CN104933740B - A kind of digital hole distinguishing method between true and false of certificate - Google Patents
A kind of digital hole distinguishing method between true and false of certificate Download PDFInfo
- Publication number
- CN104933740B CN104933740B CN201510267706.9A CN201510267706A CN104933740B CN 104933740 B CN104933740 B CN 104933740B CN 201510267706 A CN201510267706 A CN 201510267706A CN 104933740 B CN104933740 B CN 104933740B
- Authority
- CN
- China
- Prior art keywords
- digital
- hole
- circle
- digital hole
- false
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Inspection Of Paper Currency And Valuable Securities (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of digital hole distinguishing method between true and false of certificate, specifically implement according to following steps：Digital hole pattern picture is gathered in the case where amplifying 10 to 20 times to the region containing digital hole in certificate；Logarithmic code hole pattern picture carries out binaryzation and extracts each digital hole using the method for connected component analysis；The method being fitted using least squar cicular is fitted each digital hole, then extracts the center of circle and the radius of fitting circle；Then judge the true and false in digital hole with method of comparison and K mean cluster method, the digital hole distinguishing method between true and false of certificate of the present invention solves artificial present in prior art waste time and energy when differentiating the certificate number hole true and false and the problem of identification efficiency is not high.
Description
Technical field
The invention belongs to certificate false proof authentication technique field, and in particular to a kind of digital hole distinguishing method between true and false of certificate.
Background technology
Certificate is to provide according to law, assigns or authorize the proof in terms of identity, qualification, power or the honor of certain main body
Property file, such as resident identification card, passport, driver's license.Generally, can all occur some digital codes in certificate, such as
Sequence number, the page number, identity code of holder of certificate etc., often these digital codes identify the uniqueness and card of certificate
Uniformity of part and holder etc..In order to improve the security of certificate, these digital codes of additional clause are generally using punching skill
Art progress is false proof, and the hole for forming these digital codes is called digital hole.Digital hole on general real docu-ment is in the same size, hole
The smooth of the edge, digital hole site marshalling, and the digital Kong Ze forged does not have these characteristics.
Additional clause number hole, the diameter in each hole is general between 0.1-0.8mm, is typically in 10-20 times of amplification
In the case of, the normative feature such as void shape and the arrangement of digital punching is observed by way of artificial naked eyes, to recognize it
True or false.Artificial mirror method for distinguishing is wasted time and energy, and is difficult to ensure that the objectivity of discriminating.
The content of the invention
It is an object of the invention to provide a kind of digital hole distinguishing method between true and false of certificate, people present in prior art is solved
The problem of work wastes time and energy when differentiating the certificate number hole true and false and identification efficiency is not high.
The technical solution adopted in the present invention is, the digital hole distinguishing method between true and false of a kind of certificate, specifically according to following steps
Implement：
Step 1, the digital bore region of certificate amplified 10 to 20 times, the digital bore region of additional clause is entered using transmission mode
Digitized IMAQ, if the digital hole pattern picture collected is f (x, y)；
Step 2, to the digital hole pattern that is gathered in step 1 as f (x, y) carries out binaryzation and using the side of connected component analysis
Method extracts each digital hole；
Step 3, each digital hole to being extracted in step 2, the method being fitted using Least Square Circle are fitted per number
Code holes, and extract the central coordinate of circle and radius geometric parameter of fitting circle；
Step 4, the central coordinate of circle according to the fitting circle obtained in step 3 and the radius geometric parameter resulting number code holes true and false,
The specific method of the resulting number code holes true and false has two kinds, and first method is method of comparison：The each digital hole extracted in calculation procedure 3
The difference value at edge and fitting circle, and a threshold value is set, if difference value is more than threshold value, judge number of the digital hole as forgery
Code holes；Second method is K mean cluster method：Using the method for K mean cluster, all digital holes in step 3 are fitted
Central coordinate of circle clustered on line direction and column direction, then calculate the difference that cluster centre is deviateed in each digital hole center of circle,
If difference value is more than the threshold value of a setting, shows that digital hole is misaligned, belong to the digital hole of forgery.
The features of the present invention is also resided in,
The specific method of step 2 is：
Step 2.1), using maximum variance between clusters, to the digital hole pattern in step 1 as f (x, y) carries out binaryzation, if
Image after binaryzation is g (x, y)；
Step 2.2), to step 2.1) in image g (x, y) after binaryzation using 8 neighborhood connected component analysis method,
Each digital hole is extracted, if the digital hole extracted is hi(i=1,2 ..., P), wherein P represents the number in digital hole.
The specific method of step 3 is：
Step 3.1), each digital hole is fitted using circle, if i-th of number hole hiThe center of circle of the circle fitted is sat
It is designated as (Ai,Bi), radius be Ri, then circle is expressed as:
Another representation of formula (1) is：x2+y2+aix+biy+ci=0, (2)
The transforming relationship of formula (1) and formula (2) is：
Step 3.2), the point that sets after binaryzation on the profile of digital hole be expressed as (xj″′,yj″′), if Q (ai,bi,ci) it is number
The quadratic sum of rounded edge range difference is arrived on code holes edge a little, then：
L is the points on digital bore edges in formula (3)；
Step 3.3), with Gauss-Newton nonlinear least square methods object function Q (a are asked to formula (3)i,bi,ci)
Minimum, obtain parameter (ai,bi,ci), according to parameter (ai,bi,ci) calculate digital hole hiCentral coordinate of circle be (Ai,Bi)
It is R with radiusi。
The specific method of method of comparison is in step 4：
With digital hole hiThe radius R of fitting circleiFor radius, the center of circle (A fittedi,Bi) divide for origin, by whole circumference
Into K deciles, the distance difference information of corresponding each point and each point on the circle fitted on digital bore edges is calculated, is made with this
The foundation differentiated for the digital hole true and false, shown in its calculation formula such as formula (4)：
D in formula (4)jIt is the distance of the center of circle that fits to digital bore edges；
If threshold value T1=0.9, if C ＜ T1, then the digital hole is the digital hole forged.
The specific method of K mean cluster method is in step 4：
If the digital hole of digital number is made up of M rows, N row in composition certificate, the central coordinate of circle that each digital hole is fitted is
(Ai,Bi), then the logarithm code holes center of circle is clustered on line direction and column direction according to the following steps, and calculates deviation cluster centre
Difference：
Step 4.1), using the method for K mean cluster, to the ordinate B in all digital holesiM is clustered into the row direction
Class C1、C2、…、CMIf the cluster centre coordinate value of M classes is m1、m2、…、mM, to any one digital hole hiIf, threshold value T2=5,
If its ordinate BiBelong to Cj′Class, then compare BiWith mj′Value, if | Bi-mj′| ＞ T2, then it is assumed that digital hole hiIt is to forge；
Step 4.2), using the method for K mean cluster, to the abscissa A in all digital holesiN is clustered into a column direction
Class C1、C2、…、CNIf the cluster centre coordinate value of N classes is n1、n2、…、nN, to any one digital hole hiIf, its ordinate Ai
Belong to Cj″Class, then compare AiWith nj″Value, if | Ai-nj″| ＞ T2, then it is assumed that digital hole hiIt is to forge.
To the ordinate B in all digital holesiThe calculation formula of the K mean cluster clustered in the row direction is：
M in formula1、m2、…、mMThe cluster centre coordinate value of M classes is represented, by each cluster centre value of continuous iterative modifications, is made
E values are gradually reduced；
To the abscissa A in all digital holesiThe calculation formula of the K mean cluster clustered in a column direction is：
In formula, n1、n2、…、nNThe cluster centre coordinate value of N classes is represented, by each cluster centre value of continuous iterative modifications,
It is gradually reduced E values.
The beneficial effects of the invention are as follows, a kind of digital hole distinguishing method between true and false of certificate, by amplifying the digital hole pattern picture of collection,
Become apparent from digital hole pattern picture.Quantitative calculating can be carried out with the circularity of logarithm code holes, current circularity can be overcome to sentence
Disconnected not accurate enough defect, can differentiate from local shape to the digital hole of forgery.By logarithmic code hole rows direction and
The judgement of column direction, can differentiate to the digital hole of forgery on the whole.
Brief description of the drawings
Fig. 1 is a kind of flow chart of the digital hole distinguishing method between true and false of certificate of the invention.
Embodiment
The present invention is described in detail with reference to the accompanying drawings and detailed description.
A kind of digital hole distinguishing method between true and false of certificate of the present invention, steps flow chart are as shown in figure 1, specific real according to following steps
Apply：
Step 1, the digital bore region of certificate amplified 10 to 20 times, the digital bore region of additional clause is entered using transmission mode
Digitized IMAQ, if the digital hole pattern picture collected is f (x, y)；
Step 2, to the digital hole pattern that is gathered in step 1 as f (x, y) carries out binaryzation and using the side of connected component analysis
Method extracts each digital hole, and specific method is：
Step 2.1), using maximum variance between clusters, to the digital hole pattern in step 1 as f (x, y) carries out binaryzation, if
Image after binaryzation is g (x, y)；
Step 2.2), to step 2.1) in image g (x, y) after binaryzation using 8 neighborhood connected component analysis method,
Each digital hole is extracted, if the digital hole extracted is hi(i=1,2 ..., P), wherein P represents the number in digital hole；
Step 3, each digital hole to being extracted in step 2, the method being fitted using Least Square Circle are fitted per number
Code holes, and the central coordinate of circle and radius geometric parameter of fitting circle are extracted, detailed process is as follows：
Step 3.1), each digital hole is fitted using circle, if i-th of number hole hiThe center of circle of the circle fitted is sat
It is designated as (Ai,Bi), radius be Ri, then circle is expressed as:
Another representation of formula (1) is：x2+y2+aix+biy+ci=0, the conversion of (2) formula (1) and formula (2) is closed
It is to be：
Step 3.2), the point that sets after binaryzation on the profile of digital hole be expressed as (xj″′,yj″′), if Q (ai,bi,ci) it is number
The quadratic sum of rounded edge range difference is arrived on code holes edge a little, then：
L is the points on digital bore edges in formula (3)；
Step 3.3), with Gauss-Newton nonlinear least square methods object function Q (a are asked to formula (3)i,bi,ci)
Minimum, obtain parameter (ai,bi,ci), according to parameter (ai,bi,ci) calculate digital hole hiCentral coordinate of circle be (Ai,Bi)
It is R with radiusi。
Step 4, the central coordinate of circle according to the fitting circle obtained in step 3 and the radius geometric parameter resulting number code holes true and false,
The specific method of the resulting number code holes true and false has two kinds, and first method is method of comparison：The each digital hole extracted in calculation procedure 3
The difference value at edge and fitting circle, and a threshold value is set, if difference value is more than threshold value, judge number of the digital hole as forgery
Code holes；Second method is K mean cluster method：Using the method for K mean cluster, all digital holes in step 3 are fitted
Central coordinate of circle clustered on line direction and column direction, then calculate the difference that cluster centre is deviateed in each digital hole center of circle,
If difference value is more than the threshold value of setting, shows that digital hole is misaligned, belong to the digital hole of forgery, wherein, method of comparison
Specific method is：
With digital hole hiThe radius R of fitting circleiFor radius, the center of circle (A fittedi,Bi) divide for origin, by whole circumference
Into K deciles, the distance difference information of corresponding each point and each point on the circle fitted on digital bore edges is calculated, is made with this
The foundation differentiated for the digital hole true and false, shown in its calculation formula such as formula (4)：
D in formula (4)jIt is the distance of the center of circle that fits to digital hole bore edges；
If threshold value T1=0.9, if C ＜ T1, then the digital hole is the digital hole forged.
The specific method of K mean cluster method is：
If the digital hole of digital number is made up of M rows, N row in composition certificate, the central coordinate of circle that each digital hole is fitted is
(Ai,Bi), then the logarithm code holes center of circle is clustered on line direction and column direction according to the following steps, and calculates deviation cluster centre
Difference：
Step 4.1), using the method for K mean cluster, to the ordinate B in all digital holesiM is clustered into the row direction
Class C1、C2、L、CM, specifically carry out K mean cluster according to formula (5)：
M in formula1、m2、…、mMThe cluster centre coordinate value of M classes is represented, by each cluster centre value of continuous iterative modifications, is made
E values are gradually reduced；
To any one digital hole hiIf, threshold value T2=5, if its ordinate BiBelong to CjClass, then compare BiWith mjValue, if
|Bi-mj| ＞ T2, then it is assumed that digital hole hiIt is to forge；
Step 4.2), using the method for K mean cluster, to the abscissa A in all digital holesiN is clustered into a column direction
Class C1、C2、L、CN, specifically carry out K mean cluster according to formula (6)：
In formula, n1、n2、…、nNThe cluster centre coordinate value of N classes is represented, by each cluster centre value of continuous iterative modifications,
It is gradually reduced E values；
To any one digital hole hiIf, its ordinate AiBelong to CjClass, then compare AiWith njValue, if Ai-nj| ＞ T2, then
Think digital hole hiIt is to forge.
The digital hole distinguishing method between true and false of a kind of certificate of the present invention, from the shape of digital hole in itself, i.e. number bore edges and its
The difference of fitting circle is set out, can be from the local true and false for judging digital hole；Propose simultaneously from digital hole rows direction and column direction
Arrangement is set out, and the true and false in the digital hole from integral position to deviateing cluster centre coordinate differentiates that the method proposed is examined
Consider comprehensive, more preferable identification result can be obtained.
Claims (4)
1. the digital hole distinguishing method between true and false of a kind of certificate, it is characterised in that specifically implement according to following steps：
Step 1, the digital bore region of certificate amplified 10 to 20 times, line number is entered to the digital bore region of additional clause using transmission mode
Word IMAQ, if the digital hole pattern picture collected is f (x, y)；
Step 2, to the digital hole pattern that is gathered in the step 1 as f (x, y) carries out binaryzation and using the side of connected component analysis
Method extracts each digital hole；
Step 3, each digital hole to being extracted in the step 2, the method being fitted using Least Square Circle are fitted per number
Code holes, and extract the central coordinate of circle and radius geometric parameter of fitting circle；
Step 4, the central coordinate of circle according to the fitting circle obtained in the step 3 and the radius geometric parameter resulting number code holes true and false,
The specific method of the resulting number code holes true and false has two kinds, and first method is method of comparison：Calculate extracted in the step 3 every
The difference value at number code holes edge and fitting circle, and set a threshold value, if difference value be more than threshold value, judge the digital hole as
The digital hole of forgery；Second method is K mean cluster method：Using the method for K mean cluster, to all in the step 3
The central coordinate of circle that digital hole is fitted is clustered on line direction and column direction, is then calculated each digital hole center of circle and is deviateed cluster
The difference at center, if difference value is more than the threshold value of a setting, shows that digital hole is misaligned, belongs to the number of forgery
Hole；
Wherein, the specific method of method of comparison is：
With digital hole hiThe radius R of fitting circleiFor radius, i=1,2 ..., P, wherein P represent the number in digital hole, fit
The center of circle (Ai,Bi) be origin, whole circumference is divided into K equal portions, calculate corresponding each point and the circle fitted on digital bore edges
On each point distance difference information, the foundation differentiated in this, as the digital hole true and false, shown in its calculation formula such as formula (4)：
D in formula (4)jIt is the distance of the center of circle that fits to digital bore edges；
If threshold value T1=0.9, if C ＜ T1, then the digital hole is the digital hole forged；
Wherein, the specific method of K mean cluster method is：
If the digital hole of digital number is made up of M rows, N row in composition certificate, the central coordinate of circle that each digital hole is fitted is (Ai,
Bi), then the logarithm code holes center of circle is clustered on line direction and column direction according to the following steps, and calculates the difference for deviateing cluster centre
It is different：
Step 4.1), using the method for K mean cluster, to the ordinate B in all digital holesiM classes C is clustered into the row direction1、
C2、…、CMIf the cluster centre coordinate value of the M classes is m1、m2、…、mM, to any one digital hole hiIf, threshold value T2=5,
If its ordinate BiBelong to Cj′Class, then compare BiWith mj′Value, if | Bi-mj′| ＞ T2, wherein, j '=1,2 ..., M, then it is assumed that
Digital hole hiIt is to forge；
Step 4.2), using the method for K mean cluster, to the abscissa A in all digital holesiN classes C is clustered into a column direction1、
C2、…、CNIf the cluster centre coordinate value of this N class is n1、n2、…、nN, to any one digital hole hiIf, its abscissa AiCategory
In Cj″Class, then compare AiWith nj″Value, if | Ai-nj″| ＞ T2, wherein, j "=1,2 ..., N, then it is assumed that digital hole hiIt is to forge
's.
2. a kind of digital hole distinguishing method between true and false of certificate according to claim 1, it is characterised in that the tool of the step 2
Body method is：
Step 2.1), using maximum variance between clusters, to the digital hole pattern in the step 1 as f (x, y) carries out binaryzation, if
Image after binaryzation is g (x, y)；
Step 2.2), to the step 2.1) in image g (x, y) after binaryzation using 8 neighborhood connected component analysis method,
Extract each digital hole.
3. a kind of digital hole distinguishing method between true and false of certificate according to claim 1, it is characterised in that the tool of the step 3
Body method is：
Step 3.1), each digital hole is fitted using circle, if the central coordinate of circle for the circle that i-th of digital hole is fitted is
(Ai,Bi), radius be Ri, then circle is expressed as:
Another representation of formula (1) is：x2+y2+aix+biy+ci=0, (2)
The transforming relationship of formula (1) and formula (2) is：
Step 3.2), the point that sets after binaryzation on the profile of digital hole be expressed as (xj″′,yj″′), if Q (ai,bi,ci) it is digital hole side
The quadratic sum of rounded edge range difference is arrived on edge a little, then：
L is the points on digital bore edges in formula (3)；
Step 3.3), with Gauss-Newton nonlinear least square methods object function Q (a are asked to formula (3)i,bi,ci) pole
Small value, obtains parameter (ai,bi,ci), according to parameter (ai,bi,ci) calculate digital hole hiCentral coordinate of circle be (Ai,Bi) and half
Footpath is Ri。
4. a kind of digital hole distinguishing method between true and false of certificate according to claim 1, it is characterised in that step 4.1) described in
To the ordinate B in all digital holesiThe calculation formula of the K mean cluster clustered in the row direction is：
M in formula1、m2、…、mMThe cluster centre coordinate value of M classes is represented, by each cluster centre value of continuous iterative modifications, makes E values
It is gradually reduced；
Step 4.2) described in the abscissa A to all digital holesiThe calculation formula of the K mean cluster clustered in a column direction
For：
In formula, n1、n2、…、nNThe cluster centre coordinate value of N classes is represented, by each cluster centre value of continuous iterative modifications, makes E values
It is gradually reduced.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510267706.9A CN104933740B (en) | 2015-05-22 | 2015-05-22 | A kind of digital hole distinguishing method between true and false of certificate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510267706.9A CN104933740B (en) | 2015-05-22 | 2015-05-22 | A kind of digital hole distinguishing method between true and false of certificate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104933740A CN104933740A (en) | 2015-09-23 |
| CN104933740B true CN104933740B (en) | 2017-08-22 |
Family
ID=54120893
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510267706.9A Expired - Fee Related CN104933740B (en) | 2015-05-22 | 2015-05-22 | A kind of digital hole distinguishing method between true and false of certificate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104933740B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107121439A (en) * | 2017-03-23 | 2017-09-01 | 杭州兆深科技有限公司 | The method of inspection and system of a kind of liquid-transfering sucker |
| CN106871786A (en) * | 2017-03-23 | 2017-06-20 | 杭州兆深科技有限公司 | A kind of detection method and system for liquid-transfering sucker port |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101666637A (en) * | 2008-09-03 | 2010-03-10 | 鸿富锦精密工业（深圳）有限公司 | Roundness calculation and display system and method |
| CN102568007A (en) * | 2011-03-22 | 2012-07-11 | 黄海清 | Image processing method performing quasi-circular detection by improving Hough transformation |
| CN103235939A (en) * | 2013-05-08 | 2013-08-07 | 哈尔滨工业大学 | Datum point positioning method based on machine vision |
-
2015
- 2015-05-22 CN CN201510267706.9A patent/CN104933740B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101666637A (en) * | 2008-09-03 | 2010-03-10 | 鸿富锦精密工业（深圳）有限公司 | Roundness calculation and display system and method |
| CN102568007A (en) * | 2011-03-22 | 2012-07-11 | 黄海清 | Image processing method performing quasi-circular detection by improving Hough transformation |
| CN103235939A (en) * | 2013-05-08 | 2013-08-07 | 哈尔滨工业大学 | Datum point positioning method based on machine vision |
Non-Patent Citations (2)
| Title |
|---|
| 一种基于机器视觉的圆型零件检测技术;张宇 等;《计算机工程》;20081031;第34卷(第19期);185-186 * |
| 基于最小二乘拟合的PCB板焊点定位;郭亚娟 等;《计算机与数字工程》;20141130;第42卷(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104933740A (en) | 2015-09-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107967475B (en) | Verification code identification method based on window sliding and convolutional neural network | |
| CN100356388C (en) | Biocharacteristics fusioned identity distinguishing and identification method | |
| CN105095856A (en) | Method for recognizing human face with shielding based on mask layer | |
| CN106384087A (en) | Identity identification method based on multi-layer network human being features | |
| CN101408933A (en) | Method for recognizing license plate character based on wide gridding characteristic extraction and BP neural network | |
| CN104778470A (en) | Character detection and recognition method based on component tree and Hough forest | |
| CN108520212A (en) | Method for traffic sign detection based on improved convolutional neural networks | |
| Lomte et al. | Plant leaves image segmentation techniques: a review | |
| CN103824091A (en) | Vehicle license plate recognition method for intelligent transportation system | |
| CN102411711A (en) | Finger vein recognition method based on individualized weight | |
| CN102411709A (en) | Iris segmentation recognition method | |
| CN111008647B (en) | Sample extraction and image classification method based on void convolution and residual linkage | |
| CN104933740B (en) | A kind of digital hole distinguishing method between true and false of certificate | |
| CN105005565A (en) | Onsite sole trace pattern image retrieval method | |
| CN105404885B (en) | A kind of two dimension character graphics identifying code complex background noise jamming minimizing technology | |
| CN107067407B (en) | Contour detection method based on non-classical receptive field and linear nonlinear modulation | |
| Zhao et al. | Trustworthy authorization method for security in Industrial Internet of Things | |
| CN105488486A (en) | Face recognition method and device for preventing photo attack | |
| CN108831484A (en) | A kind of offline and unrelated with category of language method for recognizing sound-groove and device | |
| CN104463122A (en) | Seal recognition method based on PCNN | |
| Ming | Image segmentation algorithm research and improvement | |
| CN104462853A (en) | Population elite distribution cloud collaboration equilibrium method used for feature extraction of electronic medical record | |
| CN105160336A (en) | Sigmoid function based face recognition method | |
| CN110071884A (en) | A kind of Modulation Recognition of Communication Signal method based on improvement entropy cloud feature | |
| CN105469097A (en) | Transformer station feature extraction method based on nerve network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170822 Termination date: 20210522 |