CN104933740A - Certificate digital hole authenticity identification method - Google Patents

Abstract

The invention discloses a certificate digital hole authenticity identification method. The certificate digital hole authenticity identification method is specifically carried out by the following steps: acquiring a digital hole image under the condition that a region containing a digital hole in a certificate is amplified by 10-20 times; carrying out binaryzation on the digital hole image and extracting each digital hole by adopting a communication region analysis method; fitting each digital hole by adopting a least square fitting method; then extracting the circle center and the radius of a fitting circle; and judging the authenticity of the digital holes by adopting a comparing method and a K mean value clustering method. By virtue of the certificate digital hole authenticity identification method, the problems in the prior art that the time and labor are wasted when the authenticity of the certificate digital holes is manually identified and the identification efficiency is low are solved.

Description

A kind of certificate digital hole distinguishing method between true and false

Technical field

The invention belongs to certificate false proof authentication technique field, be specifically related to a kind of certificate digital hole distinguishing method between true and false.

Background technology

Certificate specifies according to law, gives or authorize the proof file of the aspects such as the identity of certain main body, qualification, power or honor, as resident identification card, passport, driver's license etc.Generally, all can occur some digital codes in certificate, the identity code etc. of the sequence number of such as certificate, the page number, holder, often these digital codes identify the uniqueness of certificate and the unitarity etc. of certificate and holder.In order to improve the security of certificate, it is false proof that these digital codes of additional clause adopt cheesing techniques to carry out usually, claims the hole forming these digital codes to be digital hole.Digital hole size on general real docu-ment is consistent, hole edge is smooth, digital hole site marshalling, and forge digital Kong Ze not there are these characteristics.

The digital hole of additional clause, the diameter in each hole generally between 0.1-0.8mm, normally when amplify 10-20 doubly, observed the standardization features such as the void shape of digital punching and arrangement by the mode of artificial naked eyes, with its true or false of identification.Artificial mirror method for distinguishing is wasted time and energy, and is difficult to the objectivity ensureing discriminating.

Summary of the invention

The object of this invention is to provide a kind of certificate digital hole distinguishing method between true and false, waste time and energy when solving the digital hole true and false of the artificial discriminating certificate existed in prior art and the not high problem of identification efficiency.

The technical solution adopted in the present invention is, a kind of certificate digital hole distinguishing method between true and false, specifically implements according to following steps:

Step 1, digital for certificate bore region is amplified 10 to 20 times, adopt transmission mode to carry out digitized image collection, if the digital hole pattern picture collected is f (x, y) to the digital bore region of additional clause;

Step 2, as f (x, y), binaryzation is carried out to the digital hole pattern gathered in step 1 and adopts the method for connected component analysis to extract each digital hole;

Step 3, to each digital hole of extracting in step 2, adopt each digital hole of method matching of Least Square Circle matching, and extract central coordinate of circle and the radius geometric parameter of fitting circle;

Step 4, according to the central coordinate of circle of the fitting circle obtained in step 3 and the radius geometric parameter resulting number code holes true and false, the concrete grammar of the resulting number code holes true and false has two kinds, first method is pairing comparision: each digital bore edges extracted in calculation procedure 3 and the difference value of fitting circle, and set a threshold value, if difference value is greater than threshold value, then judge that this digital hole is as the digital hole of forging; Second method is K means Method: the method adopting K mean cluster, in the row direction with on column direction cluster is carried out to the central coordinate of circle that all digital hole in step 3 simulates, then the difference that each digital hole circle heart departs from cluster centre is calculated, if difference value is greater than the threshold value of a setting, then show that digital hole is misaligned, belong to the digital hole of forgery.

Feature of the present invention is also,

The concrete grammar of step 2 is:

Step 2.1), adopt maximum variance between clusters, as f (x, y), binaryzation is carried out to the digital hole pattern in step 1, if the image after binaryzation is g (x, y);

Step 2.2), to step 2.1) in image g (x, y) after binaryzation adopt the method for 8 neighborhood connected component analysis, extract each digital hole, if the digital hole of extracting is h _i(i=1,2 ..., P), wherein P represents the number in digital hole.

The concrete grammar of step 3 is:

Step 3.1), adopt circle to carry out matching, if the i-th number code holes h to each digital hole _i(i=1,2 ..., P) and the central coordinate of circle of circle that simulates is (A _i, B _i), radius is R _i, then circle is expressed as:

${(x-A_i)}^2+{(y-B_i)}^2=R_i^2,---\left(1\right)$

The another kind of representation of formula (1) is: x ²+ y ²+ a _ix+b _iy+c _i=0, (2)

Formula (1) with the transforming relationship of formula (2) is: $A_i=-\frac{a_i}{2},B_i=-\frac{b_i}{2},R_i=\frac{1}{2}\sqrt{a_i^2+b_i^2-4c_i};$

Step 3.2), establish binaryzation after point on the profile of digital hole be expressed as (x _j, y _j), if Q is (a _i, b _i, c _i) for digital bore edges arriving a little the quadratic sum of rounded edge range difference, then:

$Q(a_i,b_i,c_i)=\underset{j=1}{\overset{L}{\Σ}}{(x_j^2+y_j^2+a_i{x}_j+b_i{y}_j+c_i)}^2,---\left(3\right)$

In formula (3), L is counting on digital bore edges;

Step 3.3), use Gauss-Newton nonlinear least square method to ask objective function Q (a to formula (3) _i, b _i, c _i) minimal value, obtain parameter (a _i, b _i, c _i), according to parameter (a _i, b _i, c _i) calculate digital hole h _icentral coordinate of circle be (A _i, B _i) and radius be R _i.

In step 4, the concrete grammar of pairing comparision is:

With digital hole h _i(i=1,2 ..., P) and the radius R of fitting circle _ifor radius, the center of circle (A simulated _i, B _i) for initial point, whole circumference is divided into K decile, calculate the distance difference information of each point on each point corresponding on digital bore edges and the circle that simulates, in this, as the foundation of digital hole True-false distinguish, its computing formula is such as formula shown in (4):

$C=1-\frac{\underset{j=1}{\overset{K}{\Σ}}{(d_j-R_i)}^2}{{\mathrm{KR}}_i},---\left(4\right)$

D in formula (4) _jit is the center of circle that the simulates distance to digital hole bore edges;

If threshold value T ₁=0.9, if C < is T ₁, then this digital hole is the digital hole of forging.

In step 4, the concrete grammar of K means Method is:

If in composition certificate, the digital hole of digital number is capable by M, N row form, the central coordinate of circle that each digital hole simulates is (A _i, B _i), then the logarithmic code hole circle heart carries out cluster in the row direction with on column direction according to the following steps, and calculates the difference departing from cluster centre:

Step 4.1), adopt the method for K mean cluster, to the ordinate B in all digital holes _ibe clustered into M class C in the row direction ₁, C ₂..., C _mif the cluster centre coordinate figure of M class is m ₁, m ₂..., m _m, to any one digital hole h _i(i=1,2 ..., P), if threshold value T ₂=5, if its ordinate B _ibelong to C _jclass, then compare B _iwith m _jvalue, if | B _i-m _j| > T ₂, then digital hole h is thought _iforge;

Step 4.2), adopt the method for K mean cluster, to the horizontal ordinate A in all digital holes _ibe clustered into N class C in a column direction ₁, C ₂..., C _nif the cluster centre coordinate figure of N class is n ₁, n ₂..., n _n, to any one digital hole h _i(i=1,2 ..., P), if its ordinate A _ibelong to C _jclass, then compare A _iwith n _jvalue, if | A _i-n _j| > T ₂, then digital hole h is thought _iforge.

The computing formula of K mean cluster is:

$E=\underset{i=1}{\overset{M}{\&Sigma;}}\underset{B_j\&Element;C_i}{\&Sigma;}{||B_j-m_i||}^2,---\left(5\right)$

M in formula (5) ₁, m ₂..., m _mrepresent each cluster centre coordinate figure, by each cluster centre value of continuous iterative modifications, E value is reduced gradually.

The invention has the beneficial effects as follows, a kind of certificate digital hole distinguishing method between true and false, gathering digital hole pattern picture by amplifying, making digital hole pattern picture more clear.The circularity of logarithm code holes can carry out quantitative calculating, current circularity can be overcome and judge accurate not defect, can differentiate the digital hole of forging from local shape.By the judgement of logarithmic code hole rows direction and column direction, can differentiate the digital hole of forging on the whole.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of a kind of certificate of the present invention digital hole distinguishing method between true and false.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.

A kind of certificate digital hole of the present invention distinguishing method between true and false, steps flow chart as shown in Figure 1, is specifically implemented according to following steps:

Step 1, digital for certificate bore region is amplified 10 to 20 times, adopt transmission mode to carry out digitized image collection, if the digital hole pattern picture collected is f (x, y) to the digital bore region of additional clause;

Step 2, carry out binaryzation and adopt the method for connected component analysis to extract each digital hole to the digital hole pattern gathered in step 1 as f (x, y), concrete grammar is:

Step 2.1), adopt maximum variance between clusters, as f (x, y), binaryzation is carried out to the digital hole pattern in step 1, if the image after binaryzation is g (x, y);

Step 2.2), to step 2.1) in image g (x, y) after binaryzation adopt the method for 8 neighborhood connected component analysis, extract each digital hole, if the digital hole of extracting is h _i(i=1,2 ..., P), wherein P represents the number in digital hole;

Step 3, to each digital hole of extracting in step 2, adopt each digital hole of method matching of Least Square Circle matching, and extract central coordinate of circle and the radius geometric parameter of fitting circle, detailed process is as follows:

Step 3.1), adopt circle to carry out matching, if the i-th number code holes h to each digital hole _i(i=1,2 ..., P) and the central coordinate of circle of circle that simulates is (A _i, B _i), radius is R _i, then circle is expressed as:

${(x-A_i)}^2+{(y-B_i)}^2=R_i^2,---\left(1\right)$

The another kind of representation of formula (1) is: x ²+ y ²+ a _ix+b _iy+c _i=0, (2)

Formula (1) with the transforming relationship of formula (2) is: $A_i=-\frac{a_i}{2},B_i=-\frac{b_i}{2},R_i=\frac{1}{2}\sqrt{a_i^2+b_i^2-4c_i};$

Step 3.2), establish binaryzation after point on the profile of digital hole be expressed as (x _j, y _j), if Q is (a _i, b _i, c _i) for digital bore edges arriving a little the quadratic sum of rounded edge range difference, then:

$Q(a_i,b_i,c_i)=\underset{j=1}{\overset{L}{\&Sigma;}}{(x_j^2+y_j^2+a_i{x}_j+b_i{y}_j+c_i)}^2,---\left(3\right)$

In formula (3), L is counting on digital bore edges;

Step 3.3), use Gauss-Newton nonlinear least square method to ask objective function Q (a to formula (3) _i, b _i, c _i) minimal value, obtain parameter (a _i, b _i, c _i), according to parameter (a _i, b _i, c _i) calculate digital hole h _icentral coordinate of circle be (A _i, B _i) and radius be R _i.

Step 4, according to the central coordinate of circle of the fitting circle obtained in step 3 and the radius geometric parameter resulting number code holes true and false, the concrete grammar of the resulting number code holes true and false has two kinds, first method is pairing comparision: each digital bore edges extracted in calculation procedure 3 and the difference value of fitting circle, and set a threshold value, if difference value is greater than threshold value, then judge that this digital hole is as the digital hole of forging; Second method is K means Method: the method adopting K mean cluster, in the row direction with on column direction cluster is carried out to the central coordinate of circle that all digital hole in step 3 simulates, then the difference that each digital hole circle heart departs from cluster centre is calculated, if difference value is greater than the threshold value of setting, then show that digital hole is misaligned, belong to the digital hole of forgery, wherein, the concrete grammar of pairing comparision is:

With digital hole h _i(i=1,2 ..., P) and the radius R of fitting circle _ifor radius, the center of circle (A simulated _i, B _i) for initial point, whole circumference is divided into K decile, calculate the distance difference information of each point on each point corresponding on digital bore edges and the circle that simulates, in this, as the foundation of digital hole True-false distinguish, its computing formula is such as formula shown in (4):

$C=1-\frac{\underset{j=1}{\overset{K}{\&Sigma;}}{(d_j-R_i)}^2}{{\mathrm{KR}}_i},---\left(4\right)$

D in formula (4) _jit is the center of circle that the simulates distance to digital hole bore edges;

If threshold value T ₁=0.9, if C < is T ₁, then this digital hole is the digital hole of forging.

The concrete grammar of K means Method is:

If in composition certificate, the digital hole of digital number is capable by M, N row form, the central coordinate of circle that each digital hole simulates is (A _i, B _i), then the logarithmic code hole circle heart carries out cluster in the row direction with on column direction according to the following steps, and calculates the difference departing from cluster centre:

Step 4.1), adopt the method for K mean cluster, to the ordinate B in all digital holes _ibe clustered into M class C in the row direction ₁, C ₂..., C _m, specifically carry out K mean cluster according to formula (5):

$E=\underset{i=1}{\overset{M}{\&Sigma;}}\underset{B_j\&Element;C_i}{\&Sigma;}{||B_j-m_i||}^2---\left(5\right)$

M in formula (5) ₁, m ₂..., m _mthe cluster centre coordinate figure of expression M class is m ₁, m ₂..., m _m, by each cluster centre value of continuous iterative modifications, E value is reduced gradually.

To any one digital hole h _i(i=1,2 ..., P), if threshold value T ₂=5, if its ordinate B _ibelong to C _jclass, then compare B _iwith m _jvalue, if | B _i-m _j| > T ₂, then digital hole h is thought _iforge;

Step 4.2), adopt the method for K mean cluster, to the horizontal ordinate A in all digital holes _ibe clustered into N class C in a column direction ₁, C ₂..., C _nif the cluster centre coordinate figure of N class is n ₁, n ₂..., n _n, to any one digital hole h _i(i=1,2 ..., P), if its ordinate A _ibelong to C _jclass, then compare A _iwith n _jvalue, if | A _i-n _j| > T ₂, then digital hole h is thought _iforge.

A kind of certificate digital hole of the present invention distinguishing method between true and false, the shape own from digital hole, namely the difference of digital bore edges and its fitting circle is set out, and can judge the true and false in digital hole from local; Propose the arrangement from digital hole rows direction and column direction simultaneously, differentiate from integral position to the true and false in the digital hole of departing from cluster centre coordinate, the method proposed is considered comprehensively, can obtain better identification result.

Claims (6)

1. a certificate digital hole distinguishing method between true and false, is characterized in that, specifically implement according to following steps:

Step 1, digital for certificate bore region is amplified 10 to 20 times, adopt transmission mode to carry out digitized image collection, if the digital hole pattern picture collected is f (x, y) to the digital bore region of additional clause;

Step 2, as f (x, y), binaryzation is carried out to the digital hole pattern gathered in described step 1 and adopts the method for connected component analysis to extract each digital hole;

Step 3, to each digital hole of extracting in described step 2, adopt each digital hole of method matching of Least Square Circle matching, and extract central coordinate of circle and the radius geometric parameter of fitting circle;

Step 4, according to the central coordinate of circle of the fitting circle obtained in described step 3 and the radius geometric parameter resulting number code holes true and false, the concrete grammar of the described resulting number code holes true and false has two kinds, first method is pairing comparision: the difference value calculating each digital bore edges and the fitting circle extracted in described step 3, and set a threshold value, if difference value is greater than threshold value, then judge that this digital hole is as the digital hole of forging; Second method is K means Method: the method adopting K mean cluster, in the row direction with on column direction cluster is carried out to the central coordinate of circle that all digital hole in described step 3 simulates, then the difference that each digital hole circle heart departs from cluster centre is calculated, if difference value is greater than the threshold value of a setting, then show that digital hole is misaligned, belong to the digital hole of forgery.

2. the digital hole of a kind of certificate according to claim 1 distinguishing method between true and false, it is characterized in that, the concrete grammar of described step 2 is:

Step 2.1), adopt maximum variance between clusters, as f (x, y), binaryzation is carried out to the digital hole pattern in described step 1, if the image after binaryzation is g (x, y);

Step 2.2), to described step 2.1) in image g (x, y) after binaryzation adopt the method for 8 neighborhood connected component analysis, extract each digital hole, if the digital hole of extracting is h _i(i=1,2 ..., P), wherein P represents the number in digital hole.

3. the digital hole of a kind of certificate according to claim 1 distinguishing method between true and false, it is characterized in that, the concrete grammar of described step 3 is:

Step 3.1), adopt circle to carry out matching, if the i-th number code holes h to each digital hole _i(i=1,2 ..., P) and the central coordinate of circle of circle that simulates is (A _i, B _i), radius is R _i, then circle is expressed as:

${(x-A_i)}^2+{(y-B_i)}^2=R_i^2,---\left(1\right)$

The another kind of representation of formula (1) is: x ²+ y ²+ a _ix+b _iy+c _i=0, (2)

Formula (1) with the transforming relationship of formula (2) is: $A_i=-\frac{a_i}{2},$ $B_i=-\frac{b_i}{2},$ $R_i=\frac{1}{2}\sqrt{a_i^2+b_i^2-4c_i};$

Step 3.2), establish binaryzation after point on the profile of digital hole be expressed as (x _j, y _j), if Q is (a _i, b _i, c _i) for digital bore edges arriving a little the quadratic sum of rounded edge range difference, then:

$Q(a_i,b_i,c_i)=\underset{j=1}{\overset{L}{\mathrm{\&Sigma;}}}{(x_j^2+y_j^2+a_i{x}_j+b_i{y}_j+c_i)}^2,---\left(3\right)$

In formula (3), L is counting on digital bore edges;

Step 3.3), use Gauss-Newton nonlinear least square method to ask objective function Q (a to formula (3) _i, b _i, c _i) minimal value, obtain parameter (a _i, b _i, c _i), according to parameter (a _i, b _i, c _i) calculate digital hole h _icentral coordinate of circle be (A _i, B _i) and radius be R _i.

4. the digital hole of a kind of certificate according to claim 1 distinguishing method between true and false, it is characterized in that, in described step 4, the concrete grammar of pairing comparision is:

With digital hole h _i(i=1,2 ..., P) and the radius R of fitting circle _ifor radius, the center of circle (A simulated _i, B _i) for initial point, whole circumference is divided into K decile, calculate the distance difference information of each point on each point corresponding on digital bore edges and the circle that simulates, in this, as the foundation of digital hole True-false distinguish, its computing formula is such as formula shown in (4):

$C=1-\frac{\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{(d_j-R_i)}^2}{K{R}_i},---\left(4\right)$

D in formula (4) _jit is the center of circle that the simulates distance to digital hole bore edges;

If threshold value T ₁=0.9, if C < is T ₁, then this digital hole is the digital hole of forging.

5. the digital hole of a kind of certificate according to claim 1 distinguishing method between true and false, it is characterized in that, in described step 4, the concrete grammar of K means Method is:

If in composition certificate, the digital hole of digital number is capable by M, N row form, the central coordinate of circle that each digital hole simulates is (A _i, B _i), then the logarithmic code hole circle heart carries out cluster in the row direction with on column direction according to the following steps, and calculates the difference departing from cluster centre:

Step 4.1), adopt the method for K mean cluster, to the ordinate B in all digital holes _ibe clustered into M class C in the row direction ₁, C ₂, L, C _mif the cluster centre coordinate figure of described M class is m ₁, m ₂, L, m _m, to any one digital hole h _i(i=1,2 ..., P), if threshold value T ₂=5, if its ordinate B _ibelong to C _jclass, then compare B _iwith m _jvalue, if | B _i-m _j| > T ₂, then digital hole h is thought _iforge;

Step 4.2), adopt the method for K mean cluster, to the horizontal ordinate A in all digital holes _ibe clustered into N class C in a column direction ₁, C ₂, L, C _nif the cluster centre coordinate figure of this N class is n ₁, n ₂, L, n _n, to any one digital hole h _i(i=1,2 ..., P), if its ordinate A _ibelong to C _jclass, then compare A _iwith n _jvalue, if | A _i-n _j| > T ₂, then digital hole h is thought _iforge.

6. the digital hole of a kind of certificate according to claim 5 distinguishing method between true and false, it is characterized in that, the computing formula of described K mean cluster is:

$E=\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{B_j\&Element;C_i}{\mathrm{\&Sigma;}}{\left|\right|B_j-m_i\left|\right|}^2,---\left(5\right)$

M in formula (5) ₁, m ₂, L, m _mrepresent each cluster centre coordinate figure, by each cluster centre value of continuous iterative modifications, E value is reduced gradually.