CN100492395C - Fingerprint characteristic quickly matching method, device and application - Google Patents

Abstract

A quick matching method of fingerprint utilizes global character information and local detailed character information of fingerprint to accurately and conveniently as well as quickly judge out whether two fingerprints are matched to each other or not. The device and intelligent card for realizing said method are also disclosed.

Description

Fingerprint characteristic quickly matching method, device and application thereof

Technical field

The present invention relates to a kind of global characteristics information (being the central point of fingerprint image) of fingerprint and local minutia information (being the bifurcation and the end points of fingerprint ridge line) utilized and realize the fingerprint characteristic method of coupling fast, also relate to the device that is used to realize this fingerprint characteristic quickly matching method, also relate to a kind of smart card (smartcard) that uses this fingerprint characteristic rapid matching apparatus to carry out the fingerprint characteristic coupling, belong to pattern-recognition and technical field of intelligent card.

Background technology

Smart card techniques is acknowledged as the effective technology means that business and government solves identification and safety problem.

Current, smart card techniques has developed into and has utilized biological characteristic to carry out the stage of identification.With PKI structure and Java ^TMOutside the technology combination, smart card techniques combines with biometrics identification technology again, can realize the strongst a kind of authentication mode.Therefore, as portable multi-function identification and telecommunication media, adopt the smart card of biometrics identification technology to be acknowledged as the preferred plan of carrying out reliable personal identification with online and offline mode with highest security.Here said biometrics identification technology is meant collection, analyzes the specific characteristic of human body, as fingerprint, retina and acoustic pattern etc., the technology of carrying out authentication.

In biometrics identification technology, fingerprint identification technology develops the earliest, and people just bring into use computing machine to handle fingerprint in nineteen sixties.Through constantly development, fingerprint recognition has been the technology of comparative maturity now, has a lot of relevant technological achievements to come out.For example publication number is that the application for a patent for invention of CN1735908 discloses a kind of fingerprint matching device and method, recording medium and can use low volume data to carry out the program that fingerprint matching is handled fast.Central point C, the bifurcation P1 that CPU detects fingerprint image is to P8 and terminal point Q1 to Q10 and by being connected and near the immediate bifurcation P1 of central point C and bifurcation are two bifurcation P2 and P3 generation triangle W1.The length on leg-of-mutton area S1 and three limits is stored in the flash memory.In addition, the position of terminal point Q3 position, bifurcation P2 and the bifurcation P2 of near CPU calculating bifurcation P1 and the bifurcation P1 terminal point Q1 near and the position of the terminal point Q3 bifurcation P3 and the bifurcation P3 near, and in flash memory, store these positions as registered template.When carrying out coupling, CPU judge leg-of-mutton area and the length of side and the position of the terminal point that obtains from matching image whether consistent with registered template.

Publication number is that the application for a patent for invention of CN1716276 also discloses a kind of algorithm for recognizing fingerprint and system, it is characterized in that: the input method of relevant fingerprint characteristic information and the fingerprint characteristic information of importing by above-mentioned input method of logining, the relevant information of the fingerprint characteristic of importing by above-mentioned input method and the accumulating method of finger print information; Calculate the similar degree of two kinds of fingerprints from the relevant information of two kinds of fingerprint characteristics being imported; The summary value of using the computing method of similar degree to calculate the existing difference in position of the login fingerprint of being remembered by the fingerprint accumulating method is called the primary importance computing method, after the difference of the position of having revised the login fingerprint that calculates by above-mentioned primary importance computing method and the position of input fingerprint, the summary value of using above-mentioned similar level calculating method to calculate the rotation difference of login fingerprint and input fingerprint is called the first rotation computing method, the difference of the position of login fingerprint and input fingerprint, and the summary value of rotation difference; Use above-mentioned primary importance with and the difference of the login fingerprint that calculates of tropometer calculation method and input fingerprint positions, and the summary value of rotation difference, and the computing method of above-mentioned similar degree, the difference of the position of the login fingerprint that above-mentioned fingerprint accumulating method is remembered and input fingerprint positions compares, its summary value is determined in small range, at length compare, be called second place computing method.Utilization said second position computing method are calculated the difference and the above-mentioned similar level calculating method of login fingerprint positions and input fingerprint positions, the difference of rotation that utilization fingerprint accumulating method is logined and the rotation of input fingerprint is mated its summary value, compare with the above-mentioned first rotation computing method, it is determined within small range, carry out detailed comparison, be called the second position of rotation computing method; Utilization second place computing method, the second position of rotation computing method are calculated fingerprint positions of being logined and the position of importing fingerprint and the difference of position of rotation and are revised, the computing method compute classes of using above-mentioned similar degree again judges according to its result whether the fingerprint of login and the fingerprint of being imported are same people like degree; Above-mentioned judged result is exported.

From the technical standpoint analysis, fingerprint recognition is typical pattern recognition problem, mainly is made up of the feature extraction of fingerprint and characteristic matching two large divisions.At present, fingerprint recognition alignment algorithm great majority adopt the details comparison method, mainly are divided into two big classes: central point alignment algorithm and no central point alignment algorithm are arranged.No central point alignment algorithm is some the redundancy feature auxiliary matched by details, does not need central point in the matching process.Its advantage is not high to the positioning requirements of the fingerprint image of gathering, but comparison efficiency is low, and incompatible higher to time requirement occasion more is not suitable for carrying out in all limited smart card of computational resource and computing power.The central point alignment algorithm is arranged is by central point auxiliary positioning fingerprint image and mate, its advantage is that the speed of comparison is very fast, but shortcoming be accuracy requirement to the center point location than higher, if the central point locating accuracy is not accurate enough, usually cause the comparison failure.

In smart card, use the fingerprint characteristic recognition technology, must fully take into account the characteristics of smart card itself.Compare with computing machine, the limited storage space in the smart card, so the characteristic of fingerprint can not be too many; In addition, the arithmetic capability of smart card inner treater is often limited, so the fingerprint characteristic matching algorithm must be simple and efficient, can judge fast whether fingerprint mates.But existing fingerprint matching algorithm is often ignored the restriction that the These characteristics of smart card is brought, and therefore the good algorithm of result of use in computing machine just may not necessarily not obtain gratifying effect in smart card.At present, just known to the applicant, the fingerprint matching algorithm that is optimized at the characteristics of smart card is still rare specially.

Summary of the invention

First purpose of the present invention provides a kind of at smart card computing and storage capacity features of limited and the fingerprint characteristic quickly matching method that is optimized specially.This method is utilized the global characteristics and the minutia of fingerprint, can carry out the fingerprint characteristic coupling quickly and accurately.

Second purpose of the present invention provides a kind of device that is used to realize this fingerprint characteristic quickly matching method.

The 3rd purpose of the present invention provides a kind of smart card that uses above-mentioned fingerprint characteristic rapid matching apparatus to carry out the fingerprint characteristic coupling.

For realizing above-mentioned goal of the invention, the present invention adopts following technical scheme:

A kind of fingerprint characteristic quickly matching method mates calculating under polar coordinate system, it is characterized in that comprising following steps:

(1) central point of detection fingerprint image;

(2) if find central point, then the central point with fingerprint image is to carry out the minutia coupling with reference to corresponding point; If can not find central point, then change step (9) over to;

(3) whether coupling successful?

(4) if the match is successful then to be judged as be same fingerprint, if unsuccessful then be reference center with described central point, it is right that searching can be used as the details of match reference corresponding point in template and input feature vector point set, and they are noted;

(5) with the details noted to as with reference to corresponding point, the minutia of fingerprint is mated;

(6) whether coupling successful?

(7) if the match is successful then to be judged as be same fingerprint, if unsuccessful then change step (9) over to;

(9) with untapped template and input feature vector point in the abovementioned steps to as with reference to corresponding point, carry out the minutiae point coupling;

(10) whether coupling successful?

(11) if the match is successful then to be judged as be same fingerprint, if unsuccessful then be judged as non-same fingerprint.

Wherein, in the described step (1), described central point W is as follows determined:

W1) will import fingerprint image be divided into the size be the piece of W * W, wherein, W is an integer;

W2) calculate the gradient G of each pixel in each piece _xAnd G _y

W3) calculate the local principal direction of each piece;

W4) field of direction is transformed in the continuous vector field, by low-pass filter correction crestal line direction;

W5) field of direction of calculated fingerprint image;

W6) calculate certain any Poincare index, if the Poincare index value of this point is 1/2, then this is a central point.

When candidate's central point is several, obtain final center position by k-means clustering algorithm.

Behind the block of definite central point place, this block further is divided into littler block, and uses the position of recomputating central point as method as described in the step W.

Described minutia coupling was divided into for two steps and carries out, at first be to utilize the local neighborhood structural information of a pair of reference point in the unique point set to carry out partial structurtes information slightly to mate, after thick coupling is passed through, carry out accurately coupling of the overall situation as corresponding point with this reference point.

Described minutia matching process is divided into following steps:

1) with template minutiae point and input minutiae point as the reference minutiae point, concentrated neighborhood minutiae point or the overall minutiae point of input point set and template point transformed to polar coordinate system;

2) template minutiae point in the polar coordinates and input minutiae point are increased progressively the direction ordering by polar angle, and connect bunchiness;

3) mate described string with self-adaptation gauge cassette method, find out and write down the coupling mark, with the size of coupling mark as whether carrying out accurate or the match is successful the whether foundation of the overall situation;

4) find out maximal value in each time coupling mark, it is used as the coupling mark of input details point set and template details point set, if the coupling mark is higher than a pre-set threshold, then thinks input picture and template image from same fingerprint, otherwise think that they are from different fingerprints.

The size of described self-adaptation gauge box is represented with radius_size and angle_size, uses as shown in the formula calculating radius_size and the angle_size that polar radius is the template minutiae point of r:

$\mathrm{radius}\_\mathrm{size}=\left\{\begin{array}{ccc}r\_\mathrm{small}& \mathrm{if}& r\_\mathrm{size}\≤r\_\mathrm{small}\\ r\_\mathrm{size}& \mathrm{if}& r\_\mathrm{small}\≤r\_\mathrm{size}\≤r\_\mathrm{large}\\ r\_\mathrm{large}& \mathrm{if}& r\_\mathrm{size}\≥r\_\mathrm{large}\end{array}\right.$

r_size＝α ₁+β ₁r

$\mathrm{angle}\_\mathrm{size}=\left\{\begin{array}{ccc}a\_\mathrm{small}& \mathrm{if}& a\_\mathrm{size}\≤a\_\mathrm{small}\\ a\_\mathrm{size}& \mathrm{if}& a\_\mathrm{small}\≤a\_\mathrm{size}\≤a\_\mathrm{large}\\ a\_\mathrm{large}& \mathrm{if}& a\_\mathrm{size}\≥a\_\mathrm{large}\end{array}\right.$

a_size＝α ₂-β ₂r

α wherein ₁, β ₁, α ₂, β ₂Be predefined empirical parameter, and all greater than zero, r is the polar radius of template minutiae point, r_small, r_large, a_small, a_large are respectively the upper bound and the lower bounds of radius_size and angle_size, and they are values of setting by test in advance.

Described coupling mark Ms obtains by following formula:

$\mathrm{Ms}=100\×\sqrt{\frac{\mathrm{nb}\_\mathrm{pair}*\mathrm{nb}\_\mathrm{pair}}{M_c*N_c}}+\mathrm{\α}*\mathrm{nb}\_\mathrm{pair}-\mathrm{\β}*\mathrm{match}\_\mathrm{error}$

Wherein nb_pair is the right number of minutiae point on the coupling, and match_error is the right cumulative matches error of each match point, M _c, N _cBe respectively template and input fingerprint image unique point number in the public domain, α, β is predetermined weighting coefficient.

Wherein, the value of described α is 2, and the value of β is 3.

A kind of fingerprint characteristic rapid matching apparatus is characterized in that comprising:

The central point judging unit is used for seeking and judging the central point of fingerprint;

Coupling corresponding point selected cell, it is right to be used for seeking the possible template and the input feature vector point that mate corresponding point of can be used as at template and input feature vector point, and they are write down;

The matching judgment unit is used to carry out the coupling calculating of unique point;

The polar coordinates converting unit is used for the locus of features relevant point and central point is converted to polar coordinates;

,, be corresponding point then at first, mate calculating, draw matching result by the matching judgment unit with this central point if there is central point by the central point in searching of central point judging unit and the judgement fingerprint; If central point fails to find, then other templates do not considered and input feature vector point are mated as the reference corresponding point two pieces of fingerprints being carried out minutiae point by coupling corresponding point selected cell and matching judgment unit, draw matching result; If it is not match that central point mates result calculated, be reference with described two central points then by coupling corresponding point selected cell, it is right to seek the possible template and the input feature vector point that can be used as the coupling corresponding point in template and input feature vector point, with these unique points to reference corresponding point as matching algorithm, by the matching judgment unit template and input feature vector point are carried out Feature Points Matching, draw matching result.

Described matching judgment unit mates before the calculating, by the polar coordinates converting unit locus about central point and unique point is converted to polar coordinates.

A kind of smart card has microprocessor, storer, telecommunication circuit, it is characterized in that:

Also has above-mentioned fingerprint characteristic rapid matching apparatus in the described smart card.

A kind of method of using above-mentioned smart card to carry out authentication, the user imports its fingerprint by the fingerprint collecting equipment on the card reader, after fingerprint sensor collects finger print data, submits to characteristic extracting module, and fingerprint characteristic is delivered to described smart card; Described smart card mates the fingerprint characteristic of input and the fingerprint characteristic of preservation, if the match is successful, then finishes authentication, it is characterized in that:

Described smart card adopts above-mentioned fingerprint characteristic quickly matching method to carry out the fingerprint characteristic coupling.

Fingerprint characteristic quickly matching method provided by the present invention has following advantage:

(1) carries out under the guidance of central point information owing to the feature comparison process, thereby can determine two pieces of rotation and translation transformation relations between the fingerprint image apace, therefore overcome the low shortcoming of no central point matching algorithm efficiency;

(2) utilize specified reference point when the minutia of two pieces of fingerprints is compared, adopting and carried out local neighborhood earlier and slightly mate, carrying out the accurately method of coupling of the overall situation again, accelerating the speed of fingerprint comparison greatly;

(3) owing in the calculating of coupling mark, considered of the influence of details matching error to matching result, can more accurate matching result thereby obtain.

Adopt the smart card of above-mentioned fingerprint characteristic quickly matching method can further improve its safe reliability and ease of use, thereby provide a kind of feasible approach for the fusion of biometrics identification technology and smart card application technologies.

Description of drawings

The present invention is further illustrated below in conjunction with the drawings and specific embodiments.

Fig. 1 is the discrete representation synoptic diagram of closed curve.

What Fig. 2 showed is the form of asking the singular point direction.

Fig. 3 is the local feature vectors synoptic diagram.

Fig. 4 has shown an example of the gauge box of variable-size.

Fig. 5 is the basic flow sheet of fingerprint characteristic quickly matching method provided by the present invention.

Fig. 6 is the building-block of logic of a typical smart card.

The smart card that Fig. 7 has introduced built-in fingerprint feature rapid matching apparatus carries out the canonical process that fingerprint characteristic mates.

Embodiment

For ease of understanding, at first concentrate embodiment and the device for carrying out said of introducing fingerprint characteristic quickly matching method provided by the present invention below, specifically introduce the application of this method in smart card again.

This fingerprint characteristic quickly matching method mainly comprises the technology contents of following several respects: Core Point in Fingerprint detects, the fingerprint image registration, details based on partial structurtes information is slightly mated, details based on global structure information is accurately mated, based on the characteristic matching and the coupling fractional computation of singular point information.Below it is introduced one by one.

1. Core Point in Fingerprint detects

Central points in order to obtain the field of direction of fingerprint image accurately, adopts following steps based on the field of direction:

1) will import fingerprint image and be divided into the big or small piece of W * W that is.Wherein, W is an integer;

2) calculate the gradient G of each pixel in each piece _xAnd G _y

3) calculate the local principal direction of each piece:

$\mathrm{\θ}(i,j)=\frac{1}{2}{\tan }^{-1}\left(\frac{\underset{u=i-W/2}{\overset{i+W/2}{\mathrm{\Σ}}}\underset{v=j-W/2}{\overset{j+W/2}{\mathrm{\Σ}}}2G_x(u,v)G_y(u,v)}{\underset{u=i-W/2}{\overset{i+W/2}{\mathrm{\Σ}}}\underset{v=j-W/2}{\overset{j+W/2}{\mathrm{\Σ}}}(G_x^2(u,v)-G_y^2(u,v))}\right)---\left(1\right)$

G wherein _xAnd G _yBe respectively the gradient on x and the y direction, W is the width that is used for estimating the piece of the field of direction, and (i j) is point (i, j) principal direction of place piece to θ.

4) owing to noise, the crestal line of fracture and the existence of valley line, (i may not be total correct j) to the crestal line direction θ of estimation.In not having unusual neighborhood of a point, local crestal line direction is slowly to change, and can revise incorrect crestal line direction with a low-pass filter.In order to do this part thing, the field of direction need be transformed in the continuous vector field:

φ _x(i，j)＝cos(2θ(，j)) (2)

φ _y(i，j)＝sin(2θ(，j)) (3)

φ _x, φ _yBe the x of vector field, the y component, low-pass filtering can followingly be represented:

${\mathrm{\φ}}_x^{\′}(i,j)=\underset{u=-w_{\mathrm{\φ}}/2}{\overset{w_{\mathrm{\φ}}/2}{\mathrm{\Σ}}}\underset{v=-w_{\mathrm{\φ}}/2}{\overset{w_{\mathrm{\φ}}}{\mathrm{\Σ}}}h(u,v){\mathrm{\φ}}_x(i-u{w}_{\mathrm{\φ}},j-v{w}_{\mathrm{\φ}})---\left(4\right)$

${\mathrm{\φ}}_y^{\′}(i,j)=\underset{u=-w_{\mathrm{\φ}}/2}{\overset{w_{\mathrm{\φ}}/2}{\mathrm{\Σ}}}\underset{v=-w_{\mathrm{\φ}}/2}{\overset{w_{\mathrm{\φ}}}{\mathrm{\Σ}}}h(u,v){\mathrm{\φ}}_y(i-u{w}_{\mathrm{\φ}},j-v{w}_{\mathrm{\φ}})---\left(5\right)$

Wherein h is a two-dimensional low-pass filter, and its integration is 1, w _φ* w _φIt is the size of wave filter.This smooth operation is carried out on piece.Its default size is 5 * 5.

5) calculate (i, the local direction field of j) locating:

$O(i,j)=\frac{1}{2}{\tan }^{-1}\left(\frac{{\mathrm{\φ}}_y^{\′}(i,j)}{{\mathrm{\φ}}_x^{\′}(i,j)}\right)---\left(6\right)$

We have just obtained the field of direction of fingerprint image like this.

6) calculating of singular point

If O is the field of direction of fingerprint image, point (i, the following calculating of the Poincare index of j) locating:

$\mathrm{Poincare}(i,j)=\frac{1}{2\mathrm{\π}}\underset{k=0}{\overset{N_{\mathrm{\ψ}}}{\mathrm{\Σ}}}\mathrm{\Δ}\left(k\right)---\left(7\right)$

$\mathrm{\&Delta;}\left(k\right)=\left\{\begin{array}{cc}\mathrm{\&delta;}\left(k\right),& \mathrm{if}\left|\mathrm{\&delta;}\left(k\right)\right|<\frac{\mathrm{\&pi;}}{2}\\ \mathrm{\&pi;}+\mathrm{\&delta;}\left(k\right),& \mathrm{if\; \&delta;}\left(k\right)\&le;-\frac{\mathrm{\&pi;}}{2}\\ \mathrm{\&pi;}-\mathrm{\&delta;}\left(k\right),& \mathrm{otherwise}\end{array}\right.---\left(8\right)$

$\mathrm{\&delta;}\left(k\right)=O({\mathrm{\&psi;}}_x(i\&prime;),{\mathrm{\&psi;}}_y(i\&prime;))-O({\mathrm{\&psi;}}_x\left(i\right),{\mathrm{\&psi;}}_y\left(i\right))---\left(9\right)$

$i\&prime;=(i+1)\mathrm{mod}{N}_{\mathrm{\&psi;}}---\left(10\right)$

ψ wherein _x(i) and ψ _y(i) being respectively is the N that has at center with the set point _ψOn the closed curve of individual pixel i the point x and y coordinate. if the Poincare index value is 1/2, so this set point (i j) just is confirmed as central point.

In the present invention, closed curve is to get in one 5 * 5 grid.As shown in Figure 1, then in curve D ₁D ₂... D ₁₂D ₁The value that goes up its Poincare index is as follows:

$\mathrm{Poincare}(i,j)=\underset{i=1}{\overset{12}{\mathrm{\&Sigma;}}}|D_i-D_{(i+1)\mathrm{mod}12}|---\left(11\right)$

Obtain adjacent several candidate's central points like this, can obtain final center position with k-means clustering algorithm.

In some fingerprint images, because may there be pseudo-central point in the influence of factors such as noise.In order to eliminate pseudo-central point, also calculated the inside curve d of same point ₁d ₂... d ₈d ₁The Poincare value.Have only when 3 * 3 identical with 5 * 5 Poincare value (1/2) in, this candidate's central point ability is as real central point.

To each central point, determine that its direction is as follows: in 5 * 5 the field of direction, calculating eight directions (is 0 among Fig. 2,1 ... 7) difference each other of the direction on and, minimum direction is the direction of central point just, as shown in Figure 2.In computation process, may have both direction be the same, this moment can get the direction of the mean value of this both direction as central point.

The central point of fingerprint image for the coupling of fingerprint details provides can be for the information of utilizing, but this degree of accuracy with the central point detection is relevant.The central point detection method that provides above, the center position that can only be based on piecemeal that obtains, if will utilize the corresponding point of these center position as two pieces of fingerprints, precision obviously is not enough. in order to obtain more accurate center position, the accurate position fixing process that also needs a center position promptly recomputates the position of central point with littler block size in the local neighborhood of above-mentioned position.Because this calculating is carried out in very little image range, so speed can be very fast.

2. fingerprint image registration

For template fingerprint to be matched and input fingerprint, just coordinate position, direction and the type information of detected minutiae point and central point from fingerprint image that we obtain.Owing to do not know the corresponding relation between this two width of cloth fingerprint in advance, at first to find suitable conversion that they are mapped, this process is exactly the registration of fingerprint.We use a pair of unique point, and (one comes from template fingerprint, one comes from the input fingerprint) as with reference to point, utilize coordinate translation and direction rotation relationship between them to construct similarity transformation, will import fingerprint feature point and transform in the template fingerprint characteristic point coordinates space.Because the non-linear deformation of fingerprint image often is radial, the deformation ratio in certain zone is bigger, non-linearly expand outwardly then, thereby, in polar coordinates, non-linear deformation can be described better.In addition, in polar coordinate system, we do not need to consider the translation between the reference point of input picture and template image, because the translation between input picture and template image is fixed, that is to say that the translation between the other a pair of corresponding point is the same with translation between the reference point, like this, when the coordinate of other a pair of corresponding point is converted to polar coordinates with respect to reference point, translation just has been cancelled, and, in polar coordinate system, obviously handle the rotation between two width of cloth images than in rectangular coordinate system, being more convenient for.Comprehensive above-mentioned reason, the present invention will carry out the details coupling in polar coordinate system.

Even input fingerprint and template fingerprint, still have picture translation, rotation, the deformation that dimensional variation is such from same finger between them.Before two width of cloth images are mated, to estimate the deformation parameter between them earlier, and this two width of cloth image be calibrated with this.Because two width of cloth fingerprint images are normally gathered with same instrument, can suppose that the dimensional variation coefficient between them is 1.In addition, in polar coordinates, can not consider translation between two width of cloth images.Thereby need make an estimate rotation parameter between input picture and template image only arranged.

Order

$P=({(x_1^p,y_1^p,{\mathrm{\&theta;}}_1^p)}^T,......,{(x_M^p,y_M^p,{\mathrm{\&theta;}}_M^p)}^T)---\left(12\right)$

M in a representation template fingerprint minutiae point,

$Q=({(x_1^q,y_1^p,{\mathrm{\&theta;}}_1^q)}^T,......,{(x_N^q,y_N^q,{\mathrm{\&theta;}}_N^q)}^T)---\left(13\right)$

N minutiae point in the expression input fingerprint.

Go for minutiae point is transformed in the polar coordinate system, we will and import at template details point set, and minutiae point is concentrated respectively selects a reference point as the initial point in the corresponding polar coordinate system, and calculates the polar coordinates of other minutiae point with respect to reference point.If it is all have central point in two pieces of fingerprints, then right with central point earlier as the reference point; If have one piece or two pieces of fingerprints all not have central point, then can't know the corresponding relation of template details point set and input details point set, will consider that all possible reference point is right this moment.

To the concentrated every bit P of template point _i(the every bit Q that 1≤i≤M) and input point are concentrated _j(1≤j≤N),, then calculate Euclidean distance and P between them if they are minutiae point of same type _iDirection is with respect to Q _jThe anglec of rotation rotate[i of direction] [j].If the two satisfies certain threshold condition (be no more than 80 as distance, differential seat angle is no more than 45), then they are carried out ensuing thick coupling and accurate matching process as a pair of reference point.Otherwise it is right then to investigate other possible reference point.

Will be in polar coordinate system with input picture and template image calibration, only need other all input minutiae point and template minutiae point all respectively with respect to reference point P _iAnd Q _jBe transformed in the polar coordinate system, on the polar angle of all input minutiae point, add an angle rotate[i then] [j].That is to say, will import minutiae point and template minutiae point all respectively with respect to reference point P _iAnd Q _jBe transformed in the polar coordinate system with following formula

$\left(\begin{array}{c}{r}_i\\ e_i\\ {\mathrm{\&theta;}}_i\end{array}\right)=\left(\begin{array}{c}\sqrt{{(x_i-x^r)}^2+{(y_i-y^r)}^2}\\ {\tan }^{-1}\left(\frac{y_i-y^r}{x_i-x^r}\right)\\ {\mathrm{\&theta;}}_i-{\mathrm{\&theta;}}^r\end{array}\right)---\left(14\right)$

(x wherein _i, y _i, θ _t) ^TBe the coordinate of minutiae point to be converted, (x ^r, y ^r, θ ^r) ^TBe coordinate with reference to minutiae point, (r _i, e _i, θ _i) ^TBe the expression (r of minutiae point in polar coordinates _iThe expression polar radius, e _iThe expression polar angle, θ _iThe expression minutiae point direction and with reference to the angle between the minutiae point direction).Then, we are to the e of each input minutiae point _iAdd an angle rotate[i] [j].

3. slightly mate based on the details of partial structurtes information

For given a pair of reference point, utilize their local neighborhood structural informations in the unique point set to judge whether they can be used as real corresponding point and carry out accurately coupling of the overall situation earlier.Have only the reference point of having passed through the thick coupling of partial structurtes information right, just utilize them to carry out accurately coupling of the real overall situation, so just greatly reduce the calculated amount of minutiae point coupling, improved the precision of matching algorithm and the efficient of computing simultaneously as corresponding point.

For a certain unique point P _k=(x _k, y _k, φ _k) ^T, utilize and oneself set up proper vector from its other m nearest unique point (the general value of m is 6～9) and it, as partial structurtes information, this local feature is that rotation is translation invariant this proper vector.Fig. 3 has illustrated the situation of m=2, and its proper vector is described below:

Fl _k＝((d _ki，θ _ki，φ _ki) ^T，(d _kj，θ _kj，φ _kj) ^T) (15)

$d_{\mathrm{ki}}=\sqrt{{(x_k-x_i)}^2+{(y_k-y_i)}^2}$

Wherein

Wherein, d _KiBe two distances between the unique point, θ _KiBe unique point P _k, P _iLine direction and unique point P _kAngle between the direction, φ _KiIt is the angle of two unique point directions.Above computation process comes down to unique point P _kThe polar coordinate transform that carries out for initial point.

For any two fixed reference feature points, calculate their local feature vectors separately after, utilize following some matching algorithm to determine the match condition of neighborhood minutiae point.Surpass certain threshold value if the neighborhood matching details is counted, then utilize them to carry out accurately coupling of the overall situation, otherwise the reference point that continuation is investigated other is right as corresponding point.

4. accurately mate based on the minutiae point of global structure information

In a certain reference point to P _b(b=b1, b2) passed through the thick coupling of neighbour structure information after, utilize these two unique points as the corresponding reference point of further carrying out global registration, with the further feature point P except that nearest neighbor point on the image again _kWith respect to corresponding point P _bCarry out coordinate transform:

Fg _kRepresentation feature point P _kWith respect to corresponding point P _bCarry out the vector that obtains behind the polar coordinate transform.All Fg _kThe P that has calculated _bLocal feature vectors constitute the global characteristics vector together.Thereby utilize following minutiae point matching algorithm that the global characteristics vector is mated and obtain the global registration result.

5. the coupling of minutiae point

In the accurate matching process of thick coupling of above-mentioned partial structurtes information and global information, all relate to the matching problem of the minutiae feature vector in the polar coordinate system.Minutiae point matching algorithm among the present invention is as follows:

1) with template minutiae point P _iWith input minutiae point Q _jAs the reference minutiae point, the method that utilization is introduced above will import point set and concentrated neighborhood minutiae point or the overall minutiae point of template point transforms to polar coordinate system;

2) template minutiae point in the polar coordinates and input minutiae point are increased progressively the direction ordering by polar angle, and connect bunchiness, be expressed as follows:

M wherein, N is respectively the minutiae point number in template and the input feature value;

3) the self-adaptation gauge cassette method coupling string that will introduce with the back With

Find out and write down the coupling mark, with the size of coupling mark as whether carrying out accurate or the match is successful the whether foundation of the overall situation;

4) find out maximal value in each time coupling mark, it is used as the coupling mark of input details point set and template details point set.If the coupling mark is higher than a pre-set threshold, then thinks input picture and template image from same fingerprint, otherwise think that they are from different fingerprints.

Self-adaptation gauge box and size thereof as shown in Figure 4, a gauge box is a box that is placed on the template minutiae point.The size of gauge box represents that with radius_size and angle_size their value will change along with the polar radius size of minutiae point.Calculate radius_size and the angle that polar radius is the template minutiae point of r with following formula.

$\mathrm{radius}\_\mathrm{size}=\left\{\begin{array}{ccc}r\_\mathrm{small}& \mathrm{if}& r\_\mathrm{size}\&le;r\_\mathrm{small}\\ r\_\mathrm{size}& \mathrm{if}& r\_\mathrm{small}\&le;r\_\mathrm{size}\&le;r\_\mathrm{large}\\ r\_\mathrm{large}& \mathrm{if}& r\_\mathrm{size}\&GreaterEqual;r\_\mathrm{large}\end{array}\right.---\left(20\right)$

r_size＝α ₁+β ₁r (21)

$\mathrm{angle}\_\mathrm{size}=\left\{\begin{array}{ccc}a\_\mathrm{small}& \mathrm{if}& a\_\mathrm{size}\&le;a\_\mathrm{small}\\ a\_\mathrm{size}& \mathrm{if}& a\_\mathrm{small}\&le;a\_\mathrm{size}\&le;a\_\mathrm{large}\\ a\_\mathrm{large}& \mathrm{if}& a\_\mathrm{size}\&GreaterEqual;a\_\mathrm{large}\end{array}\right.---\left(22\right)$

a_size＝α ₂-β ₂r (23)

α wherein ₁, β ₁, α ₂, β ₂Be predefined empirical parameter, and all greater than zero.R is the polar radius of template minutiae point, and r_small, r_large, a_small, a_large are respectively the upper bound and the lower bounds of radius_size and angle_size, and their value also is to set by test in advance.

The gauge box that uses the gauge box of self-adaptation size rather than fixed size is in order to make algorithm to non-linear deformation robust more.Non-linear deformation is generally bigger in a specific zone, non-linearly expands outwardly then.When the utmost point of minutiae point footpath hour, little deformation just can cause the change of big polar angle, and the change of polar radius is less.So in this case the angle_size of gauge box should bigger radius_size then should be less.On the other hand, when the polar radius of minutiae point is big, the less change of polar angle will cause the minutiae point position than cataclysm, and the deformation of polar radius can be regarded this minutiae point and adding up with reference to the deformation of the All Ranges between minutiae point as.So in this case the angle_size of gauge box should smaller radius_size then should be bigger.

Coupling

With

Arthmetic statement as follows:

1) use (20)～(23) formula to determine the size of the gauge box of each template minutiae point.Put nb_pair[i] [j]=0.(nb_pair[i] when [j] expression is carried out characteristic matching with template minutiae point Pi and input minutiae point Qj as the reference minutiae point on the coupling minutiae point logarithm)

2) do following circulation:

While 1≤m≤M do

While 1≤n≤N do

If template minutiae point m and input minutiae point n satisfy condition1, then

nb_pair[i][j]＝nb_pair[i][j]+1；

End?if

Increase?n；

End?while

Increase?m；

End?while

In the said process, condition1 is defined as:

Wherein:

$\mathrm{\&Delta;r}=r_m^p-r_n^q---\left(25\right)$

$\mathrm{\&Delta;\&theta;}=\left\{\begin{array}{cc}a& \mathrm{if}(a=({\mathrm{\&theta;}}_m^p-{\mathrm{\&theta;}}_n^q+360)\mathrm{mod}360)<180\\ a-180& \end{array}\right.---\left(26\right)$

R_low[m], r_high[m], θ _ low[m] and, θ _ high[m] be respectively the lower limit and the upper limit that template minutiae point m gauge is closed polar radius and polar angle error.Condition1 regards template minutiae point m and input minutiae point n as match point right condition.Its implication is, input minutiae point n should be in the inside of the gauge box of template minutiae point m, and the direction difference of these two minutiae point should be less than ε (as ε=30).

6. the calculating of coupling mark

For with reference to after finishing above matching process, supposing that the right number of minutiae point on the coupling is nb_pair with two minutiae point (from template fingerprint, and from the input fingerprint), the right cumulative matches error of each match point is match_error.Then mate being calculated as follows of mark:

$\mathrm{Ms}=100\&times;\sqrt{\frac{\mathrm{nb}\_\mathrm{pair}*\mathrm{nb}\_\mathrm{pair}}{M_c*N_c}}+\mathrm{\&alpha;}*\mathrm{nb}\_\mathrm{pair}-\mathrm{\&beta;}*\mathrm{match}\_\mathrm{error}---\left(28\right)$

M wherein _c, N _cBe respectively template and the input fingerprint image unique point number in the public domain, match_error (comprises in formula (25)～(27) matching error for each coupling minutiae point

Weighted mean, α, β is predetermined weighting coefficient (its representative value is respectively 2.0 and 3.0).Above coupling mark and predefined matching threshold are compared, can make coupling whether judgement.

7. based on the characteristic matching of central point information

After the Core Point in Fingerprint detection, fingerprint characteristic information comprises two parts: a part is the local detail characteristic point information, is the bifurcation and the end points of fingerprint ridge line; Another part is the global characteristics information of fingerprint, is the central point of fingerprint.Therefore, the central point information of fingerprint has been arranged after, just can improve the operational efficiency of matching algorithm.Summary is got up, and the Feature Points Matching algorithm among the present invention carries out with following steps as shown in Figure 5:

1) if in template fingerprint and the input fingerprint central point is arranged all, then the central point with two pieces of fingerprints is that corresponding point are carried out Feature Points Matching, as described in Feature Points Matching algorithm wherein saves as the above-mentioned the 5th; Otherwise changeed for the 4th step;

2) if be that corresponding point are returned the coupling consistent results when carrying out Feature Points Matching with the central point, then the characteristic matching algorithm finishes; Otherwise, with above-mentioned two central points is reference, in template and input feature vector point, seek possible template that can be used as the coupling corresponding point and input feature vector point to (size that template minutiae point gauge is closed being amplified, again carrying out overall details coupling get final product), and they are recorded gather among the Point Pair;

3) utilize that each unique point point of obtaining carries out Feature Points Matching to the reference corresponding point as matching algorithm to template and input minutiae point in second step, run into the coupling consistent results and then finish the characteristic matching algorithm;

4) other original templates and input minutiae point are then finished the characteristic matching algorithm to as the reference corresponding point two pieces of fingerprints being carried out the minutiae point coupling, run into to mate consistent results.

Consider the response time of fingerprint comparison in the card, method provided by the present invention considered abundant reference corresponding point to and carry out the minutiae point coupling after, whether all will return no matter the match is successful.

Above-mentioned fingerprint characteristic quickly matching method mainly is at smart card memory storage and the limited concrete environment of arithmetic capability and special optimal design.Owing in matching process, make full use of the directive function of central point, two pieces of fingerprint images carried out rapid registering, thereby improved the time efficiency of algorithm greatly; Simultaneously, the minutiae point matching process that utilizes the self-adaptation gauge to close has effectively overcome the influence of the non-linear deformation of fingerprint image to the matching algorithm precision; For given a pair of reference point, carry out the thick coupling of local neighborhood structural information earlier, carry out the accurate coupling of overall detailed information again, thereby greatly reduce the calculated amount of minutiae point matching process, improved the precision of matching algorithm and the efficient of computing simultaneously.

The software that is used to implement this method can be made the fingerprint characteristic rapid matching apparatus of form of firmware, is integrated in the smart card.This fingerprint characteristic rapid matching apparatus comprises the central point judging unit, is used for seeking and judging the central point of fingerprint; Coupling corresponding point selected cell, it is right to be used for seeking the possible template and the input feature vector point that mate corresponding point of can be used as at template and input feature vector point, and they are write down; The matching judgment unit is used to carry out the coupling calculating of unique point, and the polar coordinates converting unit is used for the locus of features relevant point and central point is converted to polar coordinates.When carrying out the quick matching operation of fingerprint characteristic,,, be corresponding point then at first with this central point if there is central point by the central point in searching of central point judging unit and the judgement fingerprint, mate calculating by the matching judgment unit, draw matching result; If central point fails to find, then other templates do not considered and input feature vector point are mated as the reference corresponding point two pieces of fingerprints being carried out minutiae point by coupling corresponding point selected cell and matching judgment unit, draw matching result.If it is not match that central point mates result calculated, be reference with above-mentioned two central points then by coupling corresponding point selected cell, it is right to seek the possible template and the input feature vector point that can be used as the coupling corresponding point in template and input feature vector point, with these unique points to reference corresponding point as matching algorithm, by the matching judgment unit template and input feature vector point are carried out Feature Points Matching, draw matching result.The matching judgment unit mates before the calculating, by the polar coordinates converting unit locus about central point and unique point is converted to polar coordinates, and matching operation is based on that polar data carries out, and can reduce calculated amount like this, thereby obtain the result quickly.

Above-mentioned fingerprint characteristic rapid matching apparatus mainly is used in the smart card.Fig. 6 has introduced a kind of internal logic structure of typical contact type intelligent card.Use therein microprocessor chip is MPU, has logic control, management function, encrypting and decrypting function etc., and storer comprises ROM, RAM, EEPROM etc.RFC is a RF transmit-receive circuit, mainly solves the read-write communication and the power supply of card, and CAU is the cryptographic calculation coprocessor, and SL is a security logic.Certainly, to be used in contact intelligent card also be fully passable to said apparatus.

The storage space size of dissimilar smart cards does not wait from 512Byte to 16KB, and has different subregions to select.For the smart card of implementing the method for the invention, owing to need the storage fingerprint characteristic data, the capacity of smart card generally should be selected more than the 1KB.In the present invention, if the coordinate model of use characteristic point is only considered ridge tip and bifurcation two category features point, then the characteristic of one piece of fingerprint is no more than 256Byte after encrypting, and can be stored in the memory block of smart card fully.

For the smart card that application the method for the invention is carried out the quick coupling of fingerprint characteristic, above-mentioned fingerprint characteristic rapid matching apparatus can be built in the microprocessor of smart card, also can independently be provided with.The smart card that Fig. 7 has introduced built-in fingerprint feature rapid matching apparatus carries out the canonical process that fingerprint characteristic mates.The user uses the card reader that has fingerprint recognition, imports its fingerprint by the fingerprint collecting equipment on the card reader, after fingerprint sensor collects finger print data, submits to characteristic extracting module, extracts this fingerprint characteristic, and fingerprint characteristic is delivered to smart card; Smart card mates the fingerprint characteristic of input and the fingerprint characteristic of preservation, if success then allows the user to carry out follow-up operation, if coupling is unsuccessful, then points out the user to re-enter finger print data.The unsuccessful smart card that can be automatically locked of general continuous several times.

Such scheme still exists a security vulnerabilities to be, if the assailant has obtained user's finger print information by his approach of tool, so as long as after he gets access to user's smart card, can send to smart card to the finger print information that gets access in advance and finish fingerprint matching, verification process.The method that solves this security vulnerabilities has two kinds, and a kind of method is fingerprint sensor, and card reader and smart card are integrated in an equipment, makes smart card only accept the finger print data from this equipment, and it is higher to do cost like this.Another kind method is that card reader and fingerprint sensor are integrated, and smart card and card reader authenticate before swap data mutually, and data are carried out encrypted transmission, and the weak point of doing like this is the technical scheme complexity.

The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.

Claims (12)

1. a fingerprint characteristic quickly matching method mates calculating under polar coordinate system, it is characterized in that comprising following steps:

(1) central point of detection fingerprint image;

(2) if find central point, then the central point with fingerprint image is to carry out the minutia coupling with reference to corresponding point; If can not find central point, then change step (9) over to;

(3) whether coupling successful?

(4) if the match is successful then to be judged as be same fingerprint, if unsuccessful then be reference center with described central point, it is right that searching can be used as the details of match reference corresponding point in template and input feature vector point set, and they are noted;

(5) with the details noted to as with reference to corresponding point, the minutia of fingerprint is mated;

(6) whether coupling successful?

(7) if the match is successful then to be judged as be same fingerprint, if unsuccessful then change step (9) over to;

(9) with untapped template and input feature vector point in the abovementioned steps to as with reference to corresponding point, carry out the minutiae point coupling;

(10) whether coupling successful?

(11) if the match is successful then to be judged as be same fingerprint, if unsuccessful then be judged as non-same fingerprint.

2. fingerprint characteristic quickly matching method as claimed in claim 1 is characterized in that:

In the described step (1), described central point is determined as follows:

1) will import fingerprint image be divided into the size be the piece of W * W, wherein, W is an integer;

2) calculate the gradient G of each pixel in each piece _xAnd G _y

3) calculate the local principal direction of each piece;

4) field of direction is transformed in the continuous vector field, by low-pass filter correction crestal line direction;

5) field of direction of calculated fingerprint image;

6) calculate certain any Poincare index, if the Poincare index value of this point is 1/2, then this is a central point.

3. fingerprint characteristic quickly matching method as claimed in claim 2 is characterized in that:

When candidate's central point is several, obtain final center position by k-means clustering algorithm.

4. fingerprint characteristic quickly matching method as claimed in claim 2 is characterized in that:

Behind the block of definite central point place, this block further is divided into littler block, and uses the position of recomputating central point as method as described in the claim 2.

5. fingerprint characteristic quickly matching method as claimed in claim 1 is characterized in that:

Described minutia coupling was divided into for two steps and carries out, at first be to utilize the local neighborhood structural information of a pair of reference point in the unique point set to carry out partial structurtes information slightly to mate, after thick coupling is passed through, carry out accurately coupling of the overall situation as corresponding point with this reference point.

6. fingerprint characteristic quickly matching method as claimed in claim 5 is characterized in that:

Described minutia matching process is divided into following steps:

1) with template minutiae point and input minutiae point as the reference minutiae point, concentrated neighborhood minutiae point or the overall minutiae point of input point set and template point transformed to polar coordinate system;

2) template minutiae point in the polar coordinates and input minutiae point are increased progressively the direction ordering by polar angle, and connect bunchiness;

3) mate described string with self-adaptation gauge cassette method, find out and write down the coupling mark, with the size of coupling mark as whether carrying out accurate or the match is successful the whether foundation of the overall situation;

4) find out maximal value in each time coupling mark, it is used as the coupling mark of input details point set and template details point set, if the coupling mark is higher than a pre-set threshold, then thinks input picture and template image from same fingerprint, otherwise think that they are from different fingerprints.

7. fingerprint characteristic quickly matching method as claimed in claim 6 is characterized in that:

The size of described self-adaptation gauge box is represented with radius_size and angle_size, uses as shown in the formula calculating radius_size and the angle_size that polar radius is the template minutiae point of r:

$\mathrm{radius}\_\mathrm{size}=\left\{\begin{array}{cc}r\_\mathrm{small}& \mathrm{if\; r}\_\mathrm{size}\&le;r\_\mathrm{small}\\ r\_\mathrm{size}& \mathrm{if\; r}\_\mathrm{small}\&le;r\_\mathrm{size}\&le;r\_\mathrm{large}\\ r\_\mathrm{large}& \mathrm{if\; r}\_\mathrm{size}\&GreaterEqual;r\_\mathrm{large}\end{array}\right.$

r_size＝α ₁+β ₁r

$\mathrm{angle}\_\mathrm{size}=\left\{\begin{array}{cc}a\_\mathrm{small}& \mathrm{if\; a}\_\mathrm{size}\&le;a\_\mathrm{small}\\ a\_\mathrm{size}& \mathrm{if\; a}\_\mathrm{small}\&le;a\_\mathrm{size}\&le;a\_\mathrm{large}\\ a\_\mathrm{large}& \mathrm{if\; a}\_\mathrm{size}\&GreaterEqual;a\_\mathrm{large}\end{array}\right.$

a_size＝α ₂-β ₂r

α wherein ₁, β ₁, α ₂, β ₂Be predefined empirical parameter, and all greater than zero, r is the polar radius of template minutiae point, r_small, r_large, a_small, a_large are respectively the upper bound and the lower bounds of radius_size and angle_size, and they are values of setting by test in advance.

8. fingerprint characteristic quickly matching method as claimed in claim 6 is characterized in that:

Described coupling mark Ms obtains by following formula:

$\mathrm{Ms}=100\&times;\sqrt{\frac{\mathrm{nb}\_\mathrm{pair}*\mathrm{nb}\_\mathrm{pair}}{M_c*N_c}}+\mathrm{\&alpha;}*\mathrm{nb}\_\mathrm{pair}-\mathrm{\&beta;}*\mathrm{match}\_\mathrm{error}$

Wherein nb_pair is the right number of minutiae point on the coupling, and mat ch_error is the right cumulative matches error of each match point, M _c, N _cBe respectively template and input fingerprint image unique point number in the public domain, α, β is predetermined weighting coefficient.

9. fingerprint characteristic quickly matching method as claimed in claim 8 is characterized in that:

The value of described α is 2, and the value of β is 3.

10. fingerprint characteristic rapid matching apparatus is characterized in that comprising:

The central point judging unit is used for seeking and judging the central point of fingerprint;

Coupling corresponding point selected cell, it is right to be used for seeking the possible template and the input feature vector point that mate corresponding point of can be used as at template and input feature vector point, and they are write down;

The matching judgment unit is used to carry out the coupling calculating of unique point;

The polar coordinates converting unit is used for the locus of features relevant point and central point is converted to polar coordinates;

,, be corresponding point then at first, mate calculating, draw matching result by the matching judgment unit with this central point if there is central point by the central point in searching of central point judging unit and the judgement fingerprint; If central point fails to find, then other templates do not considered and input feature vector point are mated as the reference corresponding point two pieces of fingerprints being carried out minutiae point by coupling corresponding point selected cell and matching judgment unit, draw matching result; If it is not match that central point mates result calculated, be reference with described two central points then by coupling corresponding point selected cell, it is right to seek the possible template and the input feature vector point that can be used as the coupling corresponding point in template and input feature vector point, with these unique points to reference corresponding point as matching algorithm, by the matching judgment unit template and input feature vector point are carried out Feature Points Matching, draw matching result; Described matching judgment unit mates before the calculating, by the polar coordinates converting unit locus about central point and unique point is converted to polar coordinates.

11. a smart card has microprocessor, storer, telecommunication circuit, it is characterized in that:

Also has fingerprint characteristic rapid matching apparatus as claimed in claim 10 in the described smart card.

12. method of using smart card as claimed in claim 11 to carry out authentication, the user imports its fingerprint by the fingerprint collecting equipment on the card reader, after fingerprint sensor collects finger print data, submit to characteristic extracting module, and fingerprint characteristic is delivered to described smart card; Described smart card mates the fingerprint characteristic of input and the fingerprint characteristic of preservation, if the match is successful, then finishes authentication, it is characterized in that:

Described smart card adopts fingerprint characteristic quickly matching method as claimed in claim 1 to carry out the fingerprint characteristic coupling.

Images