CN110136180A - Image template matching system and algorithm based on Choquet integral - Google Patents
Image template matching system and algorithm based on Choquet integral Download PDFInfo
- Publication number
- CN110136180A CN110136180A CN201910405519.0A CN201910405519A CN110136180A CN 110136180 A CN110136180 A CN 110136180A CN 201910405519 A CN201910405519 A CN 201910405519A CN 110136180 A CN110136180 A CN 110136180A
- Authority
- CN
- China
- Prior art keywords
- image
- choquet integral
- template matching
- choquet
- integral
- 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.)
- Granted
Links
Classifications
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- 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
- G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of image template matching systems and algorithm based on Choquet integral, including detect and extract product area image；Image after extraction is pre-processed, binaryzation and morphological operation；Defect shape information and location information obtain；Location error screening and similarity screening；System probability, conviction measurement and the reasonability sieve of Choquet integral are calculated, successively removal is unsatisfactory for the point of threshold requirement, filters out match point.The security reliability of product can be improved in this programme of the invention after implementing, reduce product cost, reduces the wasting of resources, improves working efficiency.The present invention is based on the template matching methods of Choquet integral can provide better matching performance.The incomplete information environments such as the present invention is based on fuzzy algorithmic approaches, block for existing, and object is overlapping are able to achieve preferable matching.
Description
Technical field
The present invention relates to a kind of image template matching systems and algorithm based on Choquet integral, belong to vision matching skill
Art field.
Background technique
Existing template matching algorithm is difficult to realize determine reference picture (template) in larger figure in incomplete information environment
As the best match position in (scene).
However in the product image of reality, it is understood that there may be since product is overlapping, situations such as blocking leads to the product obtained
Image information is simultaneously imperfect, this increases difficulty for matching.Choquet fuzzy integral gives a kind of intuitionistic fuzzy criterion matrix
New Method of Decision, provide a kind of completely new image template matching system the present invention is based on Choquet fuzzy integral and matching calculated
Method.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of image templates based on Choquet integral
Matching system and algorithm obtain more accurate and robust by the match decision of the image template matching algorithm in conjunction with limited quantity
Matching, Choquet relevant to fuzzy measurement integral can be used for handling to be obscured as caused by incomplete image information
Property.
The object of the invention is achieved by following technical solution:
A kind of image template matching algorithm based on Choquet integral is provided, including
1) it obtains and extracts product area image；
2) image after extraction is pre-processed, binaryzation and morphological operation；
3) defect shape information and location information obtain；
4) location error screening and similarity screening；
5) system probability, conviction measurement and the reasonability sieve of Choquet integral are calculated, successively removal is unsatisfactory for threshold value and wants
The point asked, filters out match point.
Preferably, morphological operation includes the grain noise filtered out in image by etching operation, suitable using expansive working
Degree restores the edge for the target that is corroded.
Preferably, location error screening includes calculating the location error of product, if error is more than threshold value, terminates the image
Detection；If location error is less than threshold value, similarity screening is carried out.
Preferably, similarity screening includes calculating texture paging h (x(i)), filter out the pixel more than texture threshold.
Preferably, system probability calculation is as follows:
Wherein p ({ x(i)) it is pixel x(i)Weight,N is pixel number.
Preferably, p ({ x(i)) use x(i)The fog-density of point is obtained divided by maximum fog-density.
Preferably, conviction metric calculation method are as follows:
Wherein m (A(i))=g (x(i))-g(x(i-1)), g (x(i)) it is characterized the fuzzy measurement of value, X is image point set.
Preferably, reasonability calculation method are as follows:
Wherein m (A(i))=g (x(i))-g(x(i-1)), g (x(i)) it is characterized the fuzzy measurement of value, X is image point set.
Preferably, further include step 6) according to matched point calculating matching rate, and then determine whether to match with image template.
Preferably, the characteristic value includes brightness, contrast, rotation or ratio.
A kind of image template matching system based on Choquet integral is provided simultaneously, comprising:
Module is obtained, the image of product is obtained；
Extraction module extracts product area image；
Processing module pre-processes product area image, binaryzation and morphological operation and is sent to defect form
Data obtaining module and position information acquisition module；
Defect shape information obtains module, obtains product defects shape information；
Position information acquisition module obtains product space information；
Location error screening module calculates the location error of product based on product space information, if error is more than threshold value,
Terminate the detection of image；If location error is less than threshold value, similarity screening module is sent an image to；
Similarity screening module, it is general to the system for being sent to Choquet integral after the progress similarity screening of image pixel point
Rate adaptation；
The system probability match device of Choquet integral, calculates the system probability of pixel, and removal is lower than system probability threshold value
Pixel after be sent to Choquet integral conviction metrics match device；
The conviction metrics match device of Choquet integral, calculates the conviction measurement of pixel, and removal is lower than conviction metric threshold
Pixel after be sent to Choquet integral reasonability adaptation；
The reasonability adaptation of Choquet integral, calculates the reasonability of pixel, and removal is lower than the pixel of reasonability threshold value
Point.
Preferably, morphological operation includes the grain noise filtered out in image by etching operation, suitable using expansive working
Degree restores the edge for the target that is corroded.
Preferably, similarity screening module calculates texture paging h (x(i)), filter out the pixel more than texture threshold.
Preferably, system method for calculating probability are as follows:
Wherein p ({ x(i)) it is pixel x(i)Weight,N is pixel number.
Preferably, p ({ x(i)) use x(i)The fog-density of point is obtained divided by maximum fog-density.
Preferably, conviction metric calculation method are as follows:
Wherein m (A(i))=g (x(i))-g(x(i-1)), g (x(i)) it is characterized the fuzzy measurement of value, X is image point set.
Preferably, reasonability calculation method are as follows:
Wherein m (A(i))=g (x(i))-g(x(i-1)), g (x(i)) it is characterized the fuzzy measurement of value, X is image point set.
Preferably, the characteristic value includes brightness, contrast, rotation or ratio.
The invention has the following advantages over the prior art:
(1) security reliability of product can be improved in matching scheme of the invention after implementing, and reduces product cost, reduces resource
Waste improves working efficiency.
(2) the present invention is based on the template matching methods of Choquet integral can provide better matching performance.
(3) the present invention is based on fuzzy algorithmic approach, blocked for existing, the incomplete information environment such as object is overlapping be able to achieve compared with
Good matching.
Detailed description of the invention
Fig. 1 is matching algorithm flow chart of the present invention.
Specific embodiment
One, Choquet integration method polymerize individual adaptation
1, fuzzy measurement:
Give a limited discrete whole collection x={ x1, x2... xn, a probability measure P:2X→ [0,1] can pass through
Fuzzy metric attribute defines:
(1) boundedness P (φ)=0 and P (X)=1；
(2) additive property P (A ∪ B)=P (A)+P (B) of any A,With A ∩ B=φ.
Uncertain proposition has with properties:
(1) boundedness Pos (φ)=0 Pos (X)=1；
(2) any Pos (A ∪ B)=max (Pos (A), Pos (B)).Duality as uncertain proposition
Possibility measure there is polarity attributes rather than maximum attribute；
(3) Nec (A ∩ B)=min (Nec (a), Nec (B)) is any
These attributes can be regarded as the limited variant of the following monotonicity of fuzzy measurement g:
(1) g (A ∪ B) >=max (g (A), g (B))；
(2) g (A ∪ B)≤min (g (A), g (B)) is any
Conviction measurement and measuring similarity can indicate that the basic probability assignment is one with basic probability assignment (BPA)
A set function, m (A): 2X→ [0,1]；
M (φ)=0 He
Conviction measurementIt is defined as
The reasonability measurement of duality as conviction measurement is defined as:
Pl (A)=∑ B ∩ A ≠ φm(B)
It is well known that similarity and conviction measurement are the subset of high probability and low probability measurement respectively.If focus element
It is nested or auxiliary, i.e., is ranked up by the inclusion relation gathered, then confidence level and conviction measurement respectively becomes possibility
And possibility measure.Herein, focus element is a set A, obtains m (A) > 0.It is said differently, Pl (A ∪ B)=max
(Pl (A), Pl (B)) and Bel (A ∩ B)=min (Bel (A), Bel (B)).
Accordingly, it is possible to which property measurement is also referred to as consonant creditability measurement and possibility measure as auxiliary confidence measure.
2, the fuzzy measurement of Choquet integral
We will describe about probability, the discrete Choquet integral of conviction and reasonability measurement.Choquet integral is to close
It is desired extensive in the normal function of probability measure, by conventional expectation concept to degree a possibility that using other than probability measure
The expectation of amount.
Let us is since Choquet and the integral of r probability measure.Because of h ({ x(1)})≤h({x(2)})≤…≤h
({x(n)), the Choquet integral about fuzzy measurement g is defined as
CI(h1... hn)=∑I=1 ... n[g(A(i))-g(A(i+1))].h(xi)
Ai={ xi, xi+1... xnAnd An+1=φ.Known to be, as a special case, Choquet integral is surveyed about probability
Degree is reduced to weighted arithmetic mean operator.For probability measure, due to additive property, Choquet integral formula is indicated are as follows:
Simultaneously
Allow fuzzy measure according to 0≤g ({ x(1)})≤g({x(2)})≤…≤g({x(n))=1.It is assumed here thatAs constraint condition.It can by by given fog-density divided by maximum fog-density.This is false
If being the condition of being possibly realized property of g measurement.Consider a subset A(i)={ x(i)... x(n)}.The discrete desired value up and down of function
(UE and LE) h1..., hnIt is by Lebesgues-Stieljes Definitions On Integration respectively
With
Here m (A(i)) it is the basic probability assignment (BPA) for generating the focus element of reasonability and conviction measurement.For embedding
Cover focus A(i)Element,Their BPA is generated by following formula
m(A(i))=g (x(i))-g(x(i-1))
It is assumed herein that g ({ x(0))=0.Discrete Choquet is replaced to integrate template matcher CI with this BPAbel,
It is as follows with regard to the trust metrics on X:
CIbel(h1..., hn)=UE (h1..., hn)
This relationship is also considered as by desired value a possibility that Choquet integral calculation, is polymerize based on optimistic function.
Therefore, it is interpreted that (cooperate with or support) of a kind of optimism polymerize or as a kind of complementary viewpoint.
In an equivalent manner, discrete Choquet integrates template matcher CIplIt can be with relative to the creditability measurement on X
It indicates are as follows:
CIpl(h1..., hn)=LE (h1..., hn)
It is also regarded as the necessary expectation of Choquet Definitions On Integration.Polymerization is based on pessimistic function.Therefore, it is operated and is obtained by this
The result obtained can be interpreted pessimistic (i.e. destructive or redundancy) aggregation or from the evidence aggregation for replacing viewpoint.
Two, matching algorithm process
Matching algorithm the following steps are included:
1) it detects and extracts product area image；
The camera of using terminal carries out Image Acquisition, and extracts the image of product area.
2) image after extraction is pre-processed, binaryzation and morphological operation；
Median filter process is carried out to the image of extracted product area, then carries out binary conversion treatment；To binary picture
As carrying out Morphological scale-space, the grain noise in image is filtered out by etching operation, moderately reduction is corroded using expansive working
The edge of target；
3) defect shape information and location information obtain；
The defect information of product, such as defective locations, defect size etc. are obtained, determines location information and the movement of product
Direction vector.
4) location error and similarity technology；
The location error of product is calculated, if error is more than threshold value, shows not to be product to be detected, records information, and tie
The detection of the beam image；If location error is less than threshold value, texture paging h (x is calculatedi).Here texture can be set
The threshold value of similitude, the point for selecting more than threshold value continue integral matching.
5) Choquet integral matching；
Characteristics of image is matched with sample planting modes on sink characteristic, specific step is as follows for the matching of Choquet integral:
5.1 utilize texture paging h (xi) computing system probability:
p({x(i)) it is pixel x(i)Weight, using fog-density divided by maximum fog-density,Removal is lower than the point of system probability threshold value.
5.2 obtain the fuzzy measurement g (x of characteristic value (brightness, contrast, rotation, ratio etc.)(i)), calculate m (A(i))=g
(x(i))-g(x(i-1))；Calculate h (xi) upper and lower desired value:
5.3 generate conviction metrics match device and reasonability adaptation
CIbel(h1..., hn)=UE (h1..., hn)
CIpl(h1..., hn)=LE (h1..., hn)
5.4 are lower than the pixel of conviction metric threshold using the removal of conviction metrics match device, and the removal of reasonability adaptation is low
In the pixel of reasonability threshold value, residual pixel point is matched pixel.Matching rate is calculated according to matched pixel, in turn
Determine whether to match with image template.
Referring to table 1, the matched well average rate of adaptation is integrated for Choquet, wherein MOAD indicates the maximum of absolute difference
Value, SOAD indicate antipode value, and SOSD indicates the sum of difference of two squares.
The matched well average rate of 1 Choquet of table integral adaptation
| Adaptation | Brightness | Contrast | Rotation | Ratio |
| MOAD | 0.808 | 0.885 | 0.335 | 0.403 |
| SOAD | 0.180 | 0.848 | 0.758 | 0.819 |
| SOSD | 0.290 | 0.874 | 0.745 | 0.781 |
| CIBEL | 0.808 | 0.885 | 0.377 | 0.477 |
| CIPL | 0.790 | 0.923 | 0.826 | 0.868 |
| CIPR | 0.786 | 0.923 | 0.384 | 0.481 |
The above, optimal specific embodiment only of the invention, but scope of protection of the present invention is not limited thereto,
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by anyone skilled in the art,
It should be covered by the protection scope of the present invention.
The content that description in the present invention is not described in detail belongs to the well-known technique of professional and technical personnel in the field.
Claims (10)
1. a kind of image template matching algorithm based on Choquet integral characterized by comprising
1) it obtains and extracts product area image；
2) image after extraction is pre-processed, binaryzation and morphological operation；
3) defect shape information and location information obtain；
4) location error screening and similarity screening；
5) system probability, conviction measurement and the reasonability sieve of Choquet integral are calculated, successively removal is unsatisfactory for threshold requirement
Point, filters out match point.
2. the image template matching algorithm as described in claim 1 based on Choquet integral, which is characterized in that morphology behaviour
Work includes the grain noise filtered out in image by etching operation, and the edge for the target that is corroded moderately is restored using expansive working.
3. the image template matching algorithm as described in claim 1 based on Choquet integral, which is characterized in that location error
Screening includes calculating the location error of product, if error is more than threshold value, terminates the detection of the image；If location error does not surpass
Threshold value is crossed, then carries out similarity screening.
4. the image template matching algorithm as claimed in claim 3 based on Choquet integral, which is characterized in that similarity packet
It includes and calculates texture paging h (x(i)), filter out the point more than texture threshold.
5. the image template matching algorithm as claimed in claim 4 based on Choquet integral, which is characterized in that system probability
Calculation method are as follows:
Wherein p ({ x(i)) it is x(i)Weight,N is points.
6. the image template matching algorithm as claimed in claim 5 based on Choquet integral, which is characterized in that p ({ x(i)})
Using x(i)The fog-density of point is obtained divided by maximum fog-density.
7. the image template matching algorithm as claimed in claim 1 or 5 based on Choquet integral, which is characterized in that Degree of Belief
Measure calculation method are as follows:
Wherein m (A(i))=g (x(i))-g(x(i-1)), g (x(i)) it is characterized the fuzzy measurement of value, X is image point set.
8. the image template matching algorithm as claimed in claim 7 based on Choquet integral, which is characterized in that reasonability meter
Calculation method are as follows:
Wherein m (A(i))=g (x(i))-g(x(i-1)), g (x(i)) it is characterized the fuzzy measurement of value, X is image point set.
9. the image template matching algorithm as claimed in claim 8 based on Choquet integral, which is characterized in that further include step
It is rapid that matching rate 6) is calculated according to matched point, and then determine whether to match with image template.
10. a kind of image template matching system based on Choquet integral characterized by comprising
Module is obtained, the image of product is obtained；
Extraction module extracts product area image；
Processing module pre-processes product area image, binaryzation and morphological operation and is sent to defect shape information
Obtain module and position information acquisition module；
Defect shape information obtains module, obtains product defects shape information；
Position information acquisition module obtains product space information；
Location error screening module calculates the location error of product based on product space information, if error is more than threshold value, terminates
The detection of image；If location error is less than threshold value, similarity screening module is sent an image to；
Similarity screening module, to the system probability for being sent to Choquet integral after the progress similarity screening of image pixel point
Orchestration；
The system probability match device of Choquet integral, calculates the system probability of pixel, and removal is lower than the picture of system probability threshold value
The conviction metrics match device of Choquet integral is sent to after vegetarian refreshments；
The conviction metrics match device of Choquet integral, calculates the conviction measurement of pixel, and removal is lower than the picture of conviction metric threshold
The reasonability adaptation of Choquet integral is sent to after vegetarian refreshments；
The reasonability adaptation of Choquet integral, calculates the reasonability of pixel, and removal is lower than the pixel of reasonability threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910405519.0A CN110136180B (en) | 2019-05-16 | 2019-05-16 | Image template matching system and algorithm based on Choquet integral |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910405519.0A CN110136180B (en) | 2019-05-16 | 2019-05-16 | Image template matching system and algorithm based on Choquet integral |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110136180A true CN110136180A (en) | 2019-08-16 |
| CN110136180B CN110136180B (en) | 2021-07-02 |
Family
ID=67574421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910405519.0A Active CN110136180B (en) | 2019-05-16 | 2019-05-16 | Image template matching system and algorithm based on Choquet integral |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110136180B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100014755A1 (en) * | 2008-07-21 | 2010-01-21 | Charles Lee Wilson | System and method for grid-based image segmentation and matching |
| CN103839271A (en) * | 2014-03-25 | 2014-06-04 | 天津理工大学 | Method for image texture synthesis based on optimum matching |
| CN103913464A (en) * | 2014-03-20 | 2014-07-09 | 湖南大学 | High speed railway track surface defect coupling method based on machine visual inspection |
| CN106780422A (en) * | 2016-12-28 | 2017-05-31 | 深圳市美好幸福生活安全系统有限公司 | A kind of notable figure fusion method based on Choquet integrations |
| CN108109205A (en) * | 2017-12-08 | 2018-06-01 | 山东师范大学 | A kind of incomplete model index and method for reconstructing based on holes filling |
| CN109308716A (en) * | 2018-09-20 | 2019-02-05 | 珠海市君天电子科技有限公司 | A kind of image matching method, device, electronic equipment and storage medium |
| CN109657966A (en) * | 2018-12-13 | 2019-04-19 | 国网河北省电力有限公司电力科学研究院 | Transmission line of electricity risk composite valuations method based on fuzzy mearue evaluation |
| CN109727239A (en) * | 2018-12-27 | 2019-05-07 | 北京航天福道高技术股份有限公司 | Based on SURF feature to the method for registering of inspection figure and reference map |
-
2019
- 2019-05-16 CN CN201910405519.0A patent/CN110136180B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100014755A1 (en) * | 2008-07-21 | 2010-01-21 | Charles Lee Wilson | System and method for grid-based image segmentation and matching |
| CN103913464A (en) * | 2014-03-20 | 2014-07-09 | 湖南大学 | High speed railway track surface defect coupling method based on machine visual inspection |
| CN103839271A (en) * | 2014-03-25 | 2014-06-04 | 天津理工大学 | Method for image texture synthesis based on optimum matching |
| CN106780422A (en) * | 2016-12-28 | 2017-05-31 | 深圳市美好幸福生活安全系统有限公司 | A kind of notable figure fusion method based on Choquet integrations |
| CN108109205A (en) * | 2017-12-08 | 2018-06-01 | 山东师范大学 | A kind of incomplete model index and method for reconstructing based on holes filling |
| CN109308716A (en) * | 2018-09-20 | 2019-02-05 | 珠海市君天电子科技有限公司 | A kind of image matching method, device, electronic equipment and storage medium |
| CN109657966A (en) * | 2018-12-13 | 2019-04-19 | 国网河北省电力有限公司电力科学研究院 | Transmission line of electricity risk composite valuations method based on fuzzy mearue evaluation |
| CN109727239A (en) * | 2018-12-27 | 2019-05-07 | 北京航天福道高技术股份有限公司 | Based on SURF feature to the method for registering of inspection figure and reference map |
Non-Patent Citations (1)
| Title |
|---|
| 丁莹 等: "基于Choquet模糊积分的运动目标检测算法", 《电子学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110136180B (en) | 2021-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6077993B2 (en) | Image data processing method, system, and program for identifying image variants | |
| US7881494B2 (en) | Characteristic point detection of target object included in an image | |
| US20080123975A1 (en) | Abnormal Action Detector and Abnormal Action Detecting Method | |
| JP5214533B2 (en) | Person tracking method, person tracking apparatus, and person tracking program | |
| JP5227888B2 (en) | Person tracking method, person tracking apparatus, and person tracking program | |
| JP5290227B2 (en) | Object detection device and learning device thereof | |
| WO2018076392A1 (en) | Pedestrian statistical method and apparatus based on recognition of parietal region of human body | |
| JP5290229B2 (en) | Learning device and object detection device | |
| JP2009020884A (en) | Method and system for specifying feature of image data | |
| CN111695410A (en) | Violation reporting method and device, computer equipment and storage medium | |
| CN113379739B (en) | Ultrasonic image identification method, device, equipment and storage medium | |
| CN113822157A (en) | Mask wearing face recognition method based on multi-branch network and image restoration | |
| Tian et al. | Automatic histogram-based initialization of k-means clustering in CT | |
| Cazorla et al. | Two bayesian methods for junction classification | |
| CN117173590A (en) | Water body abnormality monitoring method based on multisource time sequence remote sensing image | |
| Arifin et al. | Detection of overlapping teeth on dental panoramic radiograph | |
| CN110136180A (en) | Image template matching system and algorithm based on Choquet integral | |
| US7773794B2 (en) | Method and apparatus for candidate detection using Hessian peak characteristics | |
| CN110119748A (en) | Image template match point screening system and method based on Choquet integral | |
| KR101473991B1 (en) | Method and apparatus for detecting face | |
| CN110363192A (en) | Object image identification system and object image discrimination method | |
| Oberti et al. | Performance evaluation criterion for characterizing video-surveillance systems | |
| CN114049345A (en) | Liver CT image classification method based on data set | |
| JP2012100082A (en) | Image processing device, image processing method, and program | |
| CN109389094B (en) | Stable iris feature extraction and matching method |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Xiong Changwei Inventor after: Zhang Haoyu Inventor after: Yuan Wei Inventor after: Zheng Cankai Inventor after: Zheng Weibin Inventor after: Li Haoyu Inventor after: Lin Zhirui Inventor after: Shu Yufeng Inventor after: Liu Zhiwei Inventor after: Yin Runping Inventor after: Zhong Hongjian Inventor after: Mo Zhuotian Inventor after: Liu Wanrong Inventor after: Liu Shufeng Inventor after: Tao Lixun Inventor before: Shu Yufeng Inventor before: Liang Yaorong Inventor before: Li Longgen Inventor before: Fan Sili Inventor before: Xiong Changwei Inventor before: Mei Yanghan Inventor before: Liu Zhiwei Inventor before: Zhang Junhua Inventor before: Luo Lixing Inventor before: Chen Tianyu |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |