CN104266604B - A kind of method for handwriting verification - Google Patents
A kind of method for handwriting verification Download PDFInfo
- Publication number
- CN104266604B CN104266604B CN201410528560.4A CN201410528560A CN104266604B CN 104266604 B CN104266604 B CN 104266604B CN 201410528560 A CN201410528560 A CN 201410528560A CN 104266604 B CN104266604 B CN 104266604B
- Authority
- CN
- China
- Prior art keywords
- handwriting
- person
- word
- ink depth
- depth
- 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
Landscapes
- Document Processing Apparatus (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention belongs to the technical field of the inspection of document.In order to the qualification process that the method for solving existing handwriting verification is present is complicated, the low problem of efficiency, the present invention propose a kind of method for handwriting verification, including being cut person's handwriting to be identified and former person's handwriting respectively along the thickness direction of paper；The person's handwriting after cutting is put under metallographic microscope, obtain average ink depth and the ink depth of the word；By the average ink depth ratio of identical word in the ink depth of the word in person's handwriting to be identified and former person's handwriting relatively.The method of the present invention simplifies the process of handwriting verification, improves the efficiency of handwriting verification.
Description
Technical field
The invention belongs to the technical field of the inspection of document, and in particular to a kind of method for handwriting verification.
Background technology
Handwriting verification is applied widely in fields such as judicial, finance, archaeologies as a kind of method of the inspection of document.Shen
Number please disclose a kind of method for differentiating person's handwriting using mass spectrum imaging for 201310299489.2 patent, the method first-selected with
A certain proportion of methyl alcohol and water mixed solution are as ionization reagents, using surface desorption CIMS, high in conjunction with three-dimensional
Precision automatically controls mobile system, and identifying the handwriting carries out two-dimentional scanning of the mass spectrum；Then according to characteristic compounds ion signal in person's handwriting
The change of intensity, obtains the mass spectrum image of characteristic compounds in person's handwriting；Finally the mass spectrum image to obtaining is analyzed to differentiate
The true and false of writing.The method of above-mentioned discriminating person's handwriting used methyl alcohol, water and three-dimensional high-accuracy automatically control mobile system, use
The material for arriving and instrument more, and the analysis method of mass spectrum image is very complicated so that said method process is complicated, and efficiency is low.
A kind of handwriting identification method of the disclosure of the invention of Application No. 201110123764.6, the method are used respectively
Light with paper plane included angle less than 10 ° and treat with the light that paper plane included angle is the unspecified angle between 30 ° 60 °
The stroke of measurement is irradiated；Stroke is photographed using optical observation instrument and digital photographing apparatus above paper, point
In the case of not measuring in the case of first the width of stroke and second, stroke concave edges are under light irradiation on stroke
The width of the shadow section of formation；The ratio of shadow section width and stroke width is obtained, obtains measuring index；Repeat the above steps are obtained
To the corresponding measurement index of stroke of writing different types of structure, the stroke with writing different types of structure as transverse axis, with measurement
Index is the longitudinal axis, and referring piece writing measurement index map is obtained；Part writing to be identified is obtained using same method and index map is measured, right
Than both, the true and false can be identified.Above-mentioned writing mirror method for distinguishing uses optical observation instrument and digital photographing apparatus, authenticated
Journey is complicated, inefficient.
The patent of invention of Application No. 03109813.4 discloses a kind of writing discrimination method, and the method is first to pen to be detected
Mark is pre-processed, and extracts four directional element features；Then the spy of dimensionality reduction is obtained using PCA (principal component analysis) conversion dimensionality reduction
Levy, converted with LDA (linear discriminant analysis) and extract handwriting characteristic；Finally taxonomic history is carried out using euclid-distance classifier.On
Stating writing mirror method for distinguishing is included using PCA (principal component analysis), LDA (linear discriminant analysis) and euclid-distance classifier etc.,
Discrimination process is complicated, and efficiency is low.
In sum, the method for existing writer verification has discrimination process complexity, the low problem of efficiency.
Content of the invention
In order to the qualification process that the method for solving existing handwriting verification is present is complicated, the low problem of efficiency, the present invention are carried
Go out a kind of method for handwriting verification, to realize handwriting verification process operation simply, the purpose of efficiency high.
The method for handwriting verification of the present invention, including：
(1) person's handwriting to be identified and former person's handwriting are cut along the thickness direction of paper respectively, the pen to be identified
Mark is identical with the word of the original person's handwriting；
(2) by the former person's handwriting after cutting, it is put under metallographic microscope, the section of paper is observed, every in former person's handwriting to measure
The ink depth of individual word；
Person's handwriting to be identified after cutting is put under metallographic microscope, the section of paper is observed, to be identified to measure
The ink depth of each word in person's handwriting；
(3) by the ink depth ratio of the ink depth of word in person's handwriting to be identified and identical word in former person's handwriting relatively, if
Difference account for described original person's handwriting in identical word ink depth more than 50%, then it is assumed that the word in the person's handwriting to be identified with
In former person's handwriting, identical word is not that same person is write.
Wherein, in (2nd) step, the ink depth of each word in former person's handwriting is measured, then by identical in former person's handwriting
The ink depth of word is added, then the number of words divided by the identical word, to obtain the average ink depth of the identical word；Institute
State in (3rd) step, by the average ink depth ratio of identical word in the ink depth of word in person's handwriting to be identified and former person's handwriting relatively.
Wherein, in (2nd) step, when measuring the ink depth of word in person's handwriting to be identified and former person's handwriting, measure to be measured
Word section in maximum ink depth, the maximum ink depth is the ink depth of the word to be measured.
Wherein, in (2nd) step, when measuring the ink depth of word in person's handwriting to be identified and former person's handwriting, measure to be measured
Word section base on several points ink depth, the several points are first to choose ink depth from the section base of word
The maximum point of degree, then chooses second largest point of ink depth in remaining point on the section base of word again, so selects successively
Several points are taken, the mean value of the ink depth of the several points is then tried to achieve, the mean value is the ink depth of word to be measured
Degree.
Wherein, in (1st) step, the center line with paper per a line will be cut from centre per line inscribed as line of cut
Cut.
Wherein, in (1st) step, person's handwriting to be identified is identical with the cutting position of the paper of former person's handwriting.
Wherein, the paper of person's handwriting to be identified and former person's handwriting, in (1st) step, is clamped respectively using plasticine, then
Paper is cut.
Wherein, in (1st) step, person's handwriting to be identified and former person's handwriting are cut using paper knife.
Method of the present invention qualification process step is few, first former person's handwriting and person's handwriting to be identified is cut, then will cutting
Person's handwriting afterwards is put into metallography microscope Microscopic observation respectively, to obtain the ink depth of person's handwriting on paper, by the ratio of ink depth
The true and false that can relatively verify handwriting, qualification process are simple；The instrument that the method for the present invention is used is mainly metallographic microscope, is not required to
Digital photographing apparatus and the three-dimensional high-accuracy instrument for automatically controlling the complexity such as mobile system is wanted, the instrument for using is few, the work for using
Tool is simple, consequently facilitating operation, is favorably improved operating efficiency；The method of the present invention is by comparing the ink depth of identical word
Degree, the true and false verified handwriting with the size of the difference of ink depth, the process of calculating include the difference for calculating ink depth and
Difference accounts for the percentage of former person's handwriting ink depth, simply the computing of subtraction and division, it is not necessary to which mass spectrum image analysing computer, principal component are divided
The complicated analysis method of analysis (PCA) and linear discriminant analysis (LDA) etc., method of the present invention calculating process are simple, computational efficiency
Higher.In sum, the method for the present invention simplifies the process of handwriting verification, improves the efficiency of handwriting verification.And use
The method of the present invention carries out handwriting verification together with the method for existing handwriting verification, it is possible to increase the accuracy rate of handwriting verification.
Description of the drawings
Fig. 1 is photo of the paper under metallographic microscope after cutting.
Specific embodiment
Embodiment 1：
Former person's handwriting is the signature using signature pen：Wang Yueyue；Person's handwriting to be identified is the signature using signature pen：Wang Yueyue.
Need to identify whether former person's handwriting is write by same person with person's handwriting to be identified.The method for handwriting verification of the present invention, fits
For the pen type identical situation that former person's handwriting and person's handwriting to be identified are used, such as former person's handwriting and person's handwriting to be identified all use label
The situation that word pen is write or all write with ball pen.
Step 1：
The paper of to be identified person's handwriting and former person's handwriting is clamped using plasticine respectively, and plasticine plays a part of to support paper,
And paper formation is made, so that paper can be vertically placed on the section of metallography microscope Microscopic observation paper in step 2.Then make
With paper knife, former person's handwriting and person's handwriting to be identified are cut along the thickness direction of paper respectively, former person's handwriting and treat during cutting
Center line of the person's handwriting of identification all with " Wang Yueyue " this line as line of cut, the cutting position phase of former person's handwriting and person's handwriting to be identified
Same, during the ink depth of so more former person's handwriting and person's handwriting to be identified, it is the comparison of the ink depth of same position, reduces
The impact of other factors compared result, improves the accuracy of comparative result, and then improves the accuracy rate of handwriting verification.
Step 2：
The former person's handwriting after cutting is put under metallographic microscope, observe the section of paper, to measure " king " in former person's handwriting
The ink depth of " moon " " moon " these three words, then by the ink depth of first " moon " word and the ink depth of second " moon " word
Degree is added, then divided by 2, to obtain the average ink depth of " moon " word；Because only that " king " word, the ink depth of " king " word
It is equal to the average ink depth of " king " word.Person's handwriting to be identified after cutting is put under metallographic microscope, observes paper
Section, to measure the ink depth of " king " " moon " " moon " these three words in person's handwriting to be identified.
When writing, the ink of person's handwriting can be penetrated in paper, and the ink that the depth of infiltration is flowed out with pen when writing is how many, write
The speed of word and the strength size that writes have relation, and the ink of outflow is more, and the depth of infiltration is deeper；Write slower, infiltration
Depth deeper；The strength that writes is bigger, and the depth of infiltration is deeper.Fig. 1 is photo of the paper under metallographic microscope, Fig. 1
The stripes of middle white be divided into paper, clip paper is plasticine, and on paper, the part of black is the section of word, permissible
The section for seeing word is irregular rectangle, and the base in section is not smooth straight line, but having recessed has a convex song
Line, similar to wave.During the ink depth of measurement word, maximum ink depth in the section of the word, the ink of the maximum is measured
Water depth is the ink depth of the word.
In metallography microscope Microscopic observation original person's handwriting, the section of " king " " moon " " moon " these three words, measures the section of " king " word
Middle maximum ink depth；Measure maximum ink depth in the section of first " moon " word；Measure cutting for second " moon " word
Maximum ink depth in face, the data for specifically measuring are shown in Table 1.
The ink depth of the former person's handwriting of table 1
| Person's handwriting | Maximum ink depth |
| King | 0.043mm |
| Month | 0.052mm |
| Month | 0.049mm |
Drawn in former person's handwriting by table 1, in the section of " king " word, maximum ink depth is 0.043mm, the ink of " king " word
Depth is 0.043mm.Because only that " king " word, so the average ink depth of " king " is equal to the ink depth of " king " word
Degree.In the section of first " moon " word, maximum ink depth is 0.052mm, and the ink depth of first " moon " word is
0.052mm；In the section of second " moon " word, maximum ink depth is 0.049mm, and the ink depth of second " moon " word is
0.049mm；The ink depth of ink depth and second " moon " word of the average ink depth of " moon " word for first " moon " word
Mean value, the average ink depth of " moon " word is 0.051mm.
In metallography microscope Microscopic observation person's handwriting to be identified, the section of " king " " moon " " moon " these three words, measures " king " word
Section in maximum ink depth；Measure maximum ink depth in the section of first " moon " word；Measure second " moon "
Maximum ink depth in the section of word, the data for specifically measuring are shown in Table 2.
The ink depth of the person's handwriting to be identified of table 2
| Person's handwriting | Maximum ink depth |
| King | 0.068mm |
| Month | 0.085mm |
| Month | 0.084mm |
Drawn in person's handwriting to be identified by table 2, in the section of " king " word, maximum ink depth is 0.068mm, " king "
The ink depth of word is 0.068mm.In the section of first " moon " word, maximum ink depth is 0.085mm, first " moon "
The ink depth of word is 0.085mm；In the section of second " moon " word, maximum ink depth is 0.084mm, second " moon "
The ink depth of word is 0.084mm.
Step 3：
Word in person's handwriting to be identified is compared with identical word in former person's handwriting.By " king " word in person's handwriting to be identified
Ink depth be compared with the ink depth of " king " word in former person's handwriting, by first " moon " word in person's handwriting to be identified and the
The ink depth of two " moon " words is compared with the average ink depth of " moon " word in former person's handwriting respectively, if pen to be identified
The difference of the average ink depth of the ink depth of mark and former person's handwriting accounts for more than the 50% of the average ink depth of former person's handwriting, then recognize
It is not that same person is write by person's handwriting to be identified with former person's handwriting；If difference account for the 50% of the average ink depth of former person's handwriting with
Under, as difference is less, it is proposed that carry out handwriting verification using the method for other handwriting verifications existing.
By the average ink depth ratio of word in the ink depth of word in person's handwriting to be identified and former person's handwriting relatively, mistake can be reduced
Difference, increases the accuracy of the method for the present invention.Because activity at heart when the ink depth of certain word is also write with writer etc.
Other factors have relation, and the factor such as accidental change movable at heart of writer just can cause the ink depth ratio of the person's handwriting that writes
The ink depth of the person's handwriting that often writes is bigger than normal or less than normal, is compared using average ink depth, is just being intended to reduce writer's
The impact of the ink depth that the factor such as movable accidental change is identified the handwriting at heart, to improve the accuracy of comparative result.
According to the result that step 2 is obtained, to the ink depth of person's handwriting to be identified with former person's handwriting identical word average
Relatively, comparative result is shown in Table 3 to ink depth ratio.
Table 3
Draw from the comparative result of table 3, the ink depth of " king " word and " king " word in former person's handwriting in person's handwriting to be identified
The difference of ink depth accounts for 58% of the ink depth of " king " word in former person's handwriting, " king " word more than 50%, in person's handwriting to be identified
It is not that same people is write with " king " word in former person's handwriting；The ink depth of first " moon " word in person's handwriting to be identified and former pen
In mark, the difference of the average ink depth of " moon " word accounts for 67% of the average ink depth of " moon " word in former person's handwriting, more than 50%,
First " moon " word in person's handwriting to be identified is not that same people is write with " moon " word in former person's handwriting；In person's handwriting to be identified
The ink depth of two " moons " and the difference of the average ink depth of " moon " word in former person's handwriting account for the average of " moon " word in former person's handwriting
The 65% of ink depth, more than 50%, second " moon " word in person's handwriting to be identified is not same with " moon " word in former person's handwriting
One people is write.
Embodiment 2：
Former person's handwriting is the signature using signature pen：Wang Yueyue；Person's handwriting to be identified is the signature using signature pen：Wang Yue
Month.Need to identify whether former person's handwriting is write by same person with person's handwriting to be identified.
Step 1：
The paper of to be identified person's handwriting and former person's handwriting is clamped using plasticine respectively, and plasticine plays a part of to support paper,
And paper formation is made, so that paper can be vertically placed on the section of metallography microscope Microscopic observation paper in step 2.Then make
With paper knife, former person's handwriting and person's handwriting to be identified are cut along the thickness direction of paper respectively, former person's handwriting and treat during cutting
Center line of the person's handwriting of identification all with " Wang Yueyue " this line is as line of cut.
Step 2：
The former person's handwriting after cutting is put under metallographic microscope, observe the section of paper, to measure " king " in former person's handwriting
The ink depth of " moon " " moon " these three words, then by the ink depth of first " moon " word and the ink depth of second " moon " word
Degree is added, then divided by 2, to obtain the average ink depth of " moon " word；Because only that " king " word, the ink depth of " king " word
It is equal to the average ink depth of " king " word.Person's handwriting to be identified after cutting is put under metallographic microscope, observes paper
Section, to measure the ink depth of " king " " moon " " moon " these three words in person's handwriting to be identified.
Under metallographic microscope, it can be seen that the section of each word, the section is irregular rectangle, and the base in section is not
Smooth straight line, but have recessed have a convex curve, similar to wave.Measure certain word ink depth when, lead to
Range estimation is crossed, the maximum point of the word section base inking water depth is chosen, is then being selected in remaining point again on the word section base
Second largest point of ink depth is taken, the ink depth of the two points is measured, then try to achieve the average of the two ink depth that puts
Value, the mean value is the ink depth of the word.Can certainly be according to the method for above-mentioned selected point, on the section base of word
Three or more points are chosen, then tries to achieve the mean value of the ink depth of selected point.
The section of " king " " moon " " moon " these three words in metallography microscope Microscopic observation original person's handwriting, in the section of " king " word
On base, the maximum A point of ink depth is chosen, in remaining point, is then chosen second largest B point of ink depth again, measures this
The ink depth of two points；On the base in the section of first " moon " word, the maximum A point of ink depth is chosen, then surplus
Under point in choose second largest B point of ink depth again, measure the ink depth of two points；Section in second " moon " word
Base on, choose the maximum A point of ink depth, in remaining point, then choose second largest B point of ink depth again, measure
The ink depth of two points.The data for specifically measuring are shown in Table 4.
The ink depth of the former person's handwriting of table 4
| Person's handwriting | A point ink depth | B point ink depth | Ink depth |
| King | 0.044mm | 0.040mm | 0.042mm |
| Month | 0.053mm | 0.048mm | 0.051mm |
| Month | 0.050mm | 0.046mm | 0.048mm |
Drawn in former person's handwriting by table 4, in the section of " king " word the ink depth of two points be respectively 0.044mm and
0.040mm, the mean value of the ink depth of two points is 0.042mm, and the ink depth of " king " word is 0.042mm.Because only
There is " king " word, so the average ink depth of " king " is equal to the ink depth of " king " word.The section of first " moon " word
In the ink depth of two points be respectively 0.053mm and 0.048mm, the mean value of the ink depth of two points is 0.051mm,
The ink depth of first " moon " word is 0.051mm；In the section of second " moon " word, the ink depth of two points is respectively
0.050mm and 0.046mm, the mean value of the ink depth of two points is 0.048mm, and the ink depth of second " moon " word is
0.048mm；The ink depth of ink depth and second " moon " word of the average ink depth of " moon " word for first " moon " word
Mean value, the average ink depth of " moon " word is 0.050mm.
The section of " king " " moon " " moon " these three words in metallography microscope Microscopic observation person's handwriting to be identified, in " king " word
On the base in section, the maximum A point of ink depth is chosen, in remaining point, then chooses second largest B point of ink depth again,
Measure the ink depth of two points；On the base in the section of first " moon " word, the maximum A point of ink depth is chosen, so
Choose second largest B point of ink depth afterwards in remaining point again, measure the ink depth of two points；In second " moon " word
Section base on, choose the maximum A point of ink depth, in remaining point, then choose second largest B of ink depth again
Point, measures the ink depth of two points.The data for specifically measuring are shown in Table 5.
The ink depth of the person's handwriting to be identified of table 5
| Person's handwriting | A point ink depth | B point ink depth | Ink depth |
| King | 0.067mm | 0.063mm | 0.065mm |
| Month | 0.079mm | 0.075mm | 0.077mm |
| Month | 0.083mm | 0.078mm | 0.081mm |
Drawn in person's handwriting to be identified by table 5, in the section of " king " word the ink depth of two points be respectively 0.067mm and
0.063mm, the mean value of the ink depth of two points is 0.065mm, and the ink depth of " king " word is 0.065mm.First
In the section of " moon " word the ink depth of two points be respectively 0.079mm and 0.075mm, the ink depth of two points average
It is worth for 0.077mm, the ink depth of first " moon " word is 0.077mm；The ink of two points in the section of second " moon " word
Depth is respectively 0.083mm and 0.078mm, and the mean value of the ink depth of two points is 0.081mm, second " moon " word
Ink depth is 0.081mm.
Step 3：
Word in person's handwriting to be identified is compared with identical word in former person's handwriting.By " king " word in person's handwriting to be identified
Ink depth be compared with the ink depth of " king " word in former person's handwriting, by first " moon " word in person's handwriting to be identified and the
The ink depth of two " moon " words is compared with the average ink depth of " moon " word in former person's handwriting respectively, if pen to be identified
The difference of the average ink depth of the ink depth of mark and former person's handwriting accounts for more than the 50% of the average ink depth of former person's handwriting, then recognize
It is not that same person is write by person's handwriting to be identified with former person's handwriting；If difference account for the 50% of the average ink depth of former person's handwriting with
Under, as difference is less, it is proposed that carry out handwriting verification using the method for other handwriting verifications existing.
According to the result that step 2 is obtained, to the ink depth of person's handwriting to be identified with former person's handwriting identical word average
Relatively, comparative result is shown in Table 6 to ink depth ratio.
Table 6
Draw from the comparative result of table 6, " king " word in the ink depth of " king " word in person's handwriting to be identified and former person's handwriting
The difference of ink depth accounts for 55% of the ink depth of " king " word in former person's handwriting, more than 50%, " king " in person's handwriting to be identified with
Not " king is not that same person is write " in former person's handwriting；In the ink depth of first " moon " in person's handwriting to be identified and former person's handwriting
The difference of the average ink depth of " moon " word accounts for 54% of the average ink depth of " moon " word in former person's handwriting, more than 50%, waits to reflect
It is not that same person is write to determine in person's handwriting first " moon " and former person's handwriting "；Second " moon " in person's handwriting to be identified
Ink depth accounts for the average ink depth of " moon " word in former person's handwriting with the difference of the average ink depth of " moon " word in former person's handwriting
62%, more than 50%, in second " moon " and former person's handwriting in person's handwriting to be identified is not that same person is write ".
Experiment is presented herein below, logical with the difference of the average ink depth of former person's handwriting in order to prove the ink depth of the imitative person's handwriting that writes
The 50% of the average ink depth of often greater than former person's handwriting.
Experimental procedure：
(1) that participates in experiment has 30 groups of experimenters and imitative writer, and experimenter is write with signature pen on 70g standard A4 paper
One " reality " word, the person's handwriting that experimenter writes is equivalent to former person's handwriting；
(2) writer being imitated on same A4 paper, writing " reality " word with signature pen, the person's handwriting that imitative writer writes is equivalent to treating
The person's handwriting of identification；
(3) above-mentioned A4 folder is lived using plasticine, so that paper is supported, make A4 paper that gold can be vertically placed in next step
Under phase microscope, to observe the section of paper；Then using paper knife, along the thickness direction of paper, along the direction of paper row
Paper is cut, " reality " word cutting that " reality " that experimenter is write and imitative writer are write；
(4) paper after cutting is put under metallographic microscope, the section of paper is observed, measures " reality " that experimenter is write
The ink thickness of " reality " word that the ink thickness of word and imitative writer are write, measured the results are shown in Table 7.
Table 7
Can be drawn by the data of table 7, the difference of the average ink depth of the ink depth of the imitative person's handwriting that writes and former person's handwriting
Value is typically larger than the 50% of the average ink depth of former person's handwriting.The principle of the method for the present invention is：When a people mimicker's
During person's handwriting, when the signature of such as mimicker is signed on contract, it will usually which that write is slow, often retouch again after writing several
Time, so that person's handwriting is more like really signing, and the strength during writing is will also tend to than larger, after thus having one
Really, the ink depth for exactly imitating the ink depth ratio original person's handwriting for writing person's handwriting is big because imitative when writing slowly, retouched after writing again
Several times, the ink used in signature process is certainly many, and the so imitative ink depth for writing person's handwriting is certainly big.Table 7 be according to this
The experiment that the principle of the method for invention is done, can be drawn by the experimental data of table 7, the imitative ink depth ratio original person's handwriting for writing person's handwriting
Ink depth is really big, and is typically larger than the 50% of former person's handwriting ink depth.Therefore, the present invention for handwriting verification
Method, using the imitative difference for writing person's handwriting and the ink depth of former person's handwriting more than original person's handwriting ink depth 50% as judge person's handwriting as
False condition.The factor such as type certainly in practice, by pen, paper type and psychological condition when writing is affected, and is existed and is treated
When the difference of the ink depth of the person's handwriting of identification and former person's handwriting is more than the 50% of former person's handwriting ink depth, person's handwriting to be identified is still
For genuine situation, but such case is seldom, belongs to case, does not affect the use of the method for the present invention.Using the method for the present invention
Handwriting verification is carried out, accuracy rate is more than 90%.
The method of the present invention is not suitable for identifying using the pen (such as pencil) for not occurring in paper to permeate and oozing completely
The person's handwriting that saturating pen (such as writing brush) is write.
Claims (7)
1. a kind of method for handwriting verification, it is characterised in that include：
(1) person's handwriting to be identified and former person's handwriting are cut along the thickness direction of paper respectively, the person's handwriting to be identified and
The word of the original person's handwriting is identical；
(2) by the former person's handwriting after cutting, it is put under metallographic microscope, the section of paper is observed, to measure each word in former person's handwriting
Ink depth, then the ink depth of identical word in former person's handwriting is added, then the number of words divided by the identical word, with
Average ink depth to the identical word；
Person's handwriting to be identified after cutting is put under metallographic microscope, the section of paper is observed, to measure person's handwriting to be identified
In each word ink depth；
(3) by the average ink depth ratio of the ink depth of word in person's handwriting to be identified and identical word in former person's handwriting relatively, if
Difference accounts for more than 50% of the average ink depth of identical word in the original person's handwriting, then it is assumed that in the person's handwriting to be identified
Word is not that same person is write with identical word in former person's handwriting.
2. the method for handwriting verification according to claim 1, it is characterised in that in (2nd) step, measure and wait to reflect
In fixed person's handwriting and former person's handwriting during the ink depth of word, maximum ink depth in the section of word to be measured is measured, the maximum
Ink depth is the ink depth of the word to be measured.
3. the method for handwriting verification according to claim 1, it is characterised in that in (2nd) step, measure and wait to reflect
In fixed person's handwriting and former person's handwriting during the ink depth of word, the ink depth of several points on the section base of word to be measured is measured,
The several points are first to choose the maximum point of ink depth from the section base of word, are then left on the section base of word
Point in choose second largest point of ink depth again, so successively choose several points, then try to achieve the ink of the several points
The mean value of water depth, the mean value are the ink depth of word to be measured.
4. the method for handwriting verification according to claim 1, it is characterised in that in (1st) step, with paper
Center line per a line is line of cut, will be per line inscribed from middle cutting.
5. the method for handwriting verification according to claim 1, it is characterised in that in (1st) step, to be identified
Person's handwriting is identical with the cutting position of the paper of former person's handwriting.
6. the method for handwriting verification according to claim 1, it is characterised in that in (1st) step, using rubber
Mud clamps the paper of person's handwriting to be identified and former person's handwriting respectively, then paper is cut.
7. the method for handwriting verification according to claim 1, it is characterised in that in (1st) step, using cutting a sheet of paper
Knife is cut to person's handwriting to be identified and former person's handwriting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410528560.4A CN104266604B (en) | 2014-10-09 | 2014-10-09 | A kind of method for handwriting verification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410528560.4A CN104266604B (en) | 2014-10-09 | 2014-10-09 | A kind of method for handwriting verification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104266604A CN104266604A (en) | 2015-01-07 |
| CN104266604B true CN104266604B (en) | 2017-03-08 |
Family
ID=52158146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410528560.4A Expired - Fee Related CN104266604B (en) | 2014-10-09 | 2014-10-09 | A kind of method for handwriting verification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104266604B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122766B (en) * | 2017-05-23 | 2019-09-06 | 北京瑞源智通科技有限公司 | A kind of method of handwriting verification |
| CN107239753B (en) * | 2017-05-23 | 2019-03-12 | 北京瑞源智通科技有限公司 | A kind of system of handwriting verification |
| RU2674495C1 (en) * | 2018-09-24 | 2018-12-11 | Ахмеджан Атаханович Аллаберганов | Multifunctional biological and video microscopic spectral complex |
| CN110674771B (en) * | 2019-09-29 | 2022-04-01 | 北部湾大学 | Nixing pottery ware signature identification and verification method |
| CN112117005A (en) * | 2020-09-18 | 2020-12-22 | 中国人民武装警察部队海警学院 | Handwriting interpretation system and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB1426836A (en) * | 1972-07-26 | 1976-03-03 | Sick Erwin | Method of and apparatus for measuring a light reflective measurement plane from a reference plane |
| CN1279161A (en) * | 2000-06-30 | 2001-01-10 | 郭春生 | Antiforge seal with portrait and its making method |
| US6281974B1 (en) * | 1998-03-18 | 2001-08-28 | Nova Measuring Instruments, Ltd. | Method and apparatus for measurements of patterned structures |
| CN101435826A (en) * | 2007-11-16 | 2009-05-20 | 冯德超 | Detection method in cultural relics identification |
| CN102252621A (en) * | 2011-05-11 | 2011-11-23 | 谭国清 | Handwriting identification method |
-
2014
- 2014-10-09 CN CN201410528560.4A patent/CN104266604B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB1426836A (en) * | 1972-07-26 | 1976-03-03 | Sick Erwin | Method of and apparatus for measuring a light reflective measurement plane from a reference plane |
| US6281974B1 (en) * | 1998-03-18 | 2001-08-28 | Nova Measuring Instruments, Ltd. | Method and apparatus for measurements of patterned structures |
| CN1279161A (en) * | 2000-06-30 | 2001-01-10 | 郭春生 | Antiforge seal with portrait and its making method |
| CN101435826A (en) * | 2007-11-16 | 2009-05-20 | 冯德超 | Detection method in cultural relics identification |
| CN102252621A (en) * | 2011-05-11 | 2011-11-23 | 谭国清 | Handwriting identification method |
Non-Patent Citations (3)
| Title |
|---|
| 基于笔压变换系数的手写汉字笔迹鉴定;李玉凤等;《微计算机信息》;20090531;第25卷(第5-2期);第248-250页 * |
| 用激光共聚焦显微镜鉴别可疑笔迹与印章;熊华淇;《激光生物学报》;20011231;第10卷(第4期);第277-280页 * |
| 签名笔迹笔压特征显现方法的比较研究;刘宁等;《中国司法鉴定》;20140131(第1期);第29页左栏第1-2段及图8 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104266604A (en) | 2015-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104266604B (en) | A kind of method for handwriting verification | |
| Zhu et al. | Biometric personal identification based on handwriting | |
| CN105046730B (en) | Written handwriting rendering method and device applied to writing brush | |
| US20150169951A1 (en) | Comparing documents using a trusted source | |
| Liwicki et al. | ICFHR 2012 competition on automatic forensic signature verification (4NsigComp 2012) | |
| Deviterne-Lapeyre | Interpol review of questioned documents 2016–2019 | |
| JP7251670B2 (en) | Feature quantity generation device, system, feature quantity generation method and program | |
| CN111213156A (en) | Character recognition sharpness determination | |
| Maken et al. | A study on various techniques involved in gender prediction system: a comprehensive review | |
| CN105302507A (en) | Certificate printing control method and certificate template setting method and system | |
| DE29521538U1 (en) | Device for recognizing the boundary between characters in handwritten text | |
| CN106934383B (en) | The recognition methods of picture markup information, device and server in file | |
| CN101587540A (en) | Printer verification method for detecting document source by means of geometric distortion of page document | |
| US12067796B2 (en) | Method for detecting fraud in documents | |
| US11076069B2 (en) | Pixelation density increment for embedding information | |
| JP7197009B2 (en) | Object management system | |
| CN107122766A (en) | A kind of method of handwriting verification | |
| Gonçalves et al. | Time/space based biometric handwritten signature verification | |
| DE102018107409A1 (en) | Method for recognizing handwritten characters, preferably during writing, and writing instrument for detecting handwritten characters | |
| Paturkar et al. | Handwriting Comparison and Character Analysis | |
| Carson et al. | Polygon-based drawing accuracy analysis and positive/negative space | |
| Hajamohideen et al. | Kalanjiyam: Unconstrained offline tamil handwritten database | |
| Atencio et al. | Using OMR for Grading MCQ-Type Answer Sheets Based on Bubble Marks | |
| Franke | Stroke-morphology analysis using super-imposed writing movements | |
| Gheni et al. | Uyghur Language Off-line Signature Verification Based On The Modified Directional Features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170308 Termination date: 20171009 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |