CN104809451B - A kind of person's handwriting identification system based on stroke curvature measuring - Google Patents
A kind of person's handwriting identification system based on stroke curvature measuring Download PDFInfo
- Publication number
- CN104809451B CN104809451B CN201510250547.1A CN201510250547A CN104809451B CN 104809451 B CN104809451 B CN 104809451B CN 201510250547 A CN201510250547 A CN 201510250547A CN 104809451 B CN104809451 B CN 104809451B
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- stroke
- handwriting
- msup
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/333—Preprocessing; Feature extraction
- G06V30/347—Sampling; Contour coding; Stroke extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/142—Image acquisition using hand-held instruments; Constructional details of the instruments
- G06V30/1423—Image acquisition using hand-held instruments; Constructional details of the instruments the instrument generating sequences of position coordinates corresponding to handwriting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Abstract
The invention discloses a kind of person's handwriting identification system based on stroke curvature measuring, including writing sample database module, IMAQ recording module, pretreatment module, signature analysis extraction module, feature square processing module, similarity measurement module, result output feedback module.Person's handwriting false distinguishing process is that pretreatment module pre-processes to the image of the real-time collection of image capture module, the extraction of signature analysis extraction module is horizontal, vertical, skims, the characteristic value of right-falling stroke four direction stroke, feature square processing module carries out the processing of stroke feature value and generates handwriting characteristic square, similarity measurement module needs the similitude of false distinguishing handwriting characteristic square and handwriting characteristic square in database using the method measurement of vector angle similarity measurement, as a result exports feedback module and exports corresponding result.The person's handwriting identification system of the present invention, can quickly carry out the person's handwriting false distinguishing of text locating, and false distinguishing result is stable, objective.
Description
Technical field
The present invention relates to a kind of person's handwriting identification system based on stroke curvature measuring, belongs to Digital Image Processing and is reflected with person's handwriting
Determine technical field.
Background technology
The rapid development of information technology brings great convenience to daily life, at the same time to personal identification
Accurate certification, protection information safety are carried out then into a key issue of current information age urgent need to resolve.Person's handwriting is
A kind of more stable behavioural characteristic of people, the certification for carrying out personal identification using handwriting have the non-property invaded (or non-touch
Property), the features such as being easily obtained, easily making one to receive, be a kind of identification of means of important personal identification.
Over the years, person's handwriting false distinguishing field uses traditional artificial person's handwriting false distinguishing more, but this method is gradual in practical operation
Drawback and defect are exposed, qualification result is adversely affected：As counterfeit institution lacks mutual cooperation mechanism, false distinguishing personnel money
Lattice assert that standard is lack of standardization, false distinguishing is horizontal uneven etc. can all adversely affect to qualification result.Therefore calculating is utilized
Machine automation, the mode of standardization handle the field of this non-engineering originally of person's handwriting false distinguishing, and the development to the field has very
Big impetus.
Person's handwriting false distinguishing at present can be divided into online and two kinds of off line by the difference of practical manner, widely should both have
With background, can be played a significant role in fields such as finance, insurance, the criminal investigation of police and judicial department and court judgments.
Online person's handwriting authentication detection technology is very ripe for reality and has been enter into practical stage, but off line person's handwriting false distinguishing is because of nothing
Stroke order, writing speed, the multidate information such as pressure of wieling the pen when the online person's handwriting false distinguishing of the image of Buddha gets writer's writing like that,
Undoubtedly can be bigger in the difficulty of false distinguishing.
Existing off line person's handwriting false distinguishing method is to be directed to specific character mostly, and its essence is turn the false distinguishing of personal person's handwriting
The false distinguishing work of personal specific character person's handwriting is turned to, it is necessary to specific character be gathered, to specific sample in the application process of reality
This undue dependence causes this kind of false distinguishing method to lack versatility and robustness.So off line person's handwriting false distinguishing for text locating
System is not also overripened at present, but also becomes study hotspot instantly.
The content of the invention
The technical problems to be solved by the invention are for the technology in the popular text locating person's handwriting false distinguishing of current social
Vacancy, there is provided a kind of person's handwriting identification system based on stroke curvature measuring, quickly realize the person's handwriting false distinguishing of text locating, false distinguishing knot
Fruit is stable, objective.
In order to solve the above technical problems, the present invention provides a kind of system of the person's handwriting false distinguishing based on stroke curvature measuring, its
It is characterised by, including following functions module：
Handwriting characteristic data library module：Prestore the digitization information of individual subscriber text locating, including handwriting characteristic
Matrix, personal information；
Writing sample IMAQ recording module：Writing sample to be identified is gathered using high photographing instrument, generates writing sample
Image；
Writing sample image pre-processing module：False distinguishing writing sample image, which pre-processes, to be treated to typing；
Signature analysis extraction module：Writing sample image after being handled through writing sample image pre-processing module is carried out
The extraction of stroke feature value matrix, all stroke feature value matrixs of acquisition are collected into composition stroke feature data group；
Handwriting characteristic matrix disposal module：Stroke feature data group is taken in pen of unidirectional stage casing average as the direction
Mark characteristic value is preserved to handwriting characteristic matrix, makees identical to the horizontal, vertical of stroke feature data group, slash, right-falling stroke four direction successively
Processing, obtains the handwriting characteristic matrix of four direction, then the value of handwriting characteristic matrix is normalized between -1 to 1；
Similarity measurement module：The handwriting characteristic matrix for treating false distinguishing writing sample image zooming-out is calculated using Cosin method
With the similarity of each user handwriting eigenmatrix in handwriting data storehouse；
As a result feedback module is exported：The similarity result that similarity measurement module calculates and given threshold are compared
Compared with obtaining person's handwriting true and false information, intuitively feed back the true and false information of person's handwriting to operator by operation interface.
The foregoing person's handwriting identification system based on stroke curvature measuring, it is characterised in that：The writing sample IMAQ
Recording module, to carrying out IMAQ comprising the writing sample no less than 30 characters, randomly selects three continuously using high photographing instrument
Character generates a writing sample image, and symbiosis is into ten width writing sample images.
The foregoing person's handwriting identification system based on stroke curvature measuring, it is characterised in that：The writing sample image is located in advance
Reason module sample image of identifying the handwriting first is pre-processed substantially, and pre-processing substantially includes gray processing, binaryzation and dimensional standard
Change, then the writing sample image to completing basic pretreatment carries out closure operation, extracts most representational one direction stroke, it is right
One direction stroke pattern picture carries out rapid refinement processing, image is only included one direction stroke skeleton；
The foregoing person's handwriting identification system based on stroke curvature measuring, it is characterised in that：Feedback mould is exported in the result
It is real feedback information in system interface output false distinguishing result, when less than setting when similarity reaches given threshold in block
It is false feedback information in system interface output false distinguishing result, and peripheral circuit string is connect by USB interface of computer during threshold value
Mouth sends order, starts the buzzer and warning lamp of alarm peripheral circuit.
The beneficial effect that the present invention is reached：
Person's handwriting identification system proposed by the present invention can intelligently solve the problems, such as the person's handwriting false distinguishing of text locating, operator
After member is by gathering a small amount of script character (at least 30) typing to person's handwriting identification system, it is possible to directly obtain false distinguishing knot
Fruit.Compared with traditional artificial person's handwriting false distinguishing, have it is simple to operate, real-time is good, feedback is directly perceived, result is objective, false distinguishing is accurate
The features such as rate is high.
Brief description of the drawings
Fig. 1 is present system function structure chart；
Fig. 2 is system mock-up figure；
Fig. 3 is system process chart；
Fig. 4 is generation handwriting characteristic square flow chart.
Embodiment
The person's handwriting identification system based on stroke curvature measuring of the present invention includes operation system of software and hardware effort system,
Whole system software module structure is as shown in figure 1, including handwriting data library module, writing sample IMAQ recording module, pen
Mark sample image pretreatment module, signature analysis extraction module, handwriting characteristic square processing module, similarity measurement module, as a result
Export feedback module：
1) handwriting data library module：Prestore individual subscriber text locating representative after the system that has been subjected to is handled
Digitization information, including handwriting characteristic square, personal information (such as name, sex, ID card No.), follow-up false distinguishing process
In can call these information to identify the writing sample image that need to discern the false from the genuine.
2) writing sample IMAQ recording module：By 5,000,000 pixels from focusing high photographing instrument, system is operated by software
It completes the real-time typing work of data progress of sample person's handwriting for system control, and the writing sample image of generation is carried out into analog-to-digital conversion
Transmission is preserved to the pretreatment module of system afterwards, the processing for subsequent module.
3) writing sample image pre-processing module：False distinguishing writing sample image is treated to input system with mathematical morphology
The work pre-processed, in random text (font is unlimited), the writing sample image of three continuation characters is extracted at random,
The process of pretreatment specifically includes the processing of writing sample image gray processing, binary conversion treatment, size normalised processing, micronization processes
And closure operation this five big specific steps, only the most representational stroke bone of one direction is included in pretreated handwriting image
Frame, the signature image after pretreatment is used further to the processing of subsequent module.
4) signature analysis extraction module：To after pretreatment writing sample image carry out feature extraction, it is horizontal, vertical, skim,
Right-falling stroke four direction extracts representative stroke of the maximum one direction stroke of write area as the direction respectively, then with form
The reconstruct that circle reconstructing method is justified to the representative stroke of horizontal, vertical, slash, right-falling stroke four direction respectively is learned, extracts four direction pen
The curvature of circle is drawn as stroke feature value, same processing is done to all sample images successively, by all strokes of acquisition
Eigenvalue matrix collects composition stroke feature data group.
5) handwriting characteristic matrix disposal module：Take stroke feature data group in unidirectional stage casing average as the direction
Handwriting characteristic value is preserved to final handwriting characteristic matrix, makees identical processing to the four direction of stroke feature data group successively,
Four direction handwriting characteristic value composition handwriting characteristic matrix is obtained, then between the value of handwriting characteristic matrix is normalized into -1 to 1,
The processing of subsequent module is used further to after handwriting characteristic matrix is preserved.
6) feature similarity measurement module：Calculated using Cosin method and treat that the person's handwriting of false distinguishing writing sample image zooming-out is special
The similarity of matrix and each user handwriting eigenmatrix in handwriting data storehouse is levied, gained handwriting characteristic matrix similarity is in certain threshold
The person's handwriting of the user is considered in the range of value.Included angle cosine similitude, also referred to as cosine similarity or COS distance, it is to use
Two vectorial angle cosine values judge two interindividual variation sizes in vector space.
(7) result output feedback module：The feedback operation of writing sample image false distinguishing information is completed, its operation principle is to work as
When similarity is higher than given threshold, system is true in computer interface feedback false distinguishing result, when similarity is less than given threshold
When, system is falseness in computer interface feedback false distinguishing result, and completes alert operation by peripheral circuit.
The specific works step of person's handwriting identification system of the invention based on stroke curvature measuring is followed successively by：
As shown in Fig. 2 controlling high photographing instrument by operation system of software first, false distinguishing is treated in collection to system mock-up figure in real time
Writing sample, writing sample collection needed for person's handwriting data include a small amount of hand-written character, then with often continuous three
Character forms a writing sample image (content of text is uncorrelated), ten sample images is formed altogether, by the writing sample of collection
Image extracts handwriting characteristic composition handwriting characteristic matrix after being pre-processed, after eigenmatrix of identifying the handwriting is normalized,
With angle similarity measurement method measurement and the similarity of sample, as the measurement for judging two interindividual variation sizes, grasping
Make to feed back false distinguishing information on interface, while operation system of software controls peripheral output circuit to carry out corresponding feedback operation.Entirely
The workflow of person's handwriting identification system, coordinate software systems to complete the false distinguishing of writing sample in real time for hardware system and work, can be fast
Speed treats the true and false information of false distinguishing writing sample to operator's feedback.
The system operation specific workflow of the present invention is as shown in Figure 3：
(1) when the true and false of writing sample need to differentiate, first by 5,000,000 pixels from focusing high photographing instrument, to sample pen
The data of mark carries out real-time typing work, and its process is that the writing sample data comprising a certain amount of hand-written character is positioned over into height
The target area of instrument is clapped, writing sample data is that black signature pen is write on white paper, and the size of sample person's handwriting is not wanted
Ask, maximum be no more than A3 paper sizes, computer software interface startup person's handwriting identification system after, by USB interface to
High photographing instrument transmits the order of captured in real-time, and high photographing instrument is by from the instant shooting work character for completing writing sample of focusing, passing through
After software systems complete analog-to-digital conversion work, take three continuation characters to generate a writing sample image at random, generate about ten width
Sample image, the writing sample image transmitting of generation is preserved to the writing sample pretreatment module of system, for subsequent characteristics
Vector analysis extraction module circular treatment.
(2) in writing sample image pre-processing module：
If the coloured image being stored in image acquisition data memory block is I1, replicate portion and deposit in data storage area simultaneously
Image gray processing is carried out to it to obtain gray level image I2；
Using maximum variance between clusters (otsu algorithms) to gray level image I2Enter row threshold division, obtain target binary picture
As I3；
By target binary image I3It is 600*300 (units of quantity to carry out size normalised processing：Pixel) size rule
Then image I4；
For regular image I4, selection corrosion factor is linear pattern, and length is 8 pixels, then respectively choose 0 °, 90 °,
45 °, 135 ° of four directions carry out respectively closure operation obtain only comprising one direction stroke horizontal stroke pattern as I4 is horizontal, vertical pen
Draw as I4 is perpendicular, skim to stroke pattern as I4 skim, press down to stroke pattern as I4 right-falling strokes, respectively to horizontal stroke pattern as I4 is horizontal, vertical stroke pattern picture
I4 is perpendicular, skim to stroke pattern as I4 skim, press down to stroke pattern as I4 right-falling strokesIn all strokes calculate connected domain area, select writing sample figure
The maximum one direction stroke of connected domain area saves as the representative stroke pattern picture of the direction as in, obtain successively it is horizontal, vertical, skim,
Press down the representative stroke pattern picture of four direction；Wherein representative stroke refers to that horizontal, vertical writer's writing, slash, right-falling stroke can be shown
The stroke of custom and feature during one direction stroke, 0 °, 90 °, 45 °, 135 ° of four directions correspond to respectively Chinese character it is horizontal, vertical, skim, right-falling stroke
Four stroke directions；
Rapid refinement processing is carried out to representative stroke pattern picture that is horizontal, vertical, skimming, press down four direction respectively, after the completion of processing
Horizontal stroke skeleton image I5 is horizontal, vertical stroke skeleton image I5 is perpendicular, skim to stroke skeleton image I5 skim, press down to stroke skeleton image
I5 right-falling strokesTemporarily it is stored in data storage area and is used for subsequent treatment.
The rapid refinement algorithm is to carry out target based on mathematical morphology successively to strip, rapid extraction image framework, its
Process is order, backward progress twice sweep, it is determined that each number of plies of the point in image, then judge it is to belong to border by the number of plies
Point or skeleton, (x, y) are current point, and P (x, y) is current point pixel value, it is specified that black level value is 1, white value 0, carry out with
Lower three steps：
(a) from top to bottom, each point is from left to right scanned successively, if current point is stain, surrounds at 4 points in judgement
Situation, embracing layer numerical value F (x, y) in definition：
As p (x, y)=1；
F (x, y)=Min [F (x-1, y+1), F (x-1, y), F (x-1, y-1), F (x, y-1)]+1
As p (x, y)=0；
F (x, y)=0
(b) from top to bottom, scan each point successively from right to left, if current point is stain, judge lower 4 points of encirclement
Situation, define lower embracing layer numerical value G (x, y)：
As p (x, y)=1；
G (x, y)=Min [G (x+1, y-1), G (x+1, y), G (x+1, y+1), G (x, y+1)]+1
As p (x, y)=0；
G (x, y)=0
(c) actual layer number M (x, y) is set, the actual layer number each put is the minimum value in the both the above number of plies, i.e.,：
M (x, y)=Min [F (x, y), G (x, y)]
Again from top to bottom, each point is from left to right scanned, then judges the number of plies situation of 8 points around the point, if currently
Point is the maximum of points in 9 points, then remains, if not maximum of points, then delete.It can be completed soon by above step
Speed refinement work.
(3) in signature analysis extraction module, horizontal stroke skeleton image I is taken5 is horizontal, in horizontal stroke skeleton image I5 is horizontalIn
Chosen on transverse direction stroke and halve three points, the reconstruct justified, then extraction calculates the radius of transverse direction circle, by half path length
Transverse direction stroke feature value of the curvature reciprocal of degree as the personal person's handwriting transverse direction invariant feature of reflection, and preserve；Again to vertical
Stroke skeleton image I5 is perpendicular, skim to stroke skeleton image I5 skim, press down to stroke skeleton image I5 right-falling strokesSame processing is carried out successively, respectively
Obtain all directions stroke feature value；Same treatment is made to ten width images successively, obtains the stroke feature of every width writing sample image
Value.
Choosing the step of halving at three is：
A) from top to bottom, from left to right each point on scan image successively, if current point (x, y) is stain, then sentences
The pixel value of 8 points around the disconnected point, when surrounding has the pixel value of 7 points to be 0, the coordinate of this point is extracted, if G is extraction
The mark of point, G=1 extract the point coordinates, and as P (x, y)=1, the G values of the point are calculated using XOR：
Wherein,XOR is represented, it is exactly in the eight direction pixels put around satisfaction extracted to calculate G values as 1
Stain quantity is 1 condition, that is, the end points of stroke；
B) point A (x are proposed by the first step0,y0), B (x1,y1) two camber line extreme coordinates, connect A, B at 2 points, obtain straight
Line l0, make straight line l0Vertical line l1, take vertical line l1With the intersection point C (x of camber line2,y2), then A, B, C are exactly the bisection three of camber line
Point；
C) reconstruct justified after 3 points of A, B, C is obtained, then extraction calculates the radius of transverse direction circle, by radius length r
C reciprocal the stroke feature value of the direction, circle are saved as the transverse features value for reflecting personal person's handwriting transverse direction invariant feature
Point D (x3,y3), curvature c calculating process it is as follows：
(4) in handwriting characteristic square processing module, combine extraction handwriting characteristic value for stroke feature value and person's handwriting is special
Matrix data normalized is levied, its process is that the unidirectional stroke feature value of all writing sample images first is collected into composition to be somebody's turn to do
The stroke feature data group in direction, then by the one direction stroke feature data group greatly data at end and the minimum end of data group
Reject, take final characteristic value of the data group stage casing mean data as direction stroke, reject the flow that extreme value takes stage casing average
As shown in figure 4, final handwriting characteristic matrix is formed, utilizes normalized side after the final characteristic value of acquirement four direction successively
Method, between the value of handwriting characteristic matrix is normalized into -1 to 1, handwriting characteristic matrix is preserved to similarity measurement module, used
In the processing of subsequent module.
(5) in feature similarity measurement module, co sinus vector included angle similarity measurement, if obtained by by above-mentioned steps
By c1, c2, c3, c4The handwriting characteristic matrix-vector of composition is expressed as F=(c1,c2,c3,c4), corresponding person's handwriting in handwriting data storehouse
Eigenmatrix vector is G=(x1,x2,x3,x4), for the two four dimensional vectors, judged between the two using Cosin method
Similarity degree, i.e.,：
Here n is taken to represent n-th of single stroke direction, n maximums are 4, take the number of degrees size of the angle theta between two matrix-vectors
As the reference value for judging its similarity, the angle between vector is bigger, then it is corresponding with database to represent person's handwriting to be identified
Person's handwriting similarity is smaller, if otherwise angle it is smaller, represent person's handwriting to be identified person's handwriting similarity corresponding with database compared with
Greatly, if M is person's handwriting to be identified person's handwriting similarity corresponding with database, then defining M is：
Person's handwriting similarity M values are calculated, person's handwriting similarity M values are made comparisons with given threshold T and (take T=95%), if being more than
The threshold value is then accredited as the person's handwriting of same people, if being less than the threshold value, is accredited as the person's handwriting of other people forgeries.
(6) peripheral circuit in result output feedback module obtains software according to the court verdict of similarity measurement module
The dependent instruction of system, it is real feedback in system interface output false distinguishing result if qualification result is the person's handwriting of same people
Information, if qualification result is false feedback information in system interface output false distinguishing result not to be the person's handwriting of same people, and
Peripheral circuit serial ports is connect by USB interface of computer simultaneously and sends order startup alarm peripheral circuit, utilizes the honeybee of peripheral circuit
Ring device and warning lamp real-time reminding system operator.Table 1 establishes sample table for database, according to numbering, name, age, body
Part card number waits information search initial data.
Table 1
The present invention is disclosed with preferred embodiment above, so it is not intended to limiting the invention, all to use equivalent substitution
Or the technical scheme that equivalent transformation mode is obtained, it is within the scope of the present invention.
Claims (9)
1. a kind of person's handwriting identification system based on stroke curvature measuring, it is characterised in that including following functions module：
Handwriting data library module：Prestore the digitization information of individual subscriber text locating, including handwriting characteristic matrix, individual
Information；
Writing sample IMAQ recording module：Writing sample to be identified is gathered using high photographing instrument, generates writing sample image；
Writing sample image pre-processing module：False distinguishing writing sample image, which pre-processes, to be treated to typing；
Signature analysis extraction module：Stroke is carried out to the writing sample image after being handled through writing sample image pre-processing module
The extraction of eigenvalue matrix, all stroke feature value matrixs of acquisition are formed into stroke feature data group；
Handwriting characteristic matrix disposal module：Take stroke feature data group special as the person's handwriting of the direction in unidirectional stage casing average
Value indicative is preserved to handwriting characteristic matrix, makees identical processing to the horizontal, vertical of stroke feature data group, slash, right-falling stroke four direction successively,
The handwriting characteristic matrix of four direction is obtained, then the value of handwriting characteristic matrix is normalized between -1 to 1；
Similarity measurement module：The handwriting characteristic matrix and pen for treating false distinguishing writing sample image zooming-out are calculated using Cosin method
The similarity of each user handwriting eigenmatrix in mark database；
As a result feedback module is exported：The similarity result calculated according to similarity measurement module, intuitively feedback is corresponding believes
Breath.
2. the person's handwriting identification system according to claim 1 based on stroke curvature measuring, it is characterised in that：The person's handwriting sample
This IMAQ recording module carries out IMAQ, random choosing to the writing sample comprising no less than 30 characters using high photographing instrument
Three continuation characters are taken to generate a writing sample images, symbiosis is into ten width writing sample images.
3. the person's handwriting identification system according to claim 1 based on stroke curvature measuring, it is characterised in that：The person's handwriting sample
This image pre-processing module sample image of identifying the handwriting first is pre-processed substantially, and basic pretreatment includes gray processing, binaryzation
With it is size normalised, then the writing sample image to completing basic pretreatment carries out closure operation, extracts most representational list
Direction stroke, rapid refinement processing is carried out to one direction stroke pattern picture, image is only included one direction stroke skeleton.
4. the person's handwriting identification system according to claim 3 based on stroke curvature measuring, it is characterised in that：In writing sample
In image pre-processing module：
If the coloured image being stored in image acquisition data memory block is I1, replicate portion and deposit in data storage area and to it
Image gray processing is carried out to obtain gray level image I2；
Using maximum variance between clusters to gray level image I2Enter row threshold division, obtain target binary image I3；
By target binary image I3Carry out regular image I of the size normalised processing for 600*300 pixel size4；
For regular image I4, selection corrosion factor is linear pattern, and length is 8 pixels, then respectively choose 0 °, 90 °, 45 °,
135 ° of four directions carry out closure operation and obtain only including the horizontal stroke pattern of one direction stroke as I respectively4 is horizontal, vertical stroke pattern
As I4 is perpendicular, skim to stroke pattern as I4 skim, press down to stroke pattern as I4 right-falling strokes, respectively to horizontal stroke pattern as I4 is horizontal, vertical stroke pattern is as I4 is perpendicular, skim
To stroke pattern as I4 skim, press down to stroke pattern as I4 right-falling strokesIn all strokes calculate connected domain area, it is maximum to select connected domain area
One direction stroke saves as the representative stroke pattern picture of the direction, obtains representative pen that is horizontal, vertical, skimming, press down four direction successively
Picture picture；Described 0 °, 90 °, 45 °, 135 ° of four directions correspond to respectively Chinese character it is horizontal, vertical, skim, right-falling stroke four stroke directions；
Rapid refinement processing, the horizontal stroke after the completion of processing are carried out to representative stroke pattern picture that is horizontal, vertical, skimming, press down four direction respectively
To stroke skeleton image I5 is horizontal, vertical stroke skeleton image I5 is perpendicular, skim to stroke skeleton image I5 skim, press down to stroke skeleton image I5 right-falling strokes
Temporarily it is stored in data storage area and is used for subsequent treatment.
5. the person's handwriting identification system according to claim 3 based on stroke curvature measuring, it is characterised in that：It is described quick thin
Changing algorithmic procedure is：Sequentially, backward carries out twice sweep, it is determined that the number of plies of the point in each image, then be by number of plies judgement
Belong to boundary point or skeleton, (x, y) is current point, P (x, y) be current point pixel value, it is specified that black level value is 1, white value is
0, carry out following three steps：
(a) from top to bottom, each point is from left to right scanned successively, if current point is stain, 4 points of feelings are surrounded in judgement
Condition, embracing layer numerical value F (x, y) in definition：
As p (x, y)=1；
F (x, y)=Min [F (x-1, y+1), F (x-1, y), F (x-1, y-1), F (x, y-1)]+1
As p (x, y)=0；
F (x, y)=0；
(b) from top to bottom, scan each point successively from right to left, if current point is stain, judge the lower feelings for surrounding at 4 points
Condition, define lower embracing layer numerical value G (x, y)：
As p (x, y)=1；
G (x, y)=Min [G (x+1, y-1), G (x+1, y), G (x+1, y+1), G (x, y+1)]+1
As p (x, y)=0；
G (x, y)=0
(c) actual layer number M (x, y) is set, the actual layer number each put is the minimum value in the both the above number of plies, i.e.,：
M (x, y)=Min [F (x, y), G (x, y)]
Again from top to bottom, each point is from left to right scanned, then judges the number of plies situation of 8 points around the point, if current point is 9
Maximum of points in individual point, then remain, if not maximum of points, then delete.
6. the person's handwriting identification system according to claim 1 based on stroke curvature measuring, it is characterised in that：In signature analysis
In extraction module, horizontal stroke skeleton image I is taken5 is horizontal, in horizontal stroke skeleton image I5 is horizontalMiddle horizontal square is halved to selection on stroke
Three points, the reconstruct justified, then extraction calculates the radius of transverse direction circle, using the curvature reciprocal of radius length as reflection
The transverse direction stroke feature value of people's person's handwriting transverse direction invariant feature, and preserve；Again to vertical stroke skeleton image I5 is perpendicular, skim to pen
Draw skeleton image I5 skim, press down to stroke skeleton image I5 right-falling strokesSame processing is carried out successively, respectively obtains all directions stroke feature value；
Identical processing is made to ten width images successively, obtains the stroke feature value of every width writing sample image.
7. the person's handwriting identification system according to claim 6 based on stroke curvature measuring, it is characterised in that：Choose and halve
Three the step of are：
A) from top to bottom, from left to right each point on scan image successively, if current point (x, y) is stain, then judges this
The pixel value of 8 points around point, when surrounding has the pixel value of 7 points to be 0, the coordinate of the point is extracted, if G is the mark of extraction point
Know, G=1 extracts the point coordinates, as P (x, y)=1, the G values of the point is calculated using XOR, P (x, y) is current point
Pixel value：
Wherein,Represent XOR；
B) point A (x are extracted by the first step0,y0), B (x1,y1) two camber line extreme coordinates, connect A, B at 2 points, obtain straight line
l0, make straight line l0Vertical line l1, take vertical line l1With the intersection point C (x of camber line2,y2), then point A, B, C is exactly the bisection three of camber line
Point；
C) reconstruct justified after 3 points of A, B, C is obtained, then extraction calculates the radius of transverse direction circle, by falling for radius length r
Transverse features values of the number c as the personal person's handwriting transverse direction invariant feature of reflection, and save as the stroke feature value of the direction, round dot
D(x3,y3), curvature c calculating process it is as follows：
<mrow>
<msub>
<mi>x</mi>
<mn>3</mn>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mn>2</mn>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mo>{</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>}</mo>
<mo>-</mo>
<mn>2</mn>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>2</mn>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mo>{</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>}</mo>
</mrow>
<mrow>
<mn>4</mn>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>2</mn>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mn>4</mn>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>2</mn>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mrow>
<mrow>
<msub>
<mi>y</mi>
<mn>3</mn>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mn>2</mn>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<mo>(</mo>
<mrow>
<msub>
<mi>x</mi>
<mn>2</mn>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
</mrow>
<mo>)</mo>
<mo>&times;</mo>
<mo>{</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>}</mo>
<mo>-</mo>
<mn>2</mn>
<mo>&times;</mo>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mo>{</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>}</mo>
</mrow>
<mrow>
<mn>4</mn>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>2</mn>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mn>4</mn>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>2</mn>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>1</mn>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mrow>
<mrow>
<mi>c</mi>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mi>r</mi>
</mfrac>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<msqrt>
<mrow>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>3</mn>
</msub>
<mo>-</mo>
<msub>
<mi>x</mi>
<mn>0.</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mn>3</mn>
</msub>
<mo>-</mo>
<msub>
<mi>y</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mfrac>
<mo>.</mo>
</mrow>
8. the person's handwriting identification system according to claim 7 based on stroke curvature measuring, it is characterised in that：It is similar in feature
In property metric module, if resulting by transverse direction stroke skeleton curvature c1, erect direction stroke skeleton curvature c2, skim direction stroke
Skeleton curvature c3, press down direction stroke skeleton curvature c4The handwriting characteristic matrix-vector of composition is expressed as F=(c1,c2,c3,c4), pen
Corresponding handwriting characteristic matrix-vector is G=(x in mark database1,x2,x3,x4), judged between the two using Cosin method
Similarity degree, i.e.,：
<mrow>
<mi>c</mi>
<mi>o</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>&theta;</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>c</mi>
<mi>k</mi>
</msub>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
</mrow>
<mrow>
<msqrt>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>c</mi>
<mi>k</mi>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
<msqrt>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>x</mi>
<mi>k</mi>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
Wherein n represents n-th of single stroke direction, and n maximums are 4, takes the number of degrees size of the angle theta between two matrix-vectors to be used as and sentences
The reference value of both disconnected similarities, the angle between vector is bigger, then it is corresponding with handwriting data storehouse to represent person's handwriting to be identified
Person's handwriting similarity is smaller, if angle on the contrary is smaller, it is similar to represent person's handwriting to be identified person's handwriting corresponding with handwriting data storehouse
Spend larger, if M is person's handwriting to be identified person's handwriting similarity corresponding with handwriting data storehouse, then defining M is：
<mrow>
<mi>M</mi>
<mo>=</mo>
<mo>&lsqb;</mo>
<mn>180</mn>
<mo>-</mo>
<mi>arccos</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>c</mi>
<mi>k</mi>
</msub>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
</mrow>
<mrow>
<msqrt>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>c</mi>
<mi>k</mi>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
<msqrt>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msubsup>
<mi>x</mi>
<mi>k</mi>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
<mo>&times;</mo>
<mn>100</mn>
<mi>%</mi>
</mrow>
Person's handwriting similarity M values are calculated, by person's handwriting similarity M values with being made comparisons in advance by testing drawn experimental threshold values T, if greatly
The person's handwriting of same people is then accredited as in the threshold value, if being less than the threshold value, is accredited as the person's handwriting of other people forgeries.
9. the person's handwriting identification system according to claim 1 based on stroke curvature measuring, it is characterised in that：In the result
Export in feedback module, be real in system interface output false distinguishing result when person's handwriting similarity reaches experiment experience threshold value
Feedback information, it is false feedback information in system interface output false distinguishing result, and pass through meter when less than experiment experience threshold value
Calculation machine USB interface connects peripheral circuit serial ports and sends order, starts the buzzer and warning lamp of alarm peripheral circuit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510250547.1A CN104809451B (en) | 2015-05-15 | 2015-05-15 | A kind of person's handwriting identification system based on stroke curvature measuring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510250547.1A CN104809451B (en) | 2015-05-15 | 2015-05-15 | A kind of person's handwriting identification system based on stroke curvature measuring |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104809451A CN104809451A (en) | 2015-07-29 |
| CN104809451B true CN104809451B (en) | 2017-12-26 |
Family
ID=53694262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510250547.1A Active CN104809451B (en) | 2015-05-15 | 2015-05-15 | A kind of person's handwriting identification system based on stroke curvature measuring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104809451B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292213B (en) * | 2016-03-30 | 2020-04-14 | 中国刑事警察学院 | Handwriting quantitative inspection and identification method |
| CN107256389A (en) * | 2017-05-26 | 2017-10-17 | 山西农业大学 | The certification recognized based on handwriting characteristic and encryption method |
| CN107392136B (en) * | 2017-07-17 | 2020-07-17 | 河海大学常州校区 | Signature authenticity identification system based on feature self-adaptive oscillation attenuation |
| CN110222660B (en) * | 2019-06-12 | 2022-09-23 | 河海大学常州校区 | Signature authentication method and system based on dynamic and static feature fusion |
| CN110544073A (en) * | 2019-09-06 | 2019-12-06 | 河北江著科技有限公司 | Entry and exit information acquisition method and device and terminal equipment |
| CN112990000A (en) * | 2021-03-11 | 2021-06-18 | 中国刑事警察学院 | Electronic signature handwriting characteristic typing and characteristic chain comparison method |
| CN114580429A (en) * | 2022-01-26 | 2022-06-03 | 云捷计算机软件(江苏)有限责任公司 | Artificial intelligence-based language and image understanding integrated service system |
| CN116311543B (en) * | 2023-02-03 | 2024-03-08 | 汇金智融(深圳)科技有限公司 | Handwriting analysis method and system based on image recognition technology |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101384211A (en) * | 2006-03-14 | 2009-03-11 | 奥林巴斯医疗株式会社 | Image analyzing device and image analyzing method |
| CN101895626A (en) * | 2010-05-19 | 2010-11-24 | 济南北秀信息技术有限公司 | Handwriting analyzing device and method thereof for mobile phone |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6189116B2 (en) * | 2013-07-09 | 2017-08-30 | ソニーセミコンダクタソリューションズ株式会社 | Image processing apparatus, image processing method, and program |
-
2015
- 2015-05-15 CN CN201510250547.1A patent/CN104809451B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101384211A (en) * | 2006-03-14 | 2009-03-11 | 奥林巴斯医疗株式会社 | Image analyzing device and image analyzing method |
| CN101895626A (en) * | 2010-05-19 | 2010-11-24 | 济南北秀信息技术有限公司 | Handwriting analyzing device and method thereof for mobile phone |
Non-Patent Citations (1)
| Title |
|---|
| 基于轮廓方向特征的脱机手写中文笔迹鉴别;刘海 等;《电视技术》;20111231;第35卷(第21期);第131-135页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104809451A (en) | 2015-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104809451B (en) | A kind of person's handwriting identification system based on stroke curvature measuring | |
| Yang et al. | MTD-Net: Learning to detect deepfakes images by multi-scale texture difference | |
| Shi et al. | Automatic road crack detection using random structured forests | |
| Zhao et al. | Preprocessing and postprocessing for skeleton-based fingerprint minutiae extraction | |
| CN103324944B (en) | A kind of based on SVM with the false fingerprint detection method of rarefaction representation | |
| CN106778586A (en) | Offline handwriting signature verification method and system | |
| CN107657241B (en) | Signature pen-oriented signature authenticity identification system | |
| Peralta et al. | Minutiae filtering to improve both efficacy and efficiency of fingerprint matching algorithms | |
| CN107392136B (en) | Signature authenticity identification system based on feature self-adaptive oscillation attenuation | |
| CN108846269A (en) | One kind is towards manifold identity identifying method and identification authentication system | |
| Liu et al. | Fingerprint presentation attack detector using global-local model | |
| Zhao et al. | Generalized symmetric pair model for action classification in still images | |
| Yuan et al. | Fingerprint liveness detection using histogram of oriented gradient based texture feature | |
| Sun et al. | Tattoo detection and localization using region-based deep learning | |
| CN108921006B (en) | Method for establishing handwritten signature image authenticity identification model and authenticity identification method | |
| CN116631068B (en) | Palm vein living body detection method based on deep learning feature fusion | |
| CN110222660B (en) | Signature authentication method and system based on dynamic and static feature fusion | |
| Jain et al. | Online signature verification using energy, angle and directional gradient feature with neural network | |
| Sharifi | Score-level-based face anti-spoofing system using handcrafted and deep learned characteristics | |
| Verma et al. | Static Signature Recognition System for User Authentication Based Two Level Cog, Hough Tranform and Neural Network | |
| Sivapriyan et al. | Analysis of facial recognition techniques | |
| Blessy et al. | Deep Learning Approach to Offline Signature Forgery Prevention | |
| Pervouchine et al. | Extraction and Analysis of Document Examiner Features from Vector Skeletons of Grapheme ‘th’ | |
| Joshi et al. | Offline handwritten signature verification using low level stroke features | |
| Breci et al. | A Novel Dataset for Non-Destructive Inspection of Handwritten Documents |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |