CN103593673A - On-line handwritten signature authentication method based on dynamic threshold - Google Patents
On-line handwritten signature authentication method based on dynamic threshold Download PDFInfo
- Publication number
- CN103593673A CN103593673A CN201310521026.6A CN201310521026A CN103593673A CN 103593673 A CN103593673 A CN 103593673A CN 201310521026 A CN201310521026 A CN 201310521026A CN 103593673 A CN103593673 A CN 103593673A
- Authority
- CN
- China
- Prior art keywords
- signature
- dtw
- coordinate
- sample
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The invention provides an on-line handwritten signature authentication method based on a dynamic threshold, and belongs to the field of information safety. According to the on-line handwritten signature authentication method, comprehensive modeling is carried out on coordinates, pressure and speed information of on-line signatures, effects of all factors are comprehensively considered, and the accuracy rate of signature judgment is more effectively increased. According to the method, subsection judgment strategies are adopted for the signatures, a more stable subsection method in the two subsection strategies is automatically selected according to the writing habit of the person who signs, and therefore accuracy of similarity judgment for the signatures in a signature judgment process can be improved. According to the method, the threshold utilized in the judgment process of signatures to be detected is determined according to different signing dynamic conditions, reflection of the difference characteristic of the signature of a person is facilitated, the problem of different kinds of characteristic stability of the signature of each person is considered, and therefore the optimal comparison result of each person is obtained.
Description
Technical field
The invention belongs to information security field, be specifically related to a kind of online handwriting signature authentication method based on dynamic threshold.
Background technology
In society, along with the develop rapidly of computer technology and Internet technology, information revolution has produced enormous impact to the traditional office of people and life style, and information security has become one of focal issue that people face.The application such as ecommerce, online transaction, online payment, network file transmission seem more outstanding to the requirement of information security, and authentication quickly and accurately more and more becomes a kind of exigence.The method of authentication can be divided into common several classes: the method based on knowledge, such as system password, password etc.; Method based on keepsake, as I.D., IC-card etc.; Recognition methods based on biological characteristic, as fingerprint, iris, retina, handwritten signature etc.Wherein, in the recognition methods based on biological characteristic, the behavioural characteristic that people's long-term life forms, as signature identification people's identity, is considered to natural, the most credible a kind of identity identifying method.And in daily life, authentication mode by signature is seen everywhere, signature of file, check, contract, agreement etc., signature authentication has the meaning of particular importance, in long-term social life, people have accepted signature as sign identity and a kind of mode that shows own wish.
In today of Internet and infotech develop rapidly, the mode of traditionally on paper handwritten signature cannot meet the demand of information security.By electronic device terminal, gather the signed data of signer, the handwritten signature verification technology of then being identified and being authenticated by data analysis technique more and more receives publicity.The processes such as handwritten signature verification comprises that signature obtains, pre-service, feature extraction and comparison, decision-making.The method of obtaining according to signature is different, and signature authentication can be divided into static signature authentication and the large class of Signature Verification two.Static signature authentication method, after being generally the signature image pre-service to obtaining, extracts the parameter attribute of some geometric configuration aspects, the comparison of then signing; Signature Verification method, by electronic device terminal, the signing messages in dynamic acquisition writing process, such as information such as the image of signing, the order of strokes observed in calligraphy, speed, pressure, then sets up signature authentication model, the authentication of signing.In dynamic signature information due to the collection of Signature Verification method, comprise abundant personal characteristics, be difficult to imitate, can effectively identify the identity of signer, so more and more become the main direction of studying of signature authentication.The research of online handwriting signature authentication be a kind of Signature Verification method, and it is the research field of a multidisciplinary intersection, relates to the knowledge of the multi-door subjects such as computing machine, signal processing, pattern-recognition, neuro-physiology.
The process nature of online handwriting signature authentication is the process to the extraction of signed data feature, comparison and differentiation, by extracting some characteristic information of signed data to be measured, and the signature in these characteristic informations and registered template base is compared, if comparative result is in rational error range, think that this signature is real signature in person, otherwise think the signature of forgery.Yet existing online handwriting signature authentication technology is still far from perfect, reliability and precision can't be satisfactory, reason is: 1) handwritten signature comprises abundant behavioral characteristics, comprise the abundant informations such as font, the order of strokes observed in calligraphy, speed, pressure, existing a lot of Signature Authentication System only adopts single or less several feature independences and authenticates, such as only relying on font style characteristic, imitator, likely by a large amount of imitation exercises, breaks through the feature in a certain respect of signature like this.2), because handwritten signature is biometric information, so the feature occurring in characteristic quantity depends on concrete everyone, different people has its distinctive characteristic quantity.In existing online handwriting signature system, the true or false of judgement handwritten signature, relatively to determine according to the similarity of respectively writing between the characteristic quantity of signature and the true signature character amount of Sample Storehouse, and in the process of similarity comparison, all to use fixed threshold, thereby the deviation feature of having ignored everyone signature, has ignored the different problem of feature stability of everyone signature, thereby cannot obtain everyone best comparison result.
Summary of the invention
For the deficiencies in the prior art, the present invention proposes a kind of online handwriting signature authentication method based on dynamic threshold, to reach the accuracy that has improved the similarity judgement of signing in the accuracy rate that signature judges, raising signature deterministic process and the object that obtains everyone best comparison result.
An online handwriting signature authentication method for dynamic threshold, comprises the following steps:
Step 1, gather my a plurality of signature sample data, the movement locus of lettering pen is carried out to constant duration sampling, obtain X coordinate figure, Y coordinate figure and the contact force value to clipboard of sampled point; And all sampling numbers certificates of signature are carried out to validity check, and if signed data is qualified, allow the submission of signed data, perform step 2; Otherwise return to execution step 1;
Step 2, signature sample choose the foundation with template base;
Whether the quantity of the signature sample that step 2-1, judgement gather reaches the sample of setting is chosen requirement, if so, performs step 2-2; Otherwise, return to execution step 1;
Step 2-2, using rising of signature start to write a little and the extreme point of coordinate as waypoint, maybe using the extreme point of the angular velocity extreme point that works start to write point, signature of signature and signature linear velocity as waypoint, the staging treating of realization to signed data;
Step 2-3, employing dynamic time programming DTW method judge the similarity degree of two signatures, obtain the final sample of template base;
Step 2-3-1, calculate the overall DTW value of all signatures coordinate between any two, coordinate segmentation summation DTW value, coordinate segmentation key point DTW value, the overall DTW value of pressure, pressure segmentation sue for peace DTW value, pressure segmentation key point DTW value, the overall DTW value of speed, speed segmentation and DTW value and speed waypoint DTW value respectively, and calculate above-mentioned DTW value average and deviation;
Step 2-3-2, delete the signature sample of sampling number grid deviation maximum and DTW value deviation maximum in all samples;
Step 2-3-3, repeatedly perform step 1 to step 2-3-3 until the volume residual of sample reaches the sample size of the required reservation in final sample storehouse;
Step 2-3-4, the whether satisfied signature stability condition of judgement residue sample, i.e. whether judgement residue sample between any two DTW deviation is all less than the DTW threshold value that user sets, and if so, will remain the final sample of signature sample as template base; Otherwise return to execution step 1;
Step 2-4, using the remaining stable signature sample of choosing as template base sample, calculation template storehouse sample DTW value between any two, and it is added up, determine the stable DTW value deviation dynamic threshold of signature coordinate of residue, the dynamic discriminant strategy of the DTW value deviation dynamic threshold of the DTW value deviation dynamic threshold of pressure, speed and applicable current signature, and deposit above-mentioned information in template base;
Step 3, gather tested user's signature sample data, the movement locus of lettering pen is carried out to constant duration sampling, obtain X coordinate figure, Y coordinate figure and the contact force value to clipboard of sampled point; And tested signature sampling number certificate is carried out to validity check, and if signed data is qualified, allow the submission of signed data, perform step 4; Otherwise return to execution step 3;
Step 4, judge the true and false of tested signature;
Step 4-1, the employing of adding up tested signature count, sign stroke the coordinate extreme point type of rise and fall stroke count amount, signature, the angular velocity extreme point type of the coordinate extreme point quantity of signature, signature, the linear velocity extreme point type of the angular velocity extreme point quantity of signature, signature and the linear velocity extreme point quantity of signature; In conjunction with the dynamic threshold in template base, and based on above-mentioned information, signature to be measured is carried out to dynamic filter, judge that whether tested signature meets dynamic filter condition, if so, performs step 4-2; Otherwise tested signature is pseudo-signature, finishes;
Step 4-2: select the dynamic discriminant strategy that is applicable to current signature to carry out staging treating to tested signature;
Step 4-3: according to the coordinate of tested signed data, pressure and speed, calculate the DTW value between coordinate, pressure and the speed of all signature sample data in itself and template base;
Step 4-4: according to the overall DTW value deviation of coordinate between the signature sample of having deposited in signed data to be measured and template base, coordinate segmentation summation DTW value deviation, coordinate segmentation key point DTW value deviation, the overall DTW value deviation of pressure between signature sample in signed data to be measured and template base, pressure segmentation summation DTW value deviation, pressure segmentation key point DTW value deviation, the overall DTW value of the speed of signature sample deviation in signed data and template base, speed segmentation and DTW value deviation, speed waypoint DTW value deviation, respectively in conjunction with the dynamic threshold of signature sample coordinate DTW value deviation in template base, the dynamic threshold of the dynamic threshold of pressure DTW value deviation and speed DTW value deviation, the method that adopts multi-expert ballot to mark judges the true and false of signature to be measured, the DTW value quantity that is even less than dynamic threshold is greater than half of DTW value sum, signature to be measured is judged as true signature, otherwise signature to be measured is judged as pseudo-signature, forbid that tested user logs in,
Step 4-5: differentiation result is carried out to output display.
Effect described in step 1 and step 3 checks, comprising that coordinate that signature is write crosses the border to check and signature reasonalbeness check, and whether the length and width scope of always the counting of signature, total writing time, signature reaches the threshold value of setting.
The dynamic discriminant strategy that determine to be applicable to current signature described in step 2-4 and step 4-2, described dynamic discriminant strategy comprises: rising of signature started to write a little and the extreme point of coordinate carries out segmentation and start to write the angular velocity extreme point of point, signature and the extreme point of the linear velocity of signing of signature are carried out to segmentation as waypoint as waypoint; Select segmentation method: be chosen under this segmentation method, signature sample is the method for DTW value deviation minimum between any two.
The DTW value deviation dynamic threshold of the signature coordinate that the definite residue described in step 2-4 is stable, be specially: determine to remain and stablize sample coordinate DTW distance between any two, obtain whole coordinate DTW range averaging values, and according to each coordinate DTW distance and coordinate DTW range averaging value, obtain the deviation of DTW distance, select deviation maximal value as the DTW value deviation dynamic threshold of coordinate; The DTW value deviation dynamic threshold value determination method of the DTW value deviation dynamic threshold of pressure and speed determines that with the DTW value deviation dynamic threshold of coordinate method is identical.
The coordinate extreme point type of the signature described in step 4-1 comprises X extreme point, Y extreme point, X and Y extreme point simultaneously; The angular velocity extreme point type of signature comprises angular velocity minimal value and angular velocity maximum value.
Dynamic filter condition described in step 4-1 is: specimen sample count average ± 20% scope in; The stroke count amount of rising and falling average ± 2 stroke counts within the scope of; Signature coordinate extreme point quantity average ± 20% scope in; Signature angular velocity extreme point quantity average ± 20% scope in; Signature linear velocity extreme point quantity average ± 20% scope in.
Advantage of the present invention:
A kind of online handwriting signature authentication method based on dynamic threshold of the present invention, be with the common online handwriting signature authentication of tradition method difference: 1) the inventive method is carried out comprehensive modeling to the coordinate of on-line signature, pressure, velocity information, and then consider the effect of each factor, more effectively improved the accuracy rate of signature judgement; 2) strategy that the inventive method has adopted segmentation to differentiate to signature, and according to my writing style of signature, automatically select more stable segmentation method in two kinds of partition strategies, thus can improve the accuracy of the similarity judgement of signing in signature deterministic process; 3) threshold value that the inventive method is used in signature deterministic process to be measured is dynamically to determine according to different signatures, be conducive to embody the deviation feature of everyone signature, consider the different problem of feature stability of everyone signature, thereby obtained everyone best comparison result.
Accompanying drawing explanation
Fig. 1 is every multidate information schematic diagram that the handwriting pad of the online handwriting signature authentication method and system based on dynamic threshold of an embodiment of the present invention obtains;
Fig. 2 is the system hardware connection diagram of an embodiment of the present invention;
Fig. 3 is that the system of an embodiment of the present invention forms module and functions structural representation;
Fig. 4 is the online handwriting signature operation interface schematic diagram of an embodiment of the present invention;
Fig. 5 is the online handwriting signature authentication method flow diagram based on dynamic threshold of an embodiment of the present invention;
Fig. 6 is the signature sequence data matching process schematic diagram of an embodiment of the present invention;
Fig. 7 is the signature segmentation schematic diagram of an embodiment of the present invention.
Embodiment
Below in conjunction with accompanying drawing, an embodiment of the present invention is described further.
As shown in Figure 1, in the embodiment of the present invention, the signature writing pad of employing is general touch control type flat-panel display.This product can provide coordinate, the pressure information at number of signature strong point; The subsidiary writing pencil of this equipment, to determine the current location of cursor, provides the two-dimensional coordinate information of cursor position.When this handwriting input device is placed in face of user with normal direction, true origin is positioned at the upper left corner of equipment, and X-axis is pointed to right-hand, and Y-axis is pointed to below.
As shown in Figure 2, in the embodiment of the present invention, the handwriting pad using is connected with computing machine by USB interface, and this universal touch control panel display, is transferred to computer screen on handwriting pad, uses Universal USB interface, connects simple and convenient.It comprises two data lines, a conduct output, transmission image; Article one, as input, transmit coordinate, the pressure information of the current point of writing pencil.
As shown in Figure 3, the online handwriting signature authentication of the embodiment of the present invention is comprised of signed data collection, signature pre-service, feature extraction, signature template base and signature authentication.The handwritten signature verification of the embodiment of the present invention comprises two stages, i.e. registration phase and authentication phase.Step before feature extraction is the same to registration phase with authentication phase, inputs signature, pre-service, feature extraction.After feature extraction, registration phase, by together with the identity information of the feature of extracting and input, deposits template database in, during for authentication phase comparison.In authentication phase, after feature extraction, system finds template corresponding to input signature from template database, and the two is carried out to matching ratio, then according to result relatively, carries out decision-making.If signature to be measured meets the decision condition of authentication model, accepting this signature is true signature, otherwise thinks that this signature is for pseudo-signature.
Fig. 4 is system operation interface schematic diagram of the present invention, and comprising signs writes district, signature sample list, signature list to be measured, signature to be measured differentiation result output area, signature image preview region and correlation function control knob.
An online handwriting signature authentication method for dynamic threshold, method flow as shown in Figure 5, comprises the following steps:
Step 1, gather my a plurality of signature sample data, the movement locus of lettering pen is carried out to constant duration sampling, obtain X coordinate figure, Y coordinate figure and the contact force value to clipboard of sampled point; And all sampling numbers certificates of signature are carried out to validity check, and if signed data is qualified, allow the submission of signed data, perform step 2; Otherwise return to execution step 1; Described effect checks, comprising that coordinate that signature is write crosses the border to check and signature reasonalbeness check, and whether the length and width scope of always the counting of signature, total writing time, signature reaches the threshold value of setting.
In the embodiment of the present invention, the collection of my sample data of signing.I carry out writing of signed data by handwriting pad signature, system is carried out constant duration sampling to the movement locus of lettering pen, obtain X, Y coordinate figure and the contact force value to clipboard of sampled point, the data recording that finally generates signature every bit is: (sampled point numbering, X coordinate, Y coordinate, timestamp, pressure).Then to all sampling numbers of signature according to carrying out validity check, if by checking, again write effective signature, or signer carries out carrying out the collection of sample after the signature writing training of standard again.
In the embodiment of the present invention, system provides signature training function: signer is formally being write before signature, can use training function, is familiar with the writing feature of handwriting pad and signature pen.Select signature training function, in input field, write signature, write latter 2 seconds systems of end and automatically remove signature.The user profile of input signature: before formal collecting sample and signature to be measured, first the user identity relevant information of signature is set.
Step 2, signature sample choose the foundation with template base;
Whether the quantity of the signature sample that step 2-1, judgement gather reaches the sample of setting is chosen requirement (minimum quantity requirement), if so, performs step 2-2; Otherwise, return to execution step 1;
Step 2-2, using rising of signature start to write a little and the extreme point of coordinate as waypoint, maybe, using the extreme point of the angular velocity extreme point that works start to write point, signature of signature and signature linear velocity as waypoint, realization is carried out staging treating to signed data;
Fig. 7 is the signature segmentation schematic diagram of a kind of online handwriting signature authentication method based on dynamic threshold of the present invention.The extreme point that works the coordinate of starting to write and sign according to signature in figure carries out segmentation.
Step 2-3, adopt dynamic time programming DTW(Dynamic Time Warping) the similarity degree of two signatures of method judgement, obtain the final sample of template base;
Fig. 6 is the main thought schematic diagram of signature sequence data dynamic matching method DTW in the present invention, is provided with two signature sequence S and T, and both sample frequency are the same, wherein S=(s
1, s
2..., s
n), T=(t
1, t
2..., t
m), for a signing messages, signed data is regarded as to one from a point sequence of origin-to-destination, s
1, s
2..., s
nrefer to point sequence corresponding to signature S the 1st, the 2nd ... n point, all points form the curve of top in Fig. 7; Signature sequence T is similar; The target of signature sequence data coupling is exactly to find an optimum point-to-point corresponding relation, makes the accumulative total gap of two samples minimum.In order to find so point-to-point alignment relation, define the matrix of a n * m, entry of a matrix element (i, j) records two some si of two samples and the distance between tj.Matching process will be found the path W=(w of from (1,1) to (n, m) exactly
1, w
2..., w
k), max (n, m) <K<n+m, makes the element sum on this path minimum.Fig. 6 has shown the point-to-point graph of a relation that the coordinate sequence coupling of two signatures obtains.For ease of observing, in figure, the point of mutual correspondence is coupled together with straight line.As can be seen from the figure, because two sequences relatively present irregular variation, in the point-to-point relation obtaining, there is a large amount of one-to-many situations, wherein have and much belong to irrational matching relationship, should avoid as far as possible.The present invention, in conjunction with the method for signature segmentation, has done further and has optimized the coupling of signature sequence data.
Step 2-3-1, calculate the overall DTW value of all signatures coordinate between any two, coordinate segmentation summation DTW value, coordinate segmentation key point DTW value, the overall DTW value of pressure, pressure segmentation sue for peace DTW value, pressure segmentation key point DTW value, the overall DTW value of speed, speed segmentation and DTW value and speed waypoint DTW value respectively, and calculate above-mentioned DTW value average and deviation;
In the embodiment of the present invention, the calculating of DTW value between signature sample.For coordinate, pressure, the speed of signature, calculate all signature sample DTW value between any two respectively.The calculating of DTW value considers by segmentation and two kinds of situations of not segmentation, calculates in DTW value in not segmentation situation, segmentation situation the summation of all segmentation DTW values, the DTW value of segmentation key point.Calculating on the basis of above-mentioned DTW value, then calculating average, the deviation information of all kinds of DTW values of all signature sample, the support information while choosing as follow-up signature sample.
Step 2-3-2, delete the signature sample of sampling number grid deviation maximum and DTW value deviation maximum in all samples;
Step 2-3-3, repeatedly perform step 1 to step 2-3-3 until the volume residual of sample reaches the sample size of the required reservation in final sample storehouse;
Step 2-3-4, the whether satisfied signature stability condition of judgement residue sample, i.e. whether judgement residue sample between any two DTW deviation is all less than the DTW threshold value that user sets, and if so, will remain the final sample of signature sample as template base; Otherwise return to execution step 1;
In the embodiment of the present invention, the whether stable method of judgement sample is for judging according to the DTW upper threshold setting, and when residue sample is when DTW deviation is all less than regulation DTW upper threshold between any two, remains sample and meets stability condition.
In the embodiment of the present invention, if active user does not also carry out sample collection, or active user's sample collection does not also all complete, cause template base also successfully not set up, even if there is so the signature to be measured of current signature, signature so to be measured also cannot complete authentication, and system is shown as un-authenticated state.
When the collection through sample successfully gathers after 5 samples, DTW index and the deviation of 5 samples of system-computed, judge the stability of 5 samples.When 5 samples meet stability condition, sample is chosen successfully, enters the establishment stage of template base; When 5 samples do not meet stability condition, system requirements user appends 5 new samples, proceeds to collection place of signature sample.When the collection through sample successfully gathers after 10 samples, DTW index and deviation between system-computed sample, from 10 samples, circulation weeds out the sample of 5 deviation maximums, finally retains 5 and treats judgement sample.When remaining 5 samples meet stability condition, sample is chosen successfully, enters the establishment stage of template base; When 5 samples of residue do not meet stability condition, it is too poor that system prompt active user writes the stability of signature, after suggestion user adds training again, then carries out the collection of sample.
Step 2-4, using the remaining stable signature sample of choosing as template base sample, calculation template storehouse sample DTW value between any two, and it is added up, determine the stable DTW value deviation dynamic threshold of signature coordinate of residue, the dynamic discriminant strategy of the DTW value deviation dynamic threshold of the DTW value deviation dynamic threshold of pressure, speed and applicable current signature, and deposit above-mentioned information in template base; Described determine the dynamic discriminant strategy that is applicable to current signature, described dynamic discriminant strategy comprises: rising of signature started to write a little and the extreme point of coordinate carries out segmentation and start to write the angular velocity extreme point of point, signature and the extreme point of the linear velocity of signing of signature are carried out to segmentation as waypoint as waypoint; Select segmentation method: be chosen under this segmentation method, signature sample is the method for DTW value deviation minimum between any two.
In the embodiment of the present invention, determine the method for dynamic threshold: according to residue, stablize the sample maximal value of coordinate DTW range deviation between any two, determine the dynamic threshold of coordinate DTW deviation, i.e. the deviation upper limit.In the embodiment of the present invention, it is S1, S2, S3 that residue is stablized sample, and wherein the DTW of S1 and S2 distance is d
12, the DTW distance of S1 and S3 is d
13, the DTW distance of S2 and S3 is d
23, ask the mean value DTW of DTW distance
avg=(d
12+ d
13+ d
23)/3, finally, according to each DTW distance, DTW average, solve the deviation of each DTW distance, thereby determine dynamic threshold.
Step 3, gather tested user's signature sample data, the movement locus of lettering pen is carried out to constant duration sampling, obtain X coordinate figure, Y coordinate figure and the contact force value to clipboard of sampled point; And tested signature sampling number certificate is carried out to validity check, and if signed data is qualified, allow the submission of signed data, perform step 4; Otherwise return to execution step 3;
Step 4, judge the true and false of tested signature;
Step 4-1, the employing of adding up tested signature count, sign coordinate extreme point type (X extreme point, Y extreme point, X and Y be extreme point simultaneously), the coordinate extreme point quantity of signature, angular velocity extreme point quantity, the linear velocity extreme point type of signature and the linear velocity extreme point quantity of signature of the angular velocity extreme point type (angular velocity minimal value, angular velocity maximum value) of signature, signature of rise and fall stroke count amount, signature of stroke; In conjunction with the dynamic threshold in template base, and based on above-mentioned information, signature to be measured is carried out to dynamic filter, judge that whether tested signature meets dynamic filter condition, if so, performs step 4-2; Otherwise tested signature is pseudo-signature, finishes;
Specimen sample count average ± 20% scope in; The stroke count amount of rising and falling average ± 2 stroke counts within the scope of; Signature coordinate extreme point quantity average ± 20% scope in; Signature angular velocity extreme point quantity average ± 20% scope in; Signature linear velocity extreme point quantity average ± 20% scope in.
Step 4-2: select the dynamic discriminant strategy that is applicable to current signature to carry out staging treating to tested signature;
Step 4-3: according to the coordinate of tested signed data, pressure and speed, calculate the DTW value deviation between coordinate, pressure and the speed of all signature sample data in itself and template base; Form the Score index of multi-expert marking, as the support information of the follow-up judgement signature true and false to be measured.
Step 4-4: according to the overall DTW value deviation of coordinate between the signature sample of having deposited in signed data to be measured and template base, coordinate segmentation summation DTW value deviation, coordinate segmentation key point DTW value deviation, the overall DTW value deviation of pressure between signature sample in signed data to be measured and template base, pressure segmentation summation DTW value deviation, pressure segmentation key point DTW value deviation, the overall DTW value of the speed of signature sample deviation in signed data and template base, speed segmentation and DTW value deviation, speed waypoint DTW value deviation, respectively in conjunction with the dynamic threshold of signature sample coordinate DTW value deviation in template base, the dynamic threshold of the dynamic threshold of pressure DTW value deviation and speed DTW value deviation, the method that adopts multi-expert ballot to mark judges the true and false of signature to be measured, the DTW value quantity that is even less than dynamic threshold is greater than half of DTW value sum, signature to be measured is judged as true signature, otherwise signature to be measured is judged as pseudo-signature, forbid that tested user logs in,
In the embodiment of the present invention, use the mode of many index ballots, for example, in above-mentioned 9 DTW values, if 6 are less than the corresponding dynamic threshold upper limit in template base above, so just think that signature to be measured is true, otherwise be pseudo-.
Step 4-5: differentiation result is carried out to output display.
The software systems of final the present invention's exploitation have been carried out a large amount of tests, and start to have dropped into practical application.System acquisition 10 people's signed data as experiment, everyone has set up the template base of 5 samples, each signature user has inputted my 10 true signatures and 50 other people signatures of imitating afterwards, amount to 100 really sign and 500 copy signature.Finally experimental result shows, signature of the present invention is refused rate lower than 10% by mistake, and misclassification rate is 0.25%, and the overall recognition correct rate of signing is up to 97.8%.To sum up, the online handwriting signature authentication method based on dynamic threshold of the present invention has obtained extraordinary effect.
Claims (6)
1. the online handwriting signature authentication method based on dynamic threshold, is characterized in that, comprises the following steps:
Step 1, gather my a plurality of signature sample data, the movement locus of lettering pen is carried out to constant duration sampling, obtain X coordinate figure, Y coordinate figure and the contact force value to clipboard of sampled point; And all sampling numbers certificates of signature are carried out to validity check, and if signed data is qualified, allow the submission of signed data, perform step 2; Otherwise return to execution step 1;
Step 2, signature sample choose the foundation with template base;
Whether the quantity of the signature sample that step 2-1, judgement gather reaches the sample of setting is chosen requirement, if so, performs step 2-2; Otherwise, return to execution step 1;
Step 2-2, using rising of signature start to write a little and the extreme point of coordinate as waypoint, maybe using the extreme point of the angular velocity extreme point that works start to write point, signature of signature and signature linear velocity as waypoint, the staging treating of realization to signed data;
Step 2-3, employing dynamic time programming DTW method judge the similarity degree of two signatures, obtain the final sample of template base;
Step 2-3-1, calculate the overall DTW value of all signatures coordinate between any two, coordinate segmentation summation DTW value, coordinate segmentation key point DTW value, the overall DTW value of pressure, pressure segmentation sue for peace DTW value, pressure segmentation key point DTW value, the overall DTW value of speed, speed segmentation and DTW value and speed waypoint DTW value respectively, and calculate above-mentioned DTW value average and deviation;
Step 2-3-2, delete the signature sample of sampling number grid deviation maximum and DTW value deviation maximum in all samples;
Step 2-3-3, repeatedly perform step 1 to step 2-3-3 until the volume residual of sample reaches the sample size of the required reservation in final sample storehouse;
Step 2-3-4, the whether satisfied signature stability condition of judgement residue sample, i.e. whether judgement residue sample between any two DTW deviation is all less than the DTW threshold value that user sets, and if so, will remain the final sample of signature sample as template base; Otherwise return to execution step 1;
Step 2-4, using the remaining stable signature sample of choosing as template base sample, calculation template storehouse sample DTW value between any two, and it is added up, determine the stable DTW value deviation dynamic threshold of signature coordinate of residue, the dynamic discriminant strategy of the DTW value deviation dynamic threshold of the DTW value deviation dynamic threshold of pressure, speed and applicable current signature, and deposit above-mentioned information in template base;
Step 3, gather tested user's signature sample data, the movement locus of lettering pen is carried out to constant duration sampling, obtain X coordinate figure, Y coordinate figure and the contact force value to clipboard of sampled point; And tested signature sampling number certificate is carried out to validity check, and if signed data is qualified, allow the submission of signed data, perform step 4; Otherwise return to execution step 3;
Step 4, judge the true and false of tested signature;
Step 4-1, add up tested signature sampling number, signature stroke the coordinate extreme point type of rise and fall stroke count amount, signature, the angular velocity extreme point type of the coordinate extreme point quantity of signature, signature, the linear velocity extreme point type of the angular velocity extreme point quantity of signature, signature and the linear velocity extreme point quantity of signature; In conjunction with the dynamic threshold in template base, and based on above-mentioned information, signature to be measured is carried out to dynamic filter, judge that whether tested signature meets dynamic filter condition, if so, performs step 4-2; Otherwise tested signature is pseudo-signature, finishes;
Step 4-2: select the dynamic discriminant strategy that is applicable to current signature to carry out staging treating to tested signature;
Step 4-3: according to the coordinate of tested signed data, pressure and speed, calculate the DTW value between coordinate, pressure and the speed of all signature sample data in itself and template base;
Step 4-4: according to the overall DTW value deviation of coordinate between the signature sample of having deposited in signed data to be measured and template base, coordinate segmentation summation DTW value deviation, coordinate segmentation key point DTW value deviation, the overall DTW value deviation of pressure between signature sample in signed data to be measured and template base, pressure segmentation summation DTW value deviation, pressure segmentation key point DTW value deviation, the overall DTW value of the speed of signature sample deviation in signed data and template base, speed segmentation and DTW value deviation, speed waypoint DTW value deviation, respectively in conjunction with the dynamic threshold of signature sample coordinate DTW value deviation in template base, the dynamic threshold of the dynamic threshold of pressure DTW value deviation and speed DTW value deviation, the method that adopts multi-expert ballot to mark judges the true and false of signature to be measured, the DTW value quantity that is even less than dynamic threshold is greater than half of DTW value sum, signature to be measured is judged as true signature, otherwise signature to be measured is judged as pseudo-signature, forbid that tested user logs in,
Step 4-5: differentiation result is carried out to output display.
2. the online handwriting signature authentication method based on dynamic threshold according to claim 1, it is characterized in that, effect described in step 1 and step 3 checks, comprising that the coordinate write of signature crosses the border checks and signature reasonalbeness check, and whether the length and width scope of always the counting of signature, total writing time, signature reaches the threshold value of setting.
3. the online handwriting signature authentication method based on dynamic threshold according to claim 1, it is characterized in that, the dynamic discriminant strategy that determine to be applicable to current signature described in step 2-4 and step 4-2, described dynamic discriminant strategy comprises: rising of signature started to write a little and the extreme point of coordinate carries out segmentation and start to write the angular velocity extreme point of point, signature and the extreme point of the linear velocity of signing of signature are carried out to segmentation as waypoint as waypoint; Select segmentation method: be chosen under this segmentation method, signature sample is the method for DTW value deviation minimum between any two.
4. the online handwriting signature authentication method based on dynamic threshold according to claim 1, it is characterized in that, the DTW value deviation dynamic threshold of the signature coordinate that the definite residue described in step 2-4 is stable, be specially: determine to remain and stablize sample coordinate DTW distance between any two, obtain whole coordinate DTW range averaging values, and according to each coordinate DTW distance and coordinate DTW range averaging value, obtain the deviation of DTW distance, select deviation maximal value as the DTW value deviation dynamic threshold of coordinate; The DTW value deviation dynamic threshold value determination method of the DTW value deviation dynamic threshold of pressure and speed determines that with the DTW value deviation dynamic threshold of coordinate method is identical.
5. the online handwriting signature authentication method based on dynamic threshold according to claim 1, is characterized in that, the coordinate extreme point type of the signature described in step 4-1 comprises X extreme point, Y extreme point, X and Y extreme point simultaneously; The angular velocity extreme point type of signature comprises angular velocity minimal value and angular velocity maximum value.
6. the online handwriting signature authentication method based on dynamic threshold according to claim 1, is characterized in that, the dynamic filter condition described in step 4-1 is: specimen sample count average ± 20% scope in; The stroke count amount of rising and falling average ± 2 stroke counts within the scope of; Signature coordinate extreme point quantity average ± 20% scope in; Signature angular velocity extreme point quantity average ± 20% scope in; Signature linear velocity extreme point quantity average ± 20% scope in.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310521026.6A CN103593673B (en) | 2013-10-27 | 2013-10-27 | A kind of on-line trial authentication method based on dynamic threshold |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310521026.6A CN103593673B (en) | 2013-10-27 | 2013-10-27 | A kind of on-line trial authentication method based on dynamic threshold |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103593673A true CN103593673A (en) | 2014-02-19 |
| CN103593673B CN103593673B (en) | 2016-07-20 |
Family
ID=50083805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310521026.6A Expired - Fee Related CN103593673B (en) | 2013-10-27 | 2013-10-27 | A kind of on-line trial authentication method based on dynamic threshold |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103593673B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105095709A (en) * | 2015-09-09 | 2015-11-25 | 西南大学 | On-line signature identification method and on-line signature identification system |
| CN105447350A (en) * | 2014-08-07 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Identity authentication method and device |
| CN105844726A (en) * | 2016-03-18 | 2016-08-10 | 吉林大学 | Handwritten signature sign-in management system |
| CN106548133A (en) * | 2016-10-17 | 2017-03-29 | 歌尔科技有限公司 | A kind of template matching method and device and gesture identification method and device |
| CN106778568A (en) * | 2016-12-05 | 2017-05-31 | 上海携程商务有限公司 | The processing method of the identifying code based on WEB page |
| CN107302433A (en) * | 2016-04-15 | 2017-10-27 | 平安科技（深圳）有限公司 | Method of calibration, verification server and the user terminal of electronic signature |
| CN110866499A (en) * | 2019-11-15 | 2020-03-06 | 爱驰汽车有限公司 | Handwritten text recognition method, system, device and medium |
| CN111291636A (en) * | 2020-01-19 | 2020-06-16 | 深圳壹账通智能科技有限公司 | Method, device and system for effectively identifying electronic signature and storage medium |
| CN111985319A (en) * | 2020-07-13 | 2020-11-24 | 上海眼控科技股份有限公司 | Signature identification method and device |
| CN112035825A (en) * | 2020-09-01 | 2020-12-04 | 中国银行股份有限公司 | Method for logging in application APP, client and server |
| CN113468987A (en) * | 2021-06-17 | 2021-10-01 | 傲雄在线(重庆)科技有限公司 | Electronic handwriting authentication method, system, electronic equipment and storage medium |
| CN114138166A (en) * | 2021-11-24 | 2022-03-04 | 安徽中科美络信息技术有限公司 | Hand-written signature method and terminal |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5825906A (en) * | 1994-11-30 | 1998-10-20 | Nippondenso Co., Ltd. | Signature recognition system |
| CN101079707A (en) * | 2007-06-21 | 2007-11-28 | 中国科学院合肥物质科学研究院 | Identity authentication method based on revocable handwritten signature |
| CN102592142A (en) * | 2012-01-05 | 2012-07-18 | 中国科学院合肥物质科学研究院 | Computer-system-based handwritten signature stability evaluation method |
-
2013
- 2013-10-27 CN CN201310521026.6A patent/CN103593673B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5825906A (en) * | 1994-11-30 | 1998-10-20 | Nippondenso Co., Ltd. | Signature recognition system |
| CN101079707A (en) * | 2007-06-21 | 2007-11-28 | 中国科学院合肥物质科学研究院 | Identity authentication method based on revocable handwritten signature |
| CN102592142A (en) * | 2012-01-05 | 2012-07-18 | 中国科学院合肥物质科学研究院 | Computer-system-based handwritten signature stability evaluation method |
Non-Patent Citations (1)
| Title |
|---|
| 栾方军等: "基于HMM的在线手写签名认证系统设计与实现", 《计算机应用与软件》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10614201B2 (en) | 2014-08-07 | 2020-04-07 | Alibaba Group Holding Limited | Method and device for identity authentication |
| CN105447350A (en) * | 2014-08-07 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Identity authentication method and device |
| CN105447350B (en) * | 2014-08-07 | 2019-10-01 | 阿里巴巴集团控股有限公司 | A kind of identity identifying method and device |
| US10795978B2 (en) | 2014-08-07 | 2020-10-06 | Alibaba Group Holding Limited | Method and device for identity authentication |
| CN105095709A (en) * | 2015-09-09 | 2015-11-25 | 西南大学 | On-line signature identification method and on-line signature identification system |
| CN105844726A (en) * | 2016-03-18 | 2016-08-10 | 吉林大学 | Handwritten signature sign-in management system |
| CN107302433A (en) * | 2016-04-15 | 2017-10-27 | 平安科技（深圳）有限公司 | Method of calibration, verification server and the user terminal of electronic signature |
| CN106548133A (en) * | 2016-10-17 | 2017-03-29 | 歌尔科技有限公司 | A kind of template matching method and device and gesture identification method and device |
| CN106548133B (en) * | 2016-10-17 | 2019-04-23 | 歌尔科技有限公司 | A kind of template matching method and device and gesture identification method and device |
| CN106778568A (en) * | 2016-12-05 | 2017-05-31 | 上海携程商务有限公司 | The processing method of the identifying code based on WEB page |
| CN106778568B (en) * | 2016-12-05 | 2020-08-21 | 上海携程商务有限公司 | Method for processing verification code based on WEB page |
| CN110866499A (en) * | 2019-11-15 | 2020-03-06 | 爱驰汽车有限公司 | Handwritten text recognition method, system, device and medium |
| CN111291636A (en) * | 2020-01-19 | 2020-06-16 | 深圳壹账通智能科技有限公司 | Method, device and system for effectively identifying electronic signature and storage medium |
| WO2021142973A1 (en) * | 2020-01-19 | 2021-07-22 | 深圳壹账通智能科技有限公司 | Electronic signature validity identification method and system, and apparatus and storage medium |
| CN111985319A (en) * | 2020-07-13 | 2020-11-24 | 上海眼控科技股份有限公司 | Signature identification method and device |
| CN112035825A (en) * | 2020-09-01 | 2020-12-04 | 中国银行股份有限公司 | Method for logging in application APP, client and server |
| CN113468987A (en) * | 2021-06-17 | 2021-10-01 | 傲雄在线(重庆)科技有限公司 | Electronic handwriting authentication method, system, electronic equipment and storage medium |
| CN114138166A (en) * | 2021-11-24 | 2022-03-04 | 安徽中科美络信息技术有限公司 | Hand-written signature method and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103593673B (en) | 2016-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103593673B (en) | A kind of on-line trial authentication method based on dynamic threshold | |
| US9589120B2 (en) | Behavior based authentication for touch screen devices | |
| Zheng et al. | An efficient user verification system via mouse movements | |
| CN102592152B (en) | Computer-system-based online handwriting authentication method | |
| CN100354882C (en) | Information processing apparatus and signature data input programs | |
| EP2874099A1 (en) | Dynamic handwriting verification and handwriting-based user authentication | |
| CN103023658B (en) | Identity authentication method and identity authentication system based on signature | |
| CN104318138A (en) | Method and device for verifying identity of user | |
| Xu et al. | Challenge-response authentication using in-air handwriting style verification | |
| CN103455741A (en) | Character-based on-line handwriting authentication template extension method | |
| Sanchez-Reillo et al. | Improving presentation attack detection in dynamic handwritten signature biometrics | |
| CN102592142A (en) | Computer-system-based handwritten signature stability evaluation method | |
| Sae-Bae et al. | Distinctiveness, complexity, and repeatability of online signature templates | |
| Alariki et al. | TOUCH GESTURE AUTHENTICATION FRAMEWORK FOR TOUCH SCREEN MOBILE DEVICES. | |
| AU2016236466B2 (en) | Method for identification of user's interaction signature | |
| CN103995995A (en) | Multimedia signature identification method and system | |
| JP6924770B2 (en) | Dynamic movement tracking infrastructure for spatially divided segments Signature authentication system and method | |
| WO2015032304A1 (en) | Online handwriting and identity authentication method having capability for identifying identity of attacker | |
| Patil et al. | An efficient DTW algorithm for online signature verification | |
| CN101894154B (en) | Method for extracting key pattern from image file | |
| CN103116750A (en) | Pattern identification authentication method capable of faintly matching | |
| Kang et al. | User interface-based repeated sequence detection method for authentication | |
| JP6168645B2 (en) | Reverse Turing test method and access authentication method | |
| JP6820580B2 (en) | Pen input personal recognition method | |
| Gonçalves et al. | Time/space based biometric handwritten signature verification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160720 Termination date: 20171027 |