a memory and can be on a processor
The computer program of operation, which is characterized in that the processor realizes above-mentioned iris recognition when executing the computer program
Method.
The iris authentication system of the present invention, which is characterized in that including:
Divide module, iris region segmentation is carried out to each frame image of the iris sequence being entered;
Registration module carries out image registration to the iris region after being divided;
Denoising module clearly schemes iris image sequence on the basis of carrying out image registration through the registration module
As information extraction and remove noise;And
Fusion Module, to going the clear iris image sequence after noise to merge.
Optionally, the segmentation module includes:
Detection sub-module carries out Canny edge detections to detect in image for each frame iris image to input
Marginal point;And
Submodule is searched for, searches out the interior outside of iris region using Hough transform for the marginal point to detected
Boundary.
Optionally, as the inner and outer boundary of iris region, described search sub-block searches go out the interior outer radius of iris region
Size and center location.
Optionally, the registration module is used for low quality iris image registration problems founding mathematical models and using changing
Iris image registration is carried out for closest approach algorithm.
Optionally, as the data model, including:The overlap ratio of affine transformation T=(A, t) and two iris regions
Rate factor lambda1And λ2, wherein A indicates that affine matrix, t are translation variable,
Include during using the iteration closest approach algorithm:
The overlapping ratio factor lambda of two iris images subject to registration of input is estimated according to golden section search algorithm1With
λ2;
Into the iterative process of algorithm:The correspondence between two pairs of point sets, root are established according to the initial value of transformation relation
The transformation relation of the correspondence update point set of strong point collection, until algorithm meets given stop condition.
Optionally, it in the registration module, carries out carrying out clear image to iris image sequence using following formula (1) and (2)
Information extraction and remove noise:
Wherein, matrix X and N is respectively low-rank matrix and noise sparse matrix, and Y is the square for inputting iris image sequence composition
Battle array carries out convex relaxation to above formula (1) using following formula (2):
Wherein, rank of matrix and zero norm are replaced with nuclear norm and a norm respectively.
Optionally, the Fusion Module to the clear iris image sequence after removing noise using image mean value fusion method into
Row fusion.
As described above, iris identification method according to the present invention and iris authentication system, can solve less constraints
Under iris recognition problem, the present invention uses for reference the thought of image co-registration on the basis of conventional iris identification technology, will collect
Several low quality iris images by image registration, feature extraction, go noise and image co-registration, finally obtain with high score
The iris image of resolution, so that it is guaranteed that the pinpoint accuracy of iris identification method.
It is used to illustrate the specific reality of certain principles of the present invention together with attached drawing by include this paper attached drawing and then
Apply mode, other feature and advantage possessed by methods and apparatus of the present invention will more specifically become apparent or be explained
It is bright.
Description of the drawings
Fig. 1 is the flow diagram for indicating the iris identification method of the present invention.
Fig. 2 is the organigram for indicating the iris authentication system of the present invention.
Fig. 3 is the organigram for indicating segmentation module.
Specific implementation mode
Be described below be the present invention multiple embodiments in some, it is desirable to provide to the present invention basic understanding.And
It is not intended to the crucial or conclusive element for confirming the present invention or limits scope of the claimed.
The content that the iris identification method and iris authentication system of the present invention are related to belongs to iris preprocessing, in fact
There is provided the frames of an iris preprocessing for matter, the purpose is to be decomposed by image registration and low-rank matrix that several are low
Quality iris image permeates a high-resolution iris image.The iris identification method and iris authentication system of the present invention
The iris image registration strategies for video sequence are proposed, extraction is decomposed clearly using low-rank matrix on the basis of registration
Iris texture information, while inhibiting the noise jamming such as eyelashes, eyelid to the full extent, it further also proposed after removing noise
Iris image sequence is merged to obtain a high-resolution iris image.
It is illustrated firstly, for the iris identification method of the present invention.
Fig. 1 is the flow diagram for indicating the iris identification method of the present invention.
As shown in Figure 1, the iris identification method of the present invention includes the following steps:
Segmentation step S100:Iris region segmentation is carried out to each frame image of the iris sequence of input;
Step of registration S200:Image registration is carried out to the iris region after segmentation;
Denoising step S300:Clear image information extraction is carried out to iris image sequence on the basis of step S200 and is gone
Noise;And
Fusion steps S400:To going the clear iris image sequence after noise to be merged to obtain high-resolution iris figure
Picture.
Segmentation step S100 is to carry out region segmentation, including two steps to each frame iris image of input:The first step
Canny edge detections are carried out to a frame image, second step searches out the inner and outer boundary of iris region using Hough transform.
Configuration step S200 is related to being registrated the iris image sequence after segmentation, including two steps:The first step
To low quality iris image registration problems founding mathematical models, second step carries out iris figure using iteration closest approach (ICP) algorithm
As registration.
Denoising step S300 includes that the iris image after being registrated carries out low-rank matrix decomposition, to extract clearly rainbow
Film image information.
Fusion steps S400 is merged using averaging method (mean fusion) to obtained clear iris image.
Then, these steps are specifically described.
First, an iris image sequence is obtained by image capture device, current common image acquisition mode is to adopt
With the mode of 850nm wavelength near-infrared light source floor lights.For the textural characteristics of east human eye, using 700nm, (wavelength is situated between
Between visible light and near infrared light) light source illuminating effect is more preferably.
In step S100, iris region segmentation is carried out to each frame image in the iris image sequence of input.In this hair
In bright, it is assumed that iris region is non-concentric cyclic structure.The process of iris effective coverage segmentation is to search inside and outside center of circle position
It sets and radius size.First, each frame image of input is marked all marginal points in image with Canny edge detectors
Out, then ballot is carried out to the marginal point that detected by classical Hough transform to may thereby determine that half inside and outside iris
Diameter size and center location.In the present invention, the upper lower eyelid of iris and eyelash etc. need not be blocked and is individually detected.
In step s 200, the iris image sequence after segmentation is registrated.In view of present invention process object is short
The iris image sequence being continuously shot in time, in addition, iris effective coverage suffer from lower eyelid block, eyelashes and light
The influence of spot etc., therefore, iris image registration is modeled as partial data loss in the present invention or there are points when abnormal point
Collect affine registration.
Following principle is followed in registration process:Matching double points are more in model, mean error and smaller;Non- in model
With point to more, mean error and bigger.Model includes that (wherein, A indicates affine matrix to affine transformation T=(A, t), and t is flat
Move variable) and two iris regions overlapping ratio factor lambda1And λ2。
Using classical iteration closest approach (Iterative Closest Point, ICP) algorithm in registration process.?
The starting stage of algorithm estimates the overlap ratio of two iris images subject to registration of input according to golden section search algorithm first
Rate factor lambda1And λ2, subsequently into the iterative process of algorithm:The correspondence between two pairs of point sets is established according to the initial value of transformation relation
Relationship updates the transformation relation of point set according to the correspondence of point set, until algorithm meets given stop condition.
In step S300, the iris image decomposed based on low-rank matrix removes noise.The iris image sequence of input is matched
After standard, due to the iris region stable structure of same eyes, theoretically, the matrix that the iris data collection after registration is constituted is answered
The feature with order minimum.
In addition, in actual engineer application, a common hypothesis is interested signal distributions in the linear of low-dimensional
In subspace, due to iris texture stable structure, it will be assumed that the iris information of noiseless interference is in low-rank subspace.Cause
This, to extract clear iris information while can remove the interference of various noises using the thought that low-rank matrix is decomposed.Such as following formula
It is shown:
In formula, matrix X and N are respectively low-rank matrix and noise sparse matrix, and Y is the square for inputting iris image sequence composition
Battle array.Convex relaxation (Convex Relaxation) is carried out to above formula:
Rank of matrix and zero norm are replaced with nuclear norm and a norm respectively.This problem is converted into l1Standard under norm
Convex optimization problem.For this type of optimization problem, the present invention solves (Alternating using classical alternating direction implicit
Direction Methods,ADM)。
In step S400, to previous step step S300, treated that iris image sequence is merged to recover high score
The iris image of resolution.Preferably, in the present invention simple image mean value fusion method is used.
The iris identification method of the present invention is illustrated above.Sequentially for the present invention iris authentication system into
Row explanation.
Fig. 2 is the organigram for indicating the iris authentication system of the present invention.
As shown in Fig. 2, the iris authentication system 100 of the present invention, which is characterized in that including:
Divide module 110, for carrying out iris region segmentation to each frame image of the iris sequence being entered;
Registration module 120, for carrying out image registration to the iris region after being divided;
Denoising module 130 is used on the basis of carrying out image registration through the registration module 120 to iris image sequence
It carries out clear image information extraction and removes noise;And
Fusion Module 140, for going the clear iris image sequence after noise to merge.
Fig. 3 is the organigram for indicating segmentation module.
Segmentation module 110 further comprises:
Detection sub-module 111 carries out Canny edge detections for each frame iris image to input, detects image
In marginal point;And
Submodule 112 is searched for, for being searched out inside and outside iris region using Hough transform to the marginal point that detected
Boundary.
Wherein, search submodule 112 searches out inside and outside iris region the marginal point that detected using Hough transform
Radius size and center location.
Further, registration module 130 is used for low quality iris image registration problems founding mathematical models and using changing
Iris image registration is carried out for closest approach algorithm.Wherein, as the data model, including:Affine transformation T=(A, t) and
The overlapping ratio factor lambda 1 and λ 2 of two iris regions, wherein A indicates that affine matrix, t are translation variable.
Include during using the iteration closest approach algorithm:
The overlapping ratio factor lambda 1 and λ of two iris images subject to registration of input are estimated according to golden section search algorithm
2;
Into the iterative process of algorithm:The correspondence between two pairs of point sets, root are established according to the initial value of transformation relation
The transformation relation of the correspondence update point set of strong point collection, until algorithm meets given stop condition.
Specifically, registration module 130 carries out carrying out clear image information to iris image sequence using following formula (1) and (2)
Extract and go noise:
Wherein, matrix X and N is respectively low-rank matrix and noise sparse matrix, and Y is the square for inputting iris image sequence composition
Battle array carries out convex relaxation to above formula (1) using following formula (2):
Wherein, rank of matrix and zero norm are replaced with nuclear norm and a norm respectively.
Finally, Fusion Module 140 carries out the clear iris image sequence after removing noise using image mean value fusion method
Fusion.
Further, the present invention also provides a kind of computer-readable mediums, are stored thereon with computer program, and feature exists
In the computer program realizes above-mentioned iris identification method when being executed by processor.
Further, the present invention also provides a kind of computer equipments, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, which is characterized in that the processor is realized when executing the computer program
Above-mentioned iris identification method.
Iris identification method and iris authentication system according to the present invention, the iris that can be solved under less constraints are known
Other problem, the present invention use for reference the thought of image co-registration on the basis of conventional iris identification technology, by several collected low-qualitys
Measure iris image by image registration, feature extraction, go noise and image co-registration, finally obtain with high-resolution iris
Image, so that it is guaranteed that the pinpoint accuracy of iris identification method.In iris recognition product, image capture device would generally obtain one
Then a image sequence carries out image quality measure, choose the panel height quality iris image for being suitble to follow-up link processing and lose
Abandon other images.In the case where constraining less, collected iris image is usually second-rate, but has by multiple image
Effect fusion can utilize the useful information of each frame image, to obtain the iris image of a width better quality to the full extent.With
Upper example primarily illustrates the iris identification method and iris authentication system of the present invention.Although only to the tool of some of present invention
Body embodiment is described, but those of ordinary skill in the art are it is to be appreciated that the present invention can be without departing from its spirit
With the implementation in the form of many other in range.Therefore, the example shown is considered as schematical rather than limit with embodiment
Property processed, in the case where not departing from the spirit and scope of the present invention as defined in appended claims, the present invention may be contained
Cover various modification and replacement.
Claims (16)
1. a kind of iris identification method, which is characterized in that including:
Segmentation step carries out iris region segmentation to each frame image for the iris sequence being entered;
Step of registration carries out image registration to the iris region after being divided;
Denoising step carries out clear image letter on the basis of carrying out image registration through the step of registration to iris image sequence
Breath extracts and goes noise;And
Fusion steps, to going the clear iris image sequence after noise to merge.
2. iris identification method as described in claim 1, which is characterized in that include in the segmentation step:
Canny edge detections are carried out to detect the marginal point in image to each frame iris image of input;And
The inner and outer boundary of iris region is searched out using Hough transform to the marginal point that detected.
3. iris identification method as claimed in claim 2, which is characterized in that
As the inner and outer boundary of the iris region, the inside and outside radius size and center location of iris region are searched out.
4. iris identification method as described in claim 1, which is characterized in that the step of registration includes:
To low quality iris image registration problems founding mathematical models;And
Iris image registration is carried out using iteration closest approach algorithm.
5. iris identification method as claimed in claim 4, which is characterized in that
As the data model, including:The overlapping ratio factor lambda of affine transformation T=(A, t) and two iris regions1With
λ2, wherein A indicates that affine matrix, t are translation variable,
Include during using the iteration closest approach algorithm:
The overlapping ratio factor lambda of two iris images subject to registration of input is estimated according to golden section search algorithm1And λ2;
Into the iterative process of algorithm:The correspondence between two pairs of point sets is established according to the initial value of transformation relation, according to point
The transformation relation of the correspondence update point set of collection, until algorithm meets given stop condition.
6. iris identification method as claimed in claim 5, which is characterized in that
In the step of registration, carries out carrying out clear image information extraction to iris image sequence and go using following formula (1) and (2)
Noise:
Wherein, matrix X and N is respectively low-rank matrix and noise sparse matrix, and Y is the matrix for inputting iris image sequence composition,
Convex relaxation is carried out using following formula (2) to above formula (1):
Wherein, rank of matrix and zero norm are replaced with nuclear norm and a norm respectively.
7. iris identification method as claimed in claim 6, which is characterized in that
In the fusion steps, to going the clear iris image sequence after noise to melt using image mean value fusion method
It closes.
8. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that the computer program is by processor
The iris identification method described in any one of claim 1~7 is realized when execution.
9. a kind of computer equipment, including memory, processor and storage can be run on a memory and on a processor
Computer program, which is characterized in that the processor is realized any one in claim 1~7 when executing the computer program
Iris identification method described in.
10. a kind of iris authentication system, which is characterized in that including:
Divide module, for carrying out iris region segmentation to each frame image of the iris sequence being entered;
Registration module, for carrying out image registration to the iris region after being divided;
Denoising module, for clearly being schemed to iris image sequence on the basis of carrying out image registration through the registration module
As information extraction and remove noise;And
Fusion Module, for going the clear iris image sequence after noise to merge.
11. iris authentication system as claimed in claim 10, which is characterized in that
The segmentation module includes:
Detection sub-module carries out Canny edge detections to detect the side in image for each frame iris image to input
Edge point;And
Submodule is searched for, searches out the inner and outer boundary of iris region using Hough transform for the marginal point to detected.
12. iris authentication system as claimed in claim 11, which is characterized in that
As the inner and outer boundary of iris region, described search sub-block searches go out inside and outside radius size and the center of circle position of iris region
It sets.
13. iris authentication system as claimed in claim 11, which is characterized in that
The registration module is used for low quality iris image registration problems founding mathematical models and is counted recently using iteration
Method carries out iris image registration.
14. iris authentication system as claimed in claim 13, which is characterized in that
As the data model, including:The overlapping ratio factor lambda of affine transformation T=(A, t) and two iris regions1With
λ2, wherein A indicates that affine matrix, t are translation variable,
Include during using the iteration closest approach algorithm:
The overlapping ratio factor lambda of two iris images subject to registration of input is estimated according to golden section search algorithm1And λ2;
Into the iterative process of algorithm:The correspondence between two pairs of point sets is established according to the initial value of transformation relation, according to point
The transformation relation of the correspondence update point set of collection, until algorithm meets given stop condition.
15. iris authentication system as claimed in claim 14, which is characterized in that
In the registration module, carries out carrying out clear image information extraction to iris image sequence and go using following formula (1) and (2)
Noise:
Wherein, matrix X and N is respectively low-rank matrix and noise sparse matrix, and Y is the matrix for inputting iris image sequence composition,
Convex relaxation is carried out using following formula (2) to above formula (1):
Wherein, rank of matrix and zero norm are replaced with nuclear norm and a norm respectively.
16. iris authentication system as claimed in claim 15, which is characterized in that
The Fusion Module merges the clear iris image sequence after removing noise using image mean value fusion method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810257237.6A CN108734085A (en) | 2018-03-27 | 2018-03-27 | Iris identification method and iris authentication system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810257237.6A CN108734085A (en) | 2018-03-27 | 2018-03-27 | Iris identification method and iris authentication system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108734085A true CN108734085A (en) | 2018-11-02 |
Family
ID=63940595
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810257237.6A Pending CN108734085A (en) | 2018-03-27 | 2018-03-27 | Iris identification method and iris authentication system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108734085A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1426760A (en) * | 2001-12-18 | 2003-07-02 | 中国科学院自动化研究所 | Identity discriminating method based on living body iris |
CN102043954A (en) * | 2011-01-30 | 2011-05-04 | 哈尔滨工业大学 | Quick and robust iris recognition method based on related function matching |
CN103116883A (en) * | 2012-11-07 | 2013-05-22 | 哈尔滨工程大学 | Normalized cross correlation (NCC) registration method of self-adaptation threshold |
CN103714550A (en) * | 2013-12-31 | 2014-04-09 | 鲁东大学 | Image registration automatic optimization algorithm based on matching of curve characteristic evaluation |
CN105957026A (en) * | 2016-04-22 | 2016-09-21 | 温州大学 | De-noising method based on recessive low-rank structure inside and among nonlocal similar image blocks |
CN106204530A (en) * | 2016-06-27 | 2016-12-07 | 西安交通大学 | Ancient times of based on image registration West Xia Dynasty's literary composition typography discrimination method |
US20170076146A1 (en) * | 2015-09-11 | 2017-03-16 | EyeVerify Inc. | Fusing ocular-vascular with facial and/or sub-facial information for biometric systems |
-
2018
- 2018-03-27 CN CN201810257237.6A patent/CN108734085A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1426760A (en) * | 2001-12-18 | 2003-07-02 | 中国科学院自动化研究所 | Identity discriminating method based on living body iris |
CN102043954A (en) * | 2011-01-30 | 2011-05-04 | 哈尔滨工业大学 | Quick and robust iris recognition method based on related function matching |
CN103116883A (en) * | 2012-11-07 | 2013-05-22 | 哈尔滨工程大学 | Normalized cross correlation (NCC) registration method of self-adaptation threshold |
CN103714550A (en) * | 2013-12-31 | 2014-04-09 | 鲁东大学 | Image registration automatic optimization algorithm based on matching of curve characteristic evaluation |
US20170076146A1 (en) * | 2015-09-11 | 2017-03-16 | EyeVerify Inc. | Fusing ocular-vascular with facial and/or sub-facial information for biometric systems |
CN105957026A (en) * | 2016-04-22 | 2016-09-21 | 温州大学 | De-noising method based on recessive low-rank structure inside and among nonlocal similar image blocks |
CN106204530A (en) * | 2016-06-27 | 2016-12-07 | 西安交通大学 | Ancient times of based on image registration West Xia Dynasty's literary composition typography discrimination method |
Non-Patent Citations (5)
Title |
---|
DIMITRI A. CHERNYAK: "Iris-Based Cyclotorsional Image Alignment Method for Wavefront Registration", 《IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING》 * |
吴国瑶 等: "基于B样条FFD模型配准的虹膜图像融合方法", 《山东大学学报(工学版)》 * |
吴礼洋 等: "基于Fast-match和改进AICP算法的人脸鲁棒精确仿射配准", 《光学学报》 * |
张家良 等: "基于对齐度和互信息的红外与可见光图像配准", 《现代电子技术》 * |
曾叶 等: "一种基于图像对齐的虹膜分割方法", 《计算机工程》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Raghavendra et al. | Novel image fusion scheme based on dependency measure for robust multispectral palmprint recognition | |
Ali et al. | An iris recognition system to enhance e-security environment based on wavelet theory | |
Connaughton et al. | Fusion of face and iris biometrics | |
US20060222212A1 (en) | One-dimensional iris signature generation system and method | |
WO2017059591A1 (en) | Finger vein identification method and device | |
US20120163678A1 (en) | System and method for identifying a person with reference to a sclera image | |
CN101093539A (en) | Matching identification method by extracting characters of vein from finger | |
Nithya et al. | Iris recognition techniques: a literature survey | |
Vega et al. | Biometric personal identification system based on patterns created by finger veins | |
Abidin et al. | Iris segmentation analysis using integro-differential and hough transform in biometric system | |
Aleem et al. | Fast and accurate retinal identification system: Using retinal blood vasculature landmarks | |
Malinowski et al. | An iris segmentation using harmony search algorithm and fast circle fitting with blob detection | |
Taha et al. | Iris features extraction and recognition based on the local binary pattern technique | |
Zheng | Static and dynamic analysis of near infra-red dorsal hand vein images for biometric applications | |
Suzaki et al. | A horse identification system using biometrics | |
CN108875579B (en) | Morphology-based close-range gesture recognition method | |
Szczepański et al. | Pupil and iris detection algorithm for near-infrared capture devices | |
Estudillo-Romero et al. | The Hermite transform: An alternative image representation model for iris recognition | |
Ibrahim | Iris recognition using Haar wavelet transform | |
Wasnik et al. | Improved fingerphoto verification system using multi-scale second order local structures | |
CN107153807A (en) | A kind of non-greedy face identification method of two-dimensional principal component analysis | |
CN108734085A (en) | Iris identification method and iris authentication system | |
Prema et al. | A review: face recognition techniques for differentiate similar faces and twin faces | |
Sahmoud | Enhancing iris recognition | |
Silpamol et al. | Detection and rectification of distorted fingerprints |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181102 |