CN103136546A - Multi-dimension authentication method and authentication device of on-line signature - Google Patents
Multi-dimension authentication method and authentication device of on-line signature Download PDFInfo
- Publication number
- CN103136546A CN103136546A CN2013100461041A CN201310046104A CN103136546A CN 103136546 A CN103136546 A CN 103136546A CN 2013100461041 A CN2013100461041 A CN 2013100461041A CN 201310046104 A CN201310046104 A CN 201310046104A CN 103136546 A CN103136546 A CN 103136546A
- Authority
- CN
- China
- Prior art keywords
- signature
- value
- user
- counting
- verification msg
- 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.)
- Pending
Links
Images
Abstract
The invention belongs to the technical field of communication electronics, and provides a multi-dimension authentication method and an authentication device of an on-line signature. The multi-dimension authentication method comprises the following steps: when a user signs, authentication data of the user signature are extracted; the user signature is displaced to a preset position, and points of the user signature in the preset position are resized into preset points; the size of the user signature with the point numbers set is processed in a normalization mode to a set size; and authentication data of each point of the user signature within the obtained set points and set size are obtained; the authentication data of each point within the set points are compared with the authentication data of an original signature, and the similarity between the user signature and the original signature is obtained according to the comparison result; if the similarity is smaller than a threshold value, the user signature is affirmed to be the same signature, and if the similarity is larger than or equal to the threshold value, the user signature is affirmed not to be the same signature. By means of the technical scheme, the multi-dimension authentication method and the authentication device of the on-line signature have the advantage of being high in authentication speed.
Description
Technical field
The invention belongs to the communications electronics field, relate in particular to various dimensions verification method, the device of signing on a kind of line.
Background technology
Biology provides a variety of methods of personal checking.Biologic verification method is different from common personal identification method used, and biology means that checking does not need to rely on the help of any foreign object, such as: key, magnetic card carves a seal etc., does not also need extra memory cipher etc.In the middle of numerous biometric verification of identity methods, personal identification method based on handwritten signature mainly has again two kinds of advantages with respect to the other biological verification method: without aggressiveness and widespread popularity, other biometric verification of identity method all needs to use specific instrument to gather human body certain bio signal with it, and handwritten signature need not to gather bio signal, only need to gather people's handwritten signature, therefore the aggressiveness of unmatchful human body, and adopt signature to verify that to document very long history is arranged, signature be also in daily life use very extensive.
The handwritten signature verification method can be divided into two kinds: on line (dynamically) with line under (static state).The key distinction of these two kinds of methods is the different obtain manners to hand-written signed data.It is to obtain the signature track data in hand-written process that signed data on line obtains.And the data acquisition under line is generally to obtain by scanner or camera the signature of having finished writing, and the data that obtain like this are gray level images.Each factor on the time-space domain on line that is to say is processed, and only processing on spatial domain under line.
Mainly contain aspect four in the identification checking of handwritten signature: a: the acquisition of handwritten signature related data (DataAcquisition), b: the data that obtain are carried out pre-service (Preprocessing), c: the extraction of signature character (Feature Extraction), d: contrast verification algorithm (Comparison).
Need data are carried out pre-service, to strengthen data characteristic after the acquisition signed data.Mainly use at data preprocessing phase the method that several signals are processed.For the signed data that obtains on line, at first can use wave filter to carry out filtering to data, remove noise information.Then need time domain and the spatial domain of signature signal are done standardization.
G.Dimauro, S.Impedovo, G.Pirlo; " Component-oriented algorithms forsignature verification "; IJPRAI; Vol.8; n.3; 1994; pp.771-794 has defined a kind of composition of handwritten signature, he is falling and the person's handwriting of pen between lifting splits and be used as a kind of composition at pen that the people writes, and they think that handwritten signature can be regarded as the composition of a series of hand-written unit, and hand-written unit is that the unit by a series of unexpected interruptions splits.
R.Plamondon, P.Yergeau and J.J.Brault, " A multi-level signature verificationsystem ", in From Pixels to Feaures III-Frontiers in Handwriting Recognition, S.Impedovo and J.C.Simon eds, Elsevier Publ., pp363-370,1992 people such as grade have also proposed a kind of dividing method.Their dividing method is the curve speed in motion process and angular velocity according to pen.In this dividing method, they suppose that the fundamental element of handwritten signature is part component and character string.
J.J.Braut and R.Plamondon, " Segmenting handwritten signaures at theirperceptually important points ", IEEE T-PAMI, Vol.15n.91993, pp.953-957 have also proposed another kind of method.This method is extracted some the important points above handwriting.This process is divided into two steps, at first give a weighted value to each point above the handwriting that acquires, then those are accounted for important point make marks in whole font, these points are to check by artificial means and mark, then take out these points.
G.Dimauro, S.Impedovo and G.Pirlo, " On-line Signature Verification by aDynamic Segmentation Technique ", in Proc.3th IWFHR, Buffalo, May1993, pp.262-271 have proposed the method for another dynamic cutting data.Basic thought is that the handwritten signature of appointment is assessed according to the sample of establishing in advance.This technology only has those to have the cut-point that clearly contrasts appointment just can be exported by using the signature of DP algorithm comparison appointment, and this method allows the signature of test signature and appointment is divided into some the same number of partitioning portions.
Two kinds of methods are mainly also being arranged: the parametric method sum function method aspect the feature extraction of handwritten signature.In the situation that with parametric method as feature, the element that is characterized as being a vector of signature, the value of the feature of element representation in each vector.When using function as feature, handwritten signature is characterized as being the function of a time, and functional value has formed the set of signature character.
Y.Mizukami, M.Yoshimura and H.Miik, " An off-line signature verificationsystem using an extracted displacement function ", Pattern Recognition Letters, V.23, the people such as 2002, pp.1569-1577 have proposed a kind of method of feature extraction of displacement.With the minimal value of function, signature is carried out extreme value and be shifted to obtain eigenwert.Here the function of using is the quadratic sum of the Euclidean distance of two handwritten signature data, and be provided with a penalty coefficient and come level and smooth this extremal function, and having taked by slightly avoiding calculating extreme value to the strategy of essence is to obtain local optimum solution, just can calculate the difference of the signature of signature to be verified and appointment by this displacement function, thus the authenticity of certifying signature.
C.N.Liu, N.M.Herbst and N.J.Anthony, " Automatic Signature Verification:System Description and Field Test Results ", IEEE T-SMC, Vol.91979, pp.35-38. wait the people to propose a kind of handwritten signature verification method of reinforcement, their method is based on two kinds of person's handwriting variablees, i.e. the acceleration of person's handwriting and the pressure of writing.Carry out constantly revising generator program when Characteristic Contrast adapts at the signature to the handwritten signature of input and the contrast that provides, adapting to inconsistent in the submission process, and provide the supervision to the procedure of adaptation.Waveform by contrast acceleration and pressure is signed with identification.
I.Yoshimura and M.Yoshimura, " On-line signaure verification incorporatingthe direction of pen movement – An experimental examination of the effectiveness ", in From Pixels to Feaures III-Frontiers in Handwriting Recognition, S.ImpedovoandJ.C.Simon eds, Elsevier Pul, .pp.353-362,1992. people such as grade have proposed a kind of handwritten signature method on line equally.Their method is mainly the direction of each point on the person's handwriting that extracts handwritten signature.At first they construct a method of effectively weighing the difference of signature with the person's handwriting direction, then choose some features above the signature sample that will verify of appointment and do cluster, the value of the difference of the signature by calculating signature more to be verified and appointment at last, if this value is for surpassing a threshold values, expression is proved to be successful.
H.D.Crane and J.S.Ostrem, " Automatic Signature Verification Using aThree-Axis Force-Sensitive Pen ", IEEE T-SMC, Vol.13, n.3, the people such as 1983, pp.329-337. have also proposed a kind of based on dynamic Handwritten Signature Recognition Method on line.They come dynamic handwriting is converted to electric signal with a kind of power sensor pen of three axles, to be used for doing real-time analysis.So they adopt the different powers of people when writing to do identification.
W.Nelson, W.Turin and T.Hastie, " Statistical methods for on-line signatureverification ", JJPRAI, v.8, n.3, the people such as 1994, pp.749-770. have discussed three kinds of handwritten signature verification methods based on statistical method in their article.These three kinds of methods all are based on a kind of statistical model, the signature font being carried out some dynamic statistics features of statistical hypothesis and signature summarizes, two kinds of methods have wherein only been used the statistical nature of signature, and calculate signature to be measured and the error metric of specifying signature with Euclidean distance.The third method is not only added up the statistical nature of actual signature, also calculates the statistical nature of false person's handwriting, and with the positive example classification of genuine person's handwriting as training, false person's handwriting is as negative counter-example classification.Use at last the method for machine learning to align the counter-example classification.
M.Castellano, S.Impedovo, A.mingolla, G.Pirlo; " A spectral analysis basedsignature verification system "; in Lecture Notes in Computer Science:Recent Issuesin Pattern Analysis and Recognition; eds.G.Goos and J.Hartmanis; Springer Verlag; Berlin; the people such as 1988, pp.316-323. have proposed a kind of method of handwritten signature being analyzed identification in the frequency domain space of one dimension.They are the waveforms to a comparison that at first obtain from handwriting signature, then extract harmonic content and select some unique points from waveform, use these two features of analyzing as handwritten signature.Same they analyze the signature of authentic and valid signature and forgery, with a maximum differential that calculates effective signature and forge a signature, then with this effective signature that obtains and the difference that forges a signature, the signature in future are verified.
C.F.Lam and D.Kamins, " Signaure recognition through spectral analysis ", Pattern Recog.22,1, at first the people such as 1989, pp.39-44. extract several features of handwritten signature: font, and person's handwriting moves, fine feature and time etc., then use some transforming function transformation functions to carry out conversion to the feature of extracting.The transform method that uses has hadamard matrix conversion and Walsh transform.And with the nonlinear curvature function, feature is compared with standard handwriting on time domain.At first use fast fourier transform that top several features are converted on spatial domain, re-use two top conversion and be further processed.
G.Dimauro, S.Impedovo, G.Pirlo, A.Salzo, " A multi-expert signaureverification system for bank check processing ", IJPRAI, Vol.11, n.5, the people such as 1997, pp.827-844. put forward the compound handwritten signature verification method of one kind of multiple technology in their paper.This signature system of their design is used the check signature of bank especially and is optimized.Due to the privacy level requirement of banking system, they combine three kinds of different signature verification algorithms.The first algorithm has used the method for the bulk treatment of identifying the handwriting, and the second and the third method have been used the method for element guiding.The method of structure is processed in main employing in the second approach, and the third method is used the high neural net method that adapts to.Use the ballot method that signature is verified in conjunction with these three kinds of methods at last.
D.Letjman and S.George, " On-line handwritten signature verification usingwavelets and back-propagation neural networks ", Proc.Of ICDAR01, Seattle, 2001, pp.596-598. wait the people to propose the dynamic handwritten signature verification method of a kind of use, they identify handwritten signature with wavelet transformation in conjunction with the method for recalling neural network.In their experimental result, their method relatively on other dynamic line the method discrimination relatively good.
In the comparison stage of handwritten signature, by signature extraction feature to be verified and known signature character are mated, whether successfully draw at last coupling.In the characteristic matching stage, several main algorithms are arranged also.
R.Sabourin, G.Genesi and F.Preteux, " Off-line Signaure Verification byLocal Granulometric Size Distributions ", IEEE T-PAMI, Vol.19, n.9, the people such as 1997, pp.976-988. just use Euclidean distance as the difference of differentiating between different handwritten signatures.At first select a distance function, then use this distance function to calculate the functional value of different hand-written features, calculate two distances between signature, then judge with the size of distance whether two signatures are that same person is write.
Wirtz B., " Stroke-based Time Warping for Signature Verification ", Proc.ICDAR1995, IEEE Press, the people such as pp.179-182. have proposed a kind of new dynamic signature verification method based on Elastic Matching.The force value of person's handwriting is counted as the location parameter of function, and is used to do local assessment.What they extracted the person's handwriting unit retouches the limit structure, goes retouching the verification algorithm that the limit structure puts into.
A kind of being associated property that adopts the zone, the verification method of Dynamic Time Warping and the tree-like coupling of skeleton is by M.Perizeau and R.Plamondon, " A comparative analysis of regional correlation; dynamic time warping and skeletal tree matching for signaure verification ", IEEET-PAMI, Vol.12, n.71990, the people such as pp.710-717. propose.also has Q.Z.Wu, S.Y.Lee and I.C.Jou, " On-line signature verification based on split and merge matching mechanism ", Patten Recognition Letters, Vol.18, 1997, pp.665-673. propose based on the matching strategy of cutting apart with merge algorithm, A.K.Jaim, F.D.Griess and S.D.Connell, " On-line signatureverification ", Pattern Recognition, Vol.35, 2002, the people such as pp.2963-2972 propose the matching algorithm of person's handwriting changed string, neural network algorithm L.Y.Tseng and T.H.Huang, " An on-lineChinese signature verification scheme based on the ART neural network ", Proc.OfInt.J.Conf.on NN, Maryland, 1992, pp.624-630., the Markov model algorithm M.Fuentes that hides, S.Gaci and B.Dorizzi, " On line signature Verification:Fusion of aHidden Markov Model and a Neural Network via a Support Machine ", Proc.OfIWFHR, Canada, 2002, pp.253-258. waiting is all the classification and identification algorithm that is used for the handwritten signature checking of commonly using.
More existing verification algorithms are confined to theory mostly, lack practical application, in addition, the speed of existing verification algorithm is slow, and for example neural network algorithm or hiding Markov model scheduling algorithm speed are slower, and but algorithm function is complicated and may not be certain ripe good the use.In addition, existing algorithm is not processed the font size variation, for the processing power of the signature not of uniform size of homogeneous not and recognition capability all a little less than.
Summary of the invention
The purpose of the embodiment of the present invention is to provide the various dimensions of signing on a kind of line verification method, is intended to solution large by calculated amount in prior art, and processing speed is slow, processes weak problem for the font size variation.
The embodiment of the present invention is achieved in that the various dimensions verification method of signing on a kind of line, and described method comprises:
When user's signature, extract the verification msg of described user's signature; Described verification msg comprises: the x axial coordinate of signature, y axial coordinate, force value and time value.
Described user's signature is displaced to desired location, and the quantity of counting of the described user's signature on desired location is zoomed to the designated value of set point number;
The scope square size that will have the user's signature of the designated value of setting up an office is carried out normalized to the size of setting;
After completing above-mentioned setting, obtain the verification msg of the point of completing the user's signature after setting according to described verification msg;
Compare the verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature, and obtain the similarity of user's signature and original signature according to comparison result; Less than threshold value, be defined as same signature as similarity, more than or equal to threshold value, be defined as non-same signature as similarity.
Optionally, described comparison is described completes the verification msg of each point of the verification msg of each point of the user's signature after setting and original signature, and obtains the similarity of user's signature and original signature according to comparison result, specifically comprises:
Calculate the absolute value of difference of verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature, all absolute value additions namely obtained both difference condition, this absolute value and be similarity.
Optionally, the designated value that the quantity of counting of the described user's signature on desired location is zoomed to set point number comprises:
The quantity of counting of user's signature is greater than setting value as described, and the designated value of the set point number of sampling out from the counting of this user's signature is counted;
The quantity of counting of user's signature is less than set-point value as described, inserts in from the counting of user's signature and counts so that counting of described user's signature reaches the designated value of set point number.
Optionally, the described verification msg of the described foundation verification msg of obtaining the point of completing the user's signature after setting comprises:
The quantity of counting of user's signature is greater than described designated value as described, obtain each verification msg of counting in counting after sampling, wherein x axial coordinate value in this verification msg, y axial coordinate value are for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature;
The quantity of counting of user's signature is less than described designated value as described, obtain the verification msg of counting of user's signature, wherein, wherein x axial coordinate value in this verification msg, y axial coordinate value for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature; Obtain and insert the verification msg of counting, wherein x axial coordinate value, x axial coordinate value, the y axial coordinate value of y axial coordinate value for counting from the described insertion of setting the user's signature of size in this insertion verification msg of counting; In the verification msg that this insertion is counted, force value is the mean value of the described insertion adjacent two point pressure values of counting, and in the verification msg that this insertion is counted, time value is the mean value of described insertion adjacent 2 time values of counting.
On the other hand, provide signature verification device on a kind of line, described device comprises:
Extraction unit is used for extracting the verification msg of described user's signature when user's signature; Described verification msg comprises: the x axial coordinate of signature, y axial coordinate, force value and time value.
Displacement unit is used for described user's signature is displaced to desired location;
Unit for scaling is used for and the quantity of counting of the described user's signature on desired location is zoomed to the designated value of set point number;
The size setup unit carries out normalized to the size of setting for the scope square size of the user's signature that will have the designated value of setting up an office;
Computing unit is used for after completing above-mentioned setting, obtains the verification msg of the point of completing the user's signature after setting according to described verification msg;
Comparing unit is used for comparing the verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature; And obtain the similarity of user's signature and original signature according to comparison result; Less than threshold value, be defined as same signature as similarity, more than or equal to threshold value, be defined as non-same signature as similarity.
Optionally, described comparing unit specifically is used for calculating the absolute value of difference of verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature, all absolute value additions are namely obtained both difference condition, this absolute value and be similarity.
Optionally, described unit for scaling specifically is used for
The quantity of counting of user's signature is greater than setting value as described, and the designated value of the set point number of sampling out from the counting of this user's signature is counted;
The quantity of counting of user's signature is less than set-point value as described, inserts in from the counting of user's signature and counts so that counting of described user's signature reaches the designated value of set point number.
Optionally, described computing unit specifically is used for
The quantity of counting of user's signature is greater than described designated value as described, obtain each verification msg of counting in counting after sampling, wherein x axial coordinate value in this verification msg, y axial coordinate value are for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature;
The quantity of counting of user's signature is less than described designated value as described, obtain the verification msg of counting of user's signature, wherein, wherein x axial coordinate value in this verification msg, y axial coordinate value for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature; Obtain and insert the verification msg of counting, wherein x axial coordinate value, x axial coordinate value, the y axial coordinate value of y axial coordinate value for counting from the described insertion of setting the user's signature of size in this insertion verification msg of counting; In the verification msg that this insertion is counted, force value is the mean value of the described insertion adjacent two point pressure values of counting, and in the verification msg that this insertion is counted, time value is the mean value of described insertion adjacent 2 time values of counting.
Optionally, described device is specially:
Computing machine, panel computer, mobile terminal or signature verification platform.
The embodiment of the present invention compared with prior art, beneficial effect is: technical scheme of the present invention has the font size variation and processes by force, and calculated amount is little, and speed reaches the high advantage of discrimination power soon.
Description of drawings
Fig. 1 is the process flow diagram of the various dimensions verification method of signing on a kind of line of providing of the embodiment of the present invention;
Fig. 2 is the structural drawing of signature verification device on a kind of line of providing of the embodiment of the present invention.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, is not intended to limit the present invention.
The specific embodiment of the invention provides the various dimensions of signing on a kind of line verification method, and the method is completed by Authentication devices, and this Authentication devices is specifically as follows, computing machine, panel computer, mobile terminal or independent verification platform etc.above-mentioned Authentication devices is all preserved the verification msg of original signature, above-mentioned original signature can for, the user is in before original signature, wherein counting of this original signature can be the designated value of set point number, the size of its original signature can be the size of setting, certainly this original signature can be exactly also any one signature of user, at this moment, original signature need to be carried out S12, S13, the operation of S14 makes counting of original signature be the designated value of set point number, make the position of original signature move on to desired location, make the size of scope square size for setting of original signature, the method comprises as shown in Figure 1:
S11, when user's signature, extract the verification msg of user's signature;
Above-mentioned verification msg includes but not limited to: the x axial coordinate of signature, y axial coordinate, force value and time value.
S12, user's signature is displaced to desired location, and the quantity of counting of the user's signature on desired location is zoomed to the designated value of set point number;
S13, the scope square size that will have a user's signature of the designated value of setting up an office are carried out normalized to the size of setting;
S14, complete above-mentioned setting after, obtain the verification msg of the point of completing the user's signature after setting according to described verification msg;
S15, comparison be described completes the verification msg of each point of the verification msg of each point of the user's signature after setting and original signature, and obtains the similarity of user's signature and original signature according to comparison result; Less than threshold value, be defined as same signature as similarity, more than or equal to threshold value, be defined as non-same signature as similarity.
The specific embodiment of the invention is when carrying out signature verification, at first the quantity of counting is according to counting the quantity set point convergent-divergent, then signature size also is processed into and sets size, then calculate similarity, font size or count what will reduce the impact of comparison result like this is so that its font size variation is processed is strong.In addition, method provided by the invention is only the operation that needs simple convergent-divergent, stretching or stack, and calculated amount is little, so its calculated amount is little, speed is fast.
Need to prove, the account form of above-mentioned similarity is specifically as follows: the absolute value of difference of verification msg that calculates each point of the verification msg of each point complete the user's signature after setting and original signature, all absolute value additions are namely obtained both difference condition, this absolute value and be similarity, if similarity is little, the representative of consumer signature is same signer with original signature, if similarity is large, the representative of consumer signature is different signers from original signature.
Optionally, the implementation method that in above-mentioned S12, the quantity of counting of the described user's signature on desired location is zoomed to the designated value of set point number is specifically as follows:
Counting greater than above-mentioned setting value of user's signature as described, the designated value of the set point number of sampling out from the counting of this user's signature is counted.For example, counting of user's signature is 2000, and the designated value of set point number is 1000, directly samples out 1000 points as set point number from the counting of 2000 user's signatures.Need to prove, above-mentioned set point number is specifically as follows, the pixel number of setting, and counting of above-mentioned user's signature is specifically as follows the pixel number of user's signature.
Counting less than above-mentioned setting value of user's signature, count so that user's signature reaches setting value from the interior insertion of counting of user's signature as described.For example, counting of user's signature is 500, and setting value is 1000, inserts a point and make it become 1000 set points in the middle of 2 points.Insert the position of counting this moment and specifically can counting of User signature determine with the ratio of set point number, for example, counting of insertion is identical with counting of user's signature, and the position of inserting is two mid points between point.
Optionally, the implementation method of above-mentioned S13 is specifically as follows:
The font size that the signature of writing due to same person is write at homogeneous not is difference to some extent.Although be the signature of same person, font when font size differs large the time, also can affect the accuracy of checking too, processes so this step is carried out convergent-divergent to the size of signature font, and is unified to same size font.In the scaling of fonts size, whole signature is carried out convergent-divergent.Because signature generally is comprised of several words, and people's custom is to write from left to right.So the size of X-axis is scaled to 3 times of Y-axis size in the middle of system in convergent-divergent, can certainly be other multiple, for example 2 times or 4 times.
Optionally, the method that realizes S14 is specifically as follows:
Counting greater than setting value of user's signature as described obtained each verification msg of counting in counting after sampling, and wherein x axial coordinate value in this verification msg, y axial coordinate value are the x axial coordinate value of counting, y axial coordinate value from the user's signature of setting size; Force value and time value are force value and the time value of the verification msg extracted in S11.
Counting less than setting value of user's signature as described obtained the verification msg of counting of user's signature, and wherein, wherein x axial coordinate value in this verification msg, y axial coordinate value are the x axial coordinate value of counting, y axial coordinate value from the user's signature of setting size; Force value and time value are force value and the time value of the verification msg extracted in S11; Obtain and insert the verification msg of counting, wherein x axial coordinate value in this insertion verification msg of counting, y axial coordinate value x axial coordinate value, the y axial coordinate value for counting from the insertion of the user's signature of setting size; In the verification msg that this insertion is counted, force value is the mean value of the described insertion adjacent two point pressure values of counting, and in the verification msg that this insertion is counted, time value is the mean value of described insertion adjacent 2 time values of counting.
Optionally, the computing method of above-mentioned similarity can adopt the method for curve to calculate, be calculated as example with the x axle, the data fitting of all x axles is become a curve, the data fitting of all x axles of original signature is become a curve, then calculate two curves corresponding count between the absolute value of difference, then the absolute value addition can be obtained similarity.
Optionally, said method can also comprise before S11:
Play the video ads of setting, when starting electronic signature, video ads stops automatically.Can carry out effective advertising like this.
In addition, need to prove, above-mentioned verification msg can also comprise: ID and numerical value b.Above-mentioned numerical value b can be the distance (that is induction degree) of magnetic pen with respect to the electromagnetism screen, and this distance is used for judging whether magnetic pen is in writes word state.Like this each point is recorded, be saved in the middle of a document.During near electromagnetic screen, value can slowly become large when magnetic pen, upward judges whether magnetic pen is in being specifically as follows of the method for writing word state: set a distance threshold, as the b value greater than distance threshold, determine to be in and write word state, as less than distance threshold, be in the non-word state of writing.
The specific embodiment of the invention also provides signature verification device on a kind of line, and this installs as shown in Figure 2, comprising:
Unit for scaling 23 is for the quantity of counting of the described user's signature on desired location being zoomed to the designated value of set point number;
Comparing unit 26 is used for comparing the verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature; And obtain the similarity of user's signature and original signature according to comparison result; Less than threshold value, be defined as same signature as similarity, more than or equal to threshold value, be defined as non-same signature as similarity.
Optionally, comparing unit 26 is concrete is used for calculating the absolute value of difference of verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature, all absolute value additions are namely obtained both difference condition, this absolute value and be similarity.
Optionally, unit for scaling 23 specifically is used for
Counting greater than setting value of user's signature as described, the setting designated value of sampling out from the counting of this user's signature is counted;
Counting less than setting value of user's signature, count so that described user's signature reaches designated value from the interior insertion of counting of user's signature as described.
Optionally, computing unit 25 specifically is used for
User's signature counts greater than designated value as described, obtain each verification msg of counting in counting after sampling, wherein x axial coordinate value in this verification msg, y axial coordinate value are for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature;
Counting less than designated value of user's signature as described obtained the verification msg of counting of user's signature, and wherein, wherein x axial coordinate value in this verification msg, y axial coordinate value are for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature; Obtain and insert the verification msg of counting, wherein x axial coordinate value, x axial coordinate value, the y axial coordinate value of y axial coordinate value for counting from the described insertion of setting the user's signature of size in this insertion verification msg of counting; In the verification msg that this insertion is counted, force value is the mean value of the described insertion adjacent two point pressure values of counting, and in the verification msg that this insertion is counted, time value is the mean value of described insertion adjacent 2 time values of counting.
Optionally, device is specially: computing machine, panel computer, mobile terminal or signature verification platform.
The above is only preferred embodiment of the present invention, not in order to limiting the present invention, all any modifications of doing within the spirit and principles in the present invention, is equal to and replaces and improvement etc., within all should being included in protection scope of the present invention.
Claims (9)
1. the various dimensions verification method of signing on a line is characterized in that, described method comprises:
When user's signature, extract the verification msg of described user's signature; Described verification msg comprises: the x axial coordinate of signature, y axial coordinate, force value and time value.
Described user's signature is displaced to desired location, and the quantity of counting of the described user's signature on desired location is zoomed to the designated value of set point number;
The scope square size that will have the user's signature of the designated value of setting up an office is carried out normalized to the size of setting;
After completing above-mentioned setting, obtain the verification msg of the point of completing the user's signature after setting according to described verification msg;
Compare the verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature, and obtain the similarity of user's signature and original signature according to comparison result; Less than threshold value, be defined as same signature as similarity, more than or equal to threshold value, be defined as non-same signature as similarity.
2. method according to claim 1, it is characterized in that, described comparison is described completes the verification msg of each point of the verification msg of each point of the user's signature after setting and original signature, and obtains the similarity of user's signature and original signature according to comparison result, specifically comprises:
Calculate the absolute value of difference of verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature, all absolute value additions namely obtained both difference condition, this absolute value and be similarity.
3. method according to claim 1, is characterized in that, the designated value that the quantity of counting of the described user's signature on desired location is zoomed to set point number comprises:
The quantity of counting of user's signature is greater than setting value as described, and the designated value of the set point number of sampling out from the counting of this user's signature is counted;
The quantity of counting of user's signature is less than set-point value as described, inserts in from the counting of user's signature and counts so that counting of described user's signature reaches the designated value of set point number.
4. method according to claim 3, is characterized in that, the verification msg that the described verification msg of described foundation is obtained the point of completing the user's signature after setting comprises:
The quantity of counting of user's signature is greater than described designated value as described, obtain each verification msg of counting in counting after sampling, wherein x axial coordinate value in this verification msg, y axial coordinate value are for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature;
The quantity of counting of user's signature is less than described designated value as described, obtain the verification msg of counting of user's signature, wherein, wherein x axial coordinate value in this verification msg, y axial coordinate value for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature; Obtain and insert the verification msg of counting, wherein x axial coordinate value, x axial coordinate value, the y axial coordinate value of y axial coordinate value for counting from the described insertion of setting the user's signature of size in this insertion verification msg of counting; In the verification msg that this insertion is counted, force value is the mean value of the described insertion adjacent two point pressure values of counting, and in the verification msg that this insertion is counted, time value is the mean value of described insertion adjacent 2 time values of counting.
5. signature verification device on a line, is characterized in that, described device comprises:
Extraction unit is used for extracting the verification msg of described user's signature when user's signature; Described verification msg comprises: the x axial coordinate of signature, y axial coordinate, force value and time value.
Displacement unit is used for described user's signature is displaced to desired location;
Unit for scaling is for the quantity of counting of the described user's signature on desired location being zoomed to the designated value of set point number;
The size setup unit carries out normalized to the size of setting for the scope square size of the user's signature that will have the designated value of setting up an office;
Computing unit is used for after completing above-mentioned setting, obtains the verification msg of the point of completing the user's signature after setting according to described verification msg;
Comparing unit is used for comparing the verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature; And obtain the similarity of user's signature and original signature according to comparison result; Less than threshold value, be defined as same signature as similarity, more than or equal to threshold value, be defined as non-same signature as similarity.
6. device according to claim 5, it is characterized in that, described comparing unit specifically is used for calculating the absolute value of difference of verification msg of each point of the verification msg of described each point of completing the user's signature after setting and original signature, all absolute value additions are namely obtained both difference condition, this absolute value and be similarity.
7. device according to claim 5, is characterized in that, described unit for scaling specifically is used for
The quantity of counting of user's signature is greater than setting value as described, and the described designated value of sampling out from the counting of this user's signature is counted;
The quantity of counting of user's signature is less than set-point value as described, inserts in from the counting of user's signature and counts so that counting of described user's signature reaches described designated value.
8. device according to claim 7, is characterized in that, described computing unit specifically is used for
The quantity of counting of user's signature is greater than described designated value as described, obtain each verification msg of counting in counting after sampling, wherein x axial coordinate value in this verification msg, y axial coordinate value are for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature;
The quantity of counting of user's signature is less than described designated value as described, obtain the verification msg of counting of user's signature, wherein, wherein x axial coordinate value in this verification msg, y axial coordinate value for from the described x axial coordinate value of counting, the y axial coordinate value of setting the user's signature of size; In this verification msg, force value and time value are force value and the time value in the verification msg of described user's signature; Obtain and insert the verification msg of counting, wherein x axial coordinate value, x axial coordinate value, the y axial coordinate value of y axial coordinate value for counting from the described insertion of setting the user's signature of size in this insertion verification msg of counting; In the verification msg that this insertion is counted, force value is the mean value of the described insertion adjacent two point pressure values of counting, and in the verification msg that this insertion is counted, time value is the mean value of described insertion adjacent 2 time values of counting.
9. according to claim 5-8 arbitrary described devices, is characterized in that, described device is specially:
Computing machine, panel computer, mobile terminal or signature verification platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100461041A CN103136546A (en) | 2013-02-05 | 2013-02-05 | Multi-dimension authentication method and authentication device of on-line signature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100461041A CN103136546A (en) | 2013-02-05 | 2013-02-05 | Multi-dimension authentication method and authentication device of on-line signature |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103136546A true CN103136546A (en) | 2013-06-05 |
Family
ID=48496354
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100461041A Pending CN103136546A (en) | 2013-02-05 | 2013-02-05 | Multi-dimension authentication method and authentication device of on-line signature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103136546A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927471A (en) * | 2014-04-18 | 2014-07-16 | 电子科技大学 | Authentication method and device |
| CN105577374A (en) * | 2014-10-08 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Verification method and verification device |
| CN109806591A (en) * | 2019-01-16 | 2019-05-28 | 网易（杭州）网络有限公司 | Signature generating method and device in scene of game |
| CN110083301A (en) * | 2019-04-29 | 2019-08-02 | 珠海纳金科技有限公司 | A kind of handwritten signature unlocking system and method |
| 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 |
| TWI756592B (en) * | 2019-12-02 | 2022-03-01 | 玉山商業銀行股份有限公司 | Identity verification system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 reversible handwriting signature |
| CN102479331A (en) * | 2010-11-22 | 2012-05-30 | 宏碁股份有限公司 | Method for comparing handwriting path to sample path |
-
2013
- 2013-02-05 CN CN2013100461041A patent/CN103136546A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 reversible handwriting signature |
| CN102479331A (en) * | 2010-11-22 | 2012-05-30 | 宏碁股份有限公司 | Method for comparing handwriting path to sample path |
Non-Patent Citations (1)
| Title |
|---|
| 张大海: "基于多维力信息的在线签名认证方法研究", 《中国科学技术大学博士论文》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927471A (en) * | 2014-04-18 | 2014-07-16 | 电子科技大学 | Authentication method and device |
| CN105577374A (en) * | 2014-10-08 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Verification method and verification device |
| CN105577374B (en) * | 2014-10-08 | 2019-06-21 | 阿里巴巴集团控股有限公司 | A kind of verification method and equipment |
| CN109806591A (en) * | 2019-01-16 | 2019-05-28 | 网易（杭州）网络有限公司 | Signature generating method and device in scene of game |
| CN110083301A (en) * | 2019-04-29 | 2019-08-02 | 珠海纳金科技有限公司 | A kind of handwritten signature unlocking system and method |
| TWI756592B (en) * | 2019-12-02 | 2022-03-01 | 玉山商業銀行股份有限公司 | Identity verification system |
| 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 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107563385B (en) | License plate character recognition method based on depth convolution production confrontation network | |
| Coetzer et al. | Offline signature verification using the discrete radon transform and a hidden Markov model | |
| Kashi et al. | A Hidden Markov Model approach to online handwritten signature verification | |
| CN103136546A (en) | Multi-dimension authentication method and authentication device of on-line signature | |
| CN106156766A (en) | The generation method and device of line of text grader | |
| CN108564040B (en) | Fingerprint activity detection method based on deep convolution characteristics | |
| CN103279770B (en) | Based on the person's handwriting recognition methods of stroke fragment and contour feature | |
| CN103839041A (en) | Client-side feature identification method and device | |
| CN102332084A (en) | Identity identification method based on palm print and human face feature extraction | |
| US11416592B2 (en) | Method for online signature verification using wrist-worn devices | |
| CN109255339B (en) | Classification method based on self-adaptive deep forest human gait energy map | |
| CN103761515A (en) | Human face feature extracting method and device based on LBP | |
| CN103714331A (en) | Facial expression feature extraction method based on point distribution model | |
| Chaabouni et al. | Static and dynamic features for writer identification based on multi-fractals. | |
| CN102592142A (en) | Computer-system-based handwritten signature stability evaluation method | |
| CN114220178A (en) | Signature identification system and method based on channel attention mechanism | |
| Thakare et al. | A combined feature extraction model using SIFT and LBP for offline signature verification system | |
| CN106295478A (en) | A kind of image characteristic extracting method and device | |
| Kumar et al. | FPR using machine learning with multi‐feature method | |
| Miroslav et al. | Basic on-line handwritten signature features for personal biometric authentication | |
| Piekarczyk et al. | Hierarchical Graph-Grammar Model for Secure and Efficient Handwritten Signatures Classification. | |
| Bansal et al. | Multimodal biometrics by fusion for security using genetic algorithm | |
| Yamazaki et al. | A study on vulnerability in on-line writer verification system | |
| Alqudah et al. | Shift and scale invariant recognition of printed numerals | |
| CN111933154A (en) | Method and device for identifying counterfeit voice and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130605 |