CN100576230C - Based on similar fingerprint recognition system of the twins of partial structurtes and method - Google Patents

Based on similar fingerprint recognition system of the twins of partial structurtes and method Download PDF

Info

Publication number
CN100576230C
CN100576230C CN200610113409A CN200610113409A CN100576230C CN 100576230 C CN100576230 C CN 100576230C CN 200610113409 A CN200610113409 A CN 200610113409A CN 200610113409 A CN200610113409 A CN 200610113409A CN 100576230 C CN100576230 C CN 100576230C
Authority
CN
China
Prior art keywords
fingerprint
partial structurtes
minutiae point
template
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN200610113409A
Other languages
Chinese (zh)
Other versions
CN101154263A (en
Inventor
田捷
时鹏
谢卫华
杨鑫
秦承虎
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Automation of Chinese Academy of Science
Original Assignee
Institute of Automation of Chinese Academy of Science
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Institute of Automation of Chinese Academy of Science filed Critical Institute of Automation of Chinese Academy of Science
Priority to CN200610113409A priority Critical patent/CN100576230C/en
Publication of CN101154263A publication Critical patent/CN101154263A/en
Application granted granted Critical
Publication of CN100576230C publication Critical patent/CN100576230C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The present invention discloses similar fingerprint recognition system of a kind of twins based on partial structurtes and method, and recognition system comprises: pretreatment unit, partial structurtes plate device, minutiae point partial structurtes coalignment; Method comprises: the fingerprint image that collects is carried out pre-service, obtain the refined image of fingerprint; Extract crestal line tip and bifurcated detailed points P in the refined image, the set structure partial structurtes to the detailed points P extracted generate fingerprint template with each partial structurtes combination in the fingerprint; Fingerprint template in input fingerprint partial structurtes template and the database is compared, and the relative position between the calculating partial structurtes and the difference of direction parameter judge whether fingerprint mates.Because the present invention uses the minutiae point partial structurtes, not only significantly improved the recognition performance of Automated Fingerprint Identification System for the similar fingerprint of twins, and can avoid the discrimination decline problem that causes because of the fingerprint global characteristics is similar, widen its application in this specific group of twins.

Description

Based on similar fingerprint recognition system of the twins of partial structurtes and method
Technical field
The invention belongs to the biometrics identification technology field, be specifically related to fingerprint identification method, relate in particular to similar fingerprint identification method based on the twins of partial structurtes.
Background technology
Along with the development of society and the raising of people's awareness of safety, the effect that the biometrics identification technology play more and more is important also has increasingly extensive application in the every aspect of fields of society and people's daily life.Though it is less that twins account for the ratio of population, the degree of concern that is subjected to is very high, to the difficult point that effective identity verification is studied especially of carrying out of this specific group.Because twins are the essentially identical individualities of gene, so they have quite similar biological characteristic, as people's face, body and sound etc., thus adopt means such as people's face, body or speech recognition, can't be with correct the making a distinction of twins.And because twinborn gene is almost completely identical, so the method for using DNA to identify also is difficult to twins are discerned and verify.Meanwhile, the similarity degree of twinborn fingerprint is also very high.Therefore, the technical difficulty that twinborn similar fingerprint is effectively discerned is far longer than the fingerprint recognition to the general population.
Medical research shows, people's fingerprint mainly is to form in the 13rd to 19 week of fetal period, and the factor of decision fingerprint form mainly contains two, the one, and inherent cause, i.e. DNA, the 2nd, the developing environment of fetus in parent is as the umbilical cord blood flow etc.Twins are the most close at this aspect two, and therefore the global characteristics similarity in twinborn fingerprint, the especially fingerprint is very high, and this has brought huge challenge just for twinborn fingerprint recognition.
Automatically fingerprint identification technology is that fingerprint and the database of fingerprint that will collect compares, to determine whether to belong to the process of certain finger.The main foundation of fingerprint identification method comprises the global characteristics of fingerprint, as fingerprint classification, singular point, crestal line stream etc., and local feature, as minutiae point etc.Because global characteristics is relatively stable in fingerprint collecting, and local feature quantity more, be convenient to comparison, therefore the present fingerprint identification method method that all adopts global characteristics to combine with local feature is carried out fingerprint recognition.Yet for the closely similar twins' fingerprint of global characteristics, traditional fingerprint identification method has significant disadvantages: in the global characteristics unanimity and under the similar situation of local feature, cause the mistake of fingerprint to know easily, caused the discrimination of twins colony fingerprint significantly to reduce, the range of application of Automated Fingerprint Identification System is restricted.
Summary of the invention
For the closely similar twins' fingerprint of global characteristics, traditional fingerprint identification method has significant disadvantages: in the global characteristics unanimity and under the similar situation of local feature, cause the mistake of fingerprint to know easily, caused the discrimination of twins colony fingerprint significantly to reduce, the range of application of automatic system of fingerprint recognition is restricted, in order to solve prior art problems, the purpose of this invention is to provide a kind of similar fingerprint recognition system of twins based on partial structurtes and method that can improve the discrimination of twins colony fingerprint.
In order to realize purpose of the present invention, an aspect of of the present present invention provides the similar fingerprint recognition system of a kind of twins based on partial structurtes, comprising:
With the pretreatment unit that the input picture end is connected, be used for to the fingerprint image extraction field of direction image of input, the prospect of cutting apart fingerprint image and background, to the filtering of image travel direction output fingerprint thinning image;
The partial structurtes plate device that is connected with pretreatment unit is used for extracting the refined image minutiae point, and structure detail point partial structurtes generate fingerprint template with each partial structurtes combination in the fingerprint;
The minutiae point partial structurtes coalignment that is connected with the partial structurtes plate device: fingerprint template in comparison input fingerprint partial structurtes template and the database is used to export the fingerprint recognition result.
According to embodiments of the invention, described pretreatment unit comprises:
Field of direction extraction unit: the relation that is used for analyzing the gradient direction of fingerprint gray scale point and waits to estimate direction, extract field of direction image;
The cutting unit that is connected with field of direction extraction unit: be used for removing the described field of direction image field of direction and change disorderly part, form the mask image of fingerprint;
The image filtering unit that is connected with cutting unit: be used for the filtering of mask image travel direction;
The refinement unit that is connected with the image filtering unit: adopt perpendicular to the x direction of principal axis or perpendicular to y direction thinning processing, be used for the line integral of crestal line direction gradient is carried out refinement by the section of crossing over crestal line, obtain the refinement fingerprint image.
According to embodiments of the invention, described minutiae point partial structurtes coalignment also comprises:
The minutiae feature unit is used for all minutiae point vectors are done as a whole corresponding one by one with fingerprint image;
The first matching unit that is connected with the minutiae feature unit: it is right to be used for that the minutiae point vector is matched to minutiae point, and generates fingerprint image minutiae point structure;
The template matches unit that is connected with first matching unit: minutiae point structure in the fingerprint partial structurtes template and the minutiae point structure in the database fingerprint are mated, obtain the coupling array, the assembly average that is used for calculating energy coupling array obtains translation parameters and rotation parameter;
The minutiae point that connects with the template matches unit is to the comparison unit: be used for the corresponding minutiae point of fingerprint partial structurtes template and database fingerprint template secondary compared, relatively the minutiae point of correspondence to and obtain mating mark.
In order to realize purpose of the present invention, another aspect of the present invention provides the similar fingerprint identification method of a kind of twins based on partial structurtes, comprises step:
Pre-treatment step: the fingerprint image that collects is carried out pre-service, obtain the refined image of fingerprint;
Structure partial structurtes template step: extract crestal line tip and bifurcated detailed points P in the refined image, the set structure partial structurtes to the detailed points P extracted generate fingerprint template with each partial structurtes combination in the fingerprint;
Minutiae point partial structurtes coupling step: fingerprint template in input fingerprint partial structurtes template and the database is compared, and the relative position between the calculating partial structurtes and the difference of direction parameter judge whether fingerprint mates.
According to embodiments of the invention, described pre-treatment step also comprises:
Step 1.1 field of direction image extracts: adopt the gradient direction of analyzing gray scale point in the fingerprint and the relation of waiting to estimate direction, utilize the method for fuzzy logic to go to solve the shared ratio of four direction, utilize the value of mapping direction, estimate to obtain field of direction image;
Step 1.2 is cut apart field of direction image: remove the disorderly part of field of direction variation in the described field of direction image, form the mask image of fingerprint;
Step 1.3 trend pass filtering: utilize one based on the template of parallelogram with utilize sliding window operation algorithm, to the filtering of mask image travel direction with directivity;
Step 1.4 trend pass filtering image thinning: adopt perpendicular to the x direction of principal axis or perpendicular to the thinning processing method under two kinds of situations of y direction, adopt line integral to carry out refinement to the direction filtering image: to exist for the crestal line direction by the section of crossing over crestal line π 4 3 π 4 Between line integral adopt perpendicular to the axial conversion of y, exist for the crestal line direction 0 π 4 With 3 π 4 π Between situation adopt perpendicular to the axial conversion of x.
According to embodiments of the invention, described structure partial structurtes template step also comprises:
Details point set information according to refined image is extracted generates fingerprint template to be compared, when the fingerprint image of gathering has extracted minutiae point point set M={M k, 1≤k≤L}, wherein L is the number that this point is concentrated minutiae point; For any one minutiae point wherein, its eigenvector is
Figure C20061011340900084
Constitute the fingerprint basic templates by this eigenvector.
According to embodiments of the invention, described minutiae point partial structurtes coupling step also comprises:
Step 3.1 obtains the feature of each minutiae point, comprising minutiae point x, and y coordinate, direction θ and field of direction provincial characteristics ZOD, all minutiae point vector P iDo as a whole corresponding one by one with fingerprint image, P i=[(x, y), θ, ZOD];
The first coupling of step 3.2 fingerprint image: the minutiae point that application specific details point vector obtains mating is right, and generates the minutiae point structure;
Step 3.3 template matches: minutiae point structure in the fingerprint partial structurtes template and the minutiae point structure in the database fingerprint are mated, obtain two coupling arrays, one of them record coupling array position translation energy, another one coupling An arrayed recording angle rotating energy, the assembly average of calculating energy coupling array obtains translation parameters and rotation parameter;
Step 3.4 couple fingerprint partial structurtes template Fl kWith minutiae point corresponding in the database fingerprint template to carrying out the secondary comparison, by integral position translation and angle rotation, minutiae point partial structurtes in the template are adjusted, relatively more corresponding minutiae point to and obtain mating mark.
The invention has the beneficial effects as follows:
Because twins have the most close DNA, and identical parent developing environment is arranged, its fingerprint is proved has very high similarity, and especially similarity degree is higher on the fingerprint global characteristics, and often error rate is higher in the application of twins' fingerprint recognition to cause existing fingerprint identification method.The present invention is based on the distinguishability of fingerprint local feature structure, proposed finger print matching method, can discern twinborn fingerprint effectively based on partial structurtes.Because the present invention has used the minutiae point partial structurtes, make to adopt method of the present invention not only significantly improve the recognition performance of Automated Fingerprint Identification System, and can avoid the discrimination decline problem that causes because of the fingerprint global characteristics is similar for the similar fingerprint of twins.Twins' fingerprint recognition is the difficult point in the fingerprint recognition technology, this method is when being applied to general population's fingerprint recognition, significantly improved the recognition capability of Automated Fingerprint Identification System for the similar fingerprint of twins, widened its range of application in this specific group of twins, easy, the means of twins' identity authentication efficiently are provided, have had important use and be worth.
Description of drawings
Fig. 1 is a configuration diagram of utilizing system of the present invention;
Fig. 2 the present invention is based on the similar fingerprint recognition system theory diagram of twins of partial structurtes
Fig. 3 is a pretreatment unit block scheme of the present invention
Fig. 4 minutiae point partial structurtes coalignment block scheme
Fig. 5 the present invention extracts minutiae feature synoptic diagram in the corresponding finger print of twins;
Fig. 6 minutiae point of the present invention is to synoptic diagram;
Fig. 7 minutiae point partial structurtes of the present invention synoptic diagram;
Fig. 8 overlapping region calibration of the present invention synoptic diagram.
Embodiment
The invention will be further described below in conjunction with drawings and Examples, is to be noted that described embodiment only is intended to be convenient to the understanding of the present invention, and it is not played any qualification effect.
As Fig. 1, be to utilize shown in the framework twins identification system synoptic diagram of system of the present invention:
Twins' fingerprint identity management system comprises: system manage ment computer and fingerprint acquisition instrument, wherein supervisory computer comprises: twins' fingerprint database, fingerprint recognition system.
The concrete grammar of realizing fingerprint recognition system of the present invention is as follows: the present invention is after utilizing fingerprint acquisition instrument to gather twinborn information in fingerprint to be compared, import the information in fingerprint that collects into twins' identification system supervisory computer by USB interface, carry out pre-service by the 1 pair of fingerprint image of pretreatment unit of fingerprint recognition system in the described system computer and obtain refined image, and the minutiae point in taking the fingerprint; By partial structurtes plate device 2 minutiae point is configured to the minutiae point partial structurtes then, to the minutiae point of extracting construct that relative position is stable, the fingerprint template information of anti-deformation, antimierophonic partial structurtes; Pass through the difference between the minutiae point partial structurtes coalignment 3 calculated fingerprint partial structurtes at last, thereby judge whether fingerprint mates.Fingerprint recognition system of the present invention travels through comparison with the template in template to be compared and the twins' fingerprint database, draws comparison result and output.
Shown in the similar fingerprint recognition system of twins that Fig. 2 the present invention is based on partial structurtes, comprising: pretreatment unit 1, partial structurtes plate device 2, minutiae point partial structurtes coalignment 3, wherein:
The pretreatment unit 1 that is connected with the input picture end is used for to the fingerprint image extraction field of direction image of input, the prospect of cutting apart fingerprint image and background, to the filtering of image travel direction output fingerprint thinning image;
The partial structurtes plate device 2 that is connected with pretreatment unit 1 is used for extracting the refined image minutiae point, and structure detail point partial structurtes generate fingerprint template with each partial structurtes combination in the fingerprint;
The minutiae point partial structurtes coalignment 3 that is connected with partial structurtes plate device 2: fingerprint template in comparison input fingerprint partial structurtes template and the database is used to export the fingerprint recognition result.
Each device program of the present invention is write by C Plus Plus, and running environment is Windows2000 and WindowsXP system, needs the computer configuration USB interface to connect fingerprint acquisition instrument.
As shown in Figure 3, pretreatment unit 1, be to the fingerprint image travel direction field that collects extract, cut apart, pre-service such as filtering, refinement, obtain the refined image of fingerprint, detailed points P such as crestal line tip in the extraction refined image and bifurcated, concrete implementation step is as follows:
Field of direction extraction unit 11: to the input fingerprint image is the relation of adopt analyzing 0 °, 45 °, 90 ° of the gradient directions of gray scale point in four fingerprints and 135 ° and waiting to estimate direction, utilizes the method for fuzzy logic to remove to solve four direction: 0 °, 45 °, 90 ° and 135 ° of shared ratios
Figure C20061011340900101
If mapping [0 °-180 °] is the value in [0-128], thus estimate the field of direction asked as shown in the formula, O = ∂ 1 + ∂ 2 × 32 + ∂ 3 × 64 + ∂ 4 × 96 . Wherein 1,32,64 and 96 distinguish 0 °, 45 °, 90 ° and 135 ° pairing direction values of corresponding four directions.This method is only used the elementary arithmetic computing, and execution speed is faster than the general arctan function that utilizes two gradient directions.
Cutting unit 12 to the field of direction image that above-mentioned steps extracts, is got rid of field of direction image direction field and is changed disorderly part, obtains the mask image ImgMsk of fingerprint;
Image filtering unit 13 is used for the filtering of mask image ImgMsk travel direction; Utilize a template to come travel direction filtering based on parallelogram with directivity, when carrying out filtering, utilize the method for sliding window to come the operation of accelerating algorithm, can further improve the efficient that algorithm is carried out, also can access very desirable figure image intensifying effect.
Refinement unit 14 is undertaken by the section of crossing over crestal line the line integral of crestal line direction gradient, only adopts perpendicular to the x direction of principal axis or perpendicular to the disposal route under two kinds of situations of y direction.Exist for the crestal line direction π 4 3 π 4 Between line integral adopt perpendicular to the axial conversion of y, exist for the crestal line direction 0 π 4 With 3 π 4 π Between situation adopt perpendicular to the axial conversion of x output fingerprint thinning result.
2, the step of described structure partial structurtes plate device 2:
Details point set information according to refined image is extracted generates fingerprint template to be compared, when the fingerprint image of gathering has extracted minutiae point point set M={M k, 1≤k≤L}, wherein L is the number that this point is concentrated minutiae point.For any one minutiae point wherein, its eigenvector is
Figure C20061011340900114
This eigenvector has just constituted the basic templates of fingerprint.
3, minutiae point partial structurtes coalignment as described in Figure 4, the template of fingerprint is compared in the partial structurtes template of input fingerprint and the database, by calculating relative position and the isoparametric difference of direction between the partial structurtes, judge whether fingerprint mates, and the specific implementation step is as follows:
Minutiae feature unit 31, as shown in Figure 5, also identifying with the short ray of band direction from extracting minutiae point with a pair of twinborn two width of cloth fingerprint images, wherein the end points of ray is a minutiae point, direction is the direction of minutiae point, and the detailed points P of extracting is further handled the proper vector of computational details point, obtain the feature of each minutiae point, comprising 4 characteristic quantities:
● the coordinate x of minutiae point under global coordinate system, y
● the direction θ under global coordinate system
● field of direction provincial characteristics ZOD
All minutiae point vector P iDo as a whole corresponding one by one with fingerprint image, P i=[(x, y), θ, ZOD];
The first matching unit 32 of fingerprint image: the minutiae point that application specific details point vector obtains mating is to P iP j, and generate minutiae point structure Fl k
Template matches unit 33: with the Fl of input fingerprint kAs template to Fl kFl ' with database fingerprint kMate, obtain two coupling arrays, the deviation of position translation in one of them record coupling array, the deviation of angle rotation in the another one record coupling array, by these two arrays, calculate translation parameters and rotation parameter respectively and be: nOri = Σ i = 0 M PO i × i Σ i = 0 M PO i , Wherein, i represents i minutiae point, and M is the maximal value that possible angle is rotated, and nOri is the quantity of angle rotation, PO iIt is the similarity weights array of the record i position deviation of ordering;
Minutiae point is to comparing unit 34: minutiae point corresponding in the template is compared carrying out secondary: by integral position translation and angle rotation, minutiae point partial structurtes in the template are adjusted, because the details point set of adjusted template details point set and input is closely similar, thus mate for the second time the minutiae point that only needs the comparison correspondence to and obtain mating mark.
4, described local detail dot structure structure promptly extracts the minutiae point of input fingerprint and database fingerprint, and structure detail point partial structurtes, for ensuing matching process forms template.Structure detail point partial structurtes are important steps in the fingerprint comparison, and the order of accuarcy of minutiae point structure construction directly has influence on the performance of whole fingerprint identification method.The concrete building method of described minutiae point partial structurtes is as follows:
A) detailed points P of extracting in the step 1 is further handled, the proper vector of computational details point, comprising 4 characteristic quantities:
● the coordinate x of minutiae point under global coordinate system, y
● the direction θ under global coordinate system
● field of direction provincial characteristics ZOD
All minutiae point vector P iDo as a whole corresponding one by one with fingerprint image, P i=[(x, y), θ, ZOD];
B) the vectorial minutiae point that obtains mating of application specific details point is to P iP j, and generate multiple degrees of freedom vector structure Fl k, need be calculated as follows 5 characteristic quantities, referring to accompanying drawing 6 minutiae point of the present invention to synoptic diagram, wherein:
● minutiae point i, between the j apart from d Ij
● minutiae point i, the angle θ of j i, θ j
● minutiae point i, j and minutiae point are to P iP jBetween direction difference Z i, Z j
C) according to minutiae point to P iP j, with and contiguous minutiae point P n, generate multivariant minutiae point vector structure Fl k, need be calculated as follows 8 characteristic quantities, referring to Fig. 7 minutiae point partial structurtes of the present invention synoptic diagram, wherein:
● minutiae point k and minutiae point i, between the j apart from d Ki, d Kj
● minutiae point k and minutiae point i, the direction difference Z between the j Ki, Z Kj
● vectorial ki, the angle θ of kj Ki, θ Kj
● vectorial ki, kj the crestal line number n of process Ki, n Kj
The minutiae point structure is a vector set, Fl k=(d Ki, d Kj, θ Ki, θ Kj, Z Ki, Z Kj, n Ki, n Kj, t k, t i, t j) T, t is the type of each minutiae point, as tip point, bifurcation etc., described in other parameter-definition such as the claim 4c;
D) fingerprint image in the database is taked same step, obtain its minutiae point partial structurtes Fl ' l, the partial structurtes Fl of its construction process and fingerprint to be matched kIdentical;
E) form template Fl kAnd Fl ' k, enter the partial structurtes coupling.
5, carry out minutiae point partial structurtes coupling, promptly the template of fingerprint in the partial structurtes template of input fingerprint and the database is compared, by calculating relative position and the isoparametric difference of direction between the partial structurtes, judge whether fingerprint mates, and described template matching method is as follows:
A) to Fl kAnd Fl ' kMate, obtain two coupling arrays, the energy of one of them record position translation, the energy of another one recording angular rotation, the assembly average of calculating energy array can obtain translation parameters and rotation parameter nOri:
Computing formula is, nOri = Σ i = 0 M PO i × i Σ i = 0 M PO i , Wherein, M is the maximal value that possible angle is rotated, and nOri is the quantity of angle rotation, and the quantity computation method of position translation is the same with the angle rotation;
B), treat the overlapping region of image in comparison chart picture and the database and calibrate, Fl according to translation parameters and rotation parameter nOri kAnd Fl ' kCarry out the right calculating of minutiae point in the structure, draw details on the coupling n that counts, referring to Fig. 8, calibration synoptic diagram in overlapping region of the present invention, wherein last figure is two fingerprint images to be matched, figure below is the synoptic diagram of overlapping region between two width of cloth images.
C) minutiae point is right structure represents with multidimensional vector, Calculate the similarity of two details point sets in the template of fingerprint in template to be compared and the database,
S ( k , j ) = pwr - | p i m - p j t | pwr , | p i m - p j t | < Thr 0 , otherwise
P wherein i m, p j iBe the coupling minutiae point concentrate i point to and concentrated j the minutiae point of template minutiae point right, pwr is the coefficient of normalization weights, Thr be in the template to be matched minutiae point to database in the threshold value of the right pairing of template minutiae point, this threshold value is determined respectively by four components of above-mentioned vector.
D), utilize experimental formula to obtain and the incoherent normalization coupling of minutiae point number fractional computation formula according to the details n that counts on the coupling in two templates:
S = P n &times; 2 , n < 100 P n &times; 250 / n , 100 < n < 500 P n &times; 102 / 256 500 < n
N wherein is the some logarithm on the coupling in the comparison, P nBe fingerprint coupling accumulative total weights.Distribute bigger weights at the matching ratio logarithm less than 100, and greater than 500 o'clock less weights coefficients of distribution, level and smooth thereby maintenance weights coefficient is tried one's best, draw final coupling mark S.
Coupling mark S and judgment threshold V compare, and the output judged result is judged and imported in fingerprint and the database whether same fingerprint of fingerprint.
Describing above is to be used to realize embodiments of the invention, it should be appreciated by those skilled in the art, in any modification or partial replacement that does not depart from the scope of the present invention, all belongs to claim of the present invention and comes restricted portion.

Claims (4)

1, the similar fingerprint identification method of a kind of twins based on partial structurtes is characterized in that: comprise step:
Pre-treatment step: the fingerprint image that collects is carried out pre-service, obtain the refined image of fingerprint;
Structure partial structurtes template step: extract crestal line tip and bifurcated detailed points P in the refined image, the set structure partial structurtes to the detailed points P extracted generate fingerprint template with each partial structurtes combination in the fingerprint;
Minutiae point partial structurtes coupling step: fingerprint template in input fingerprint partial structurtes template and the database is compared, and the relative position between the calculating partial structurtes and the difference of direction parameter judge whether fingerprint mates;
Described pre-treatment step also comprises:
Step 1.1 field of direction image extracts: adopt the gradient direction of analyzing gray scale point in the fingerprint and the relation of waiting to estimate direction, utilize the method for fuzzy logic to go to solve the shared ratio of four direction, utilize the value of mapping direction, estimate to obtain field of direction image;
Step 1.2 is cut apart field of direction image: remove the disorderly part of field of direction variation in the described field of direction image, form the mask image of fingerprint;
Step 1.3 trend pass filtering: utilize one based on the template of parallelogram with utilize sliding window operation algorithm, to the filtering of mask image travel direction with directivity;
Step 1.4 trend pass filtering image thinning: adopt perpendicular to the x direction of principal axis or perpendicular to the thinning processing method under two kinds of situations of y direction, adopt line integral to carry out refinement to the direction filtering image: to exist for the crestal line direction by the section of crossing over crestal line
Figure C2006101134090002C1
Between line integral adopt perpendicular to the axial conversion of y, exist for the crestal line direction
Figure C2006101134090002C2
With Between situation adopt perpendicular to the axial conversion of x;
Described structure partial structurtes template step also comprises:
Details point set information according to refined image is extracted generates fingerprint template to be compared, when the fingerprint image of gathering has extracted minutiae point point set M={M k, 1≤k≤L}, wherein L is the number that this point is concentrated minutiae point; For any one minutiae point wherein, its eigenvector is
Figure C2006101134090002C4
Constitute the fingerprint basic templates by this eigenvector;
Described minutiae point partial structurtes coupling step also comprises:
Step 3.1 obtains the feature of each minutiae point, comprising minutiae point x, and y coordinate, direction θ and field of direction provincial characteristics ZOD, all minutiae point vector P iDo as a whole corresponding one by one with fingerprint image, P i=[(x, y), θ, ZOD];
The first coupling of step 3.2 fingerprint image: the minutiae point that application specific details point vector obtains mating is right, and generates the minutiae point structure;
Step 3.3 template matches: minutiae point structure in the fingerprint partial structurtes template and the minutiae point structure in the database fingerprint are mated, obtain two coupling arrays, one of them record coupling array position translation energy, another one coupling An arrayed recording angle rotating energy, the assembly average of calculating energy coupling array obtains translation parameters and rotation parameter is nOri:
nOri = &Sigma; i = 0 M PO i &times; i &Sigma; i = 0 M PO i ;
Wherein, i represents i minutiae point, and M is the maximal value that possible angle is rotated, and nOri is the quantity of angle rotation, PO iIt is the similarity weights array of the record i position deviation of ordering;
Minutiae point corresponding in step 3.4 pair fingerprint partial structurtes template and the database fingerprint template is to carrying out the secondary comparison, by integral position translation and angle rotation, minutiae point partial structurtes in the template are adjusted, relatively more corresponding minutiae point to and obtain mating mark.
2, the similar fingerprint identification method of twins as claimed in claim 1 is characterized in that, the local detail dot structure also comprises:
Utilize local detail point combination details of construction dot structure, the structure detail dot structure adopts vector set to be expressed as: Fl k=(d Ki, d Kj, θ Ki, θ Kj, Z Ki, Z Kj, n Ki, n Kj, t k, t i, t j) T, wherein, d Ki, d KjIt is respectively the Euclidean distance between minutiae point k and minutiae point i and the minutiae point j; θ Ki, θ KjBe vectorial ki, the angle of kj; Z Ki, Z KjBe that direction between minutiae point k and minutiae point i and the minutiae point j is poor; n Ki, n KjBe vectorial ki, kj the crestal line number of process; T is the type of each minutiae point.
3, the similar fingerprint identification method of twins as claimed in claim 1 is characterized in that, described calculating translation parameters and rotation parameter are: nOri = &Sigma; i = 0 M PO i &times; i &Sigma; i = 0 M PO i , Treat the overlapping region of image in comparison chart picture and the database according to translation parameters and rotation parameter and calibrate, to fingerprint partial structurtes template Fl kWith database fingerprint Fl ' kCarry out the right calculating of minutiae point in the structure, draw details on the coupling n that counts.
4, the similar fingerprint identification method of the twins based on partial structurtes according to claim 1 is characterized in that, described template matches mark also comprises:
According to the details n that counts on the coupling, utilize and the incoherent normalization coupling of minutiae point number fractional computation formula:
S = P n &times; 2 , n < 100 P n &times; 250 / n , 100 &le; n &le; 500 P n &times; 102 / 256 500 < n
N wherein is the some logarithm on the coupling in the comparison, P nBe fingerprint coupling accumulative total weights; Be used to keep the weights coefficient as far as possible level and smooth, draw final coupling mark S and select: during less than 100 pairs, distributing bigger weights is P at the matching ratio logarithm n* 2; When mating logarithm greater than 500 pairs, distributing less weights coefficient is P n* 102/256.
CN200610113409A 2006-09-27 2006-09-27 Based on similar fingerprint recognition system of the twins of partial structurtes and method Expired - Fee Related CN100576230C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN200610113409A CN100576230C (en) 2006-09-27 2006-09-27 Based on similar fingerprint recognition system of the twins of partial structurtes and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN200610113409A CN100576230C (en) 2006-09-27 2006-09-27 Based on similar fingerprint recognition system of the twins of partial structurtes and method

Publications (2)

Publication Number Publication Date
CN101154263A CN101154263A (en) 2008-04-02
CN100576230C true CN100576230C (en) 2009-12-30

Family

ID=39255911

Family Applications (1)

Application Number Title Priority Date Filing Date
CN200610113409A Expired - Fee Related CN100576230C (en) 2006-09-27 2006-09-27 Based on similar fingerprint recognition system of the twins of partial structurtes and method

Country Status (1)

Country Link
CN (1) CN100576230C (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101777128B (en) * 2009-11-25 2012-05-30 中国科学院自动化研究所 Fingerprint minutiae matching method syncretized to global information and system thereof
CN101782965B (en) * 2010-02-11 2012-05-23 上海点佰趣信息科技有限公司 Method for treating deformed fingerprint image
EP3232369B1 (en) * 2016-04-15 2021-06-16 Nxp B.V. Fingerprint authentication system and method
CN105975914B (en) * 2016-04-28 2018-12-28 东南大学 Three people's kinship method of discrimination between Mr. and Mrs and child based on linear combination feature
KR102397812B1 (en) 2016-08-23 2022-05-13 비자 인터네셔널 서비스 어소시에이션 Remote use of locally stored biometric authentication data
CN106557748A (en) * 2016-11-21 2017-04-05 成都艾德沃传感技术有限公司 Fingerprint recognition processing method and processing device
CN111506889B (en) * 2017-07-27 2023-05-02 创新先进技术有限公司 User verification method and device based on similar user group
CN109726568B (en) * 2018-12-12 2023-08-08 西安电子科技大学 Fingerprint encryption method based on fusion feature descriptors
CN110096954B (en) * 2019-03-21 2023-04-07 同济大学 Fingerprint identification method based on neural network
CN110913143B (en) * 2019-12-09 2021-04-02 Oppo广东移动通信有限公司 Image processing method, image processing device, storage medium and electronic equipment
CN111428701B (en) * 2020-06-10 2020-09-29 深圳市诺赛特系统有限公司 Small-area fingerprint image feature extraction method, system, terminal and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6125192A (en) * 1997-04-21 2000-09-26 Digital Persona, Inc. Fingerprint recognition system
CN1421815A (en) * 2001-11-29 2003-06-04 田捷 Fingerprint image enhancement method based on knowledge

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6125192A (en) * 1997-04-21 2000-09-26 Digital Persona, Inc. Fingerprint recognition system
CN1421815A (en) * 2001-11-29 2003-06-04 田捷 Fingerprint image enhancement method based on knowledge

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Fingerprint minutiae matching based on the local and globalstructures. Xudong Jiang and Wei-Yun Yau.Pattern recognition, 2000. Proceedings. 15th international conference on,Vol.2 . 2000 *
On-line fingerprint verification. Anil Jain, Lin Hong and Ruud Bolle.IEEE transactions on pattern analysis and machine intelligence,Vol.19 No.4. 1997 *

Also Published As

Publication number Publication date
CN101154263A (en) 2008-04-02

Similar Documents

Publication Publication Date Title
CN100576230C (en) Based on similar fingerprint recognition system of the twins of partial structurtes and method
CN101057248B (en) Fingerprint biomass machine
Chen et al. A new algorithm for distorted fingerprints matching based on normalized fuzzy similarity measure
CN101539993B (en) Multi-acquisition-instrument fingerprint crossing-matching method based on size scaling estimation
CN101145196B (en) Quick fingerprint identification method based on strange topology structure
CN102222216A (en) Identification system based on biological characteristics of fingerprints
CN111507206B (en) Finger vein identification method based on multi-scale local feature fusion
CN103020626A (en) Method and system for detecting elliptical target in image
CN101819629B (en) Supervising tensor manifold learning-based palmprint identification system and method
CN112367273A (en) Knowledge distillation-based flow classification method and device for deep neural network model
Wang et al. Fingerprint matching using orientationcodes and polylines
CN103246880A (en) Human face recognizing method based on multi-level local obvious mode characteristic counting
CN109657657A (en) A kind of mobile phone small area high-precision fingerprint identification method and system based on image procossing
CN101414350A (en) Fingerprint image matching method syncretizing multi-reference node integral justification and direction field
CN103714159A (en) Coarse-to-fine fingerprint identification method fusing second-level and third-level features
CN114842524A (en) Face false distinguishing method based on irregular significant pixel cluster
CN110532856A (en) A kind of face identification method of more algorithm fusions
CN110598522A (en) Identity comparison method based on face and palm print palm vein recognition
US20220335639A1 (en) Ellipse detection acceleration method based on generalized pascal mapping
CN103942572A (en) Method and device for extracting facial expression features based on bidirectional compressed data space dimension reduction
CN102819754A (en) Fingerprint score fusion system and method based on Sigmoid expansion
Soltane et al. Soft decision level fusion approach to a combined behavioral speech-signature biometrics verification
CN105528591B (en) Living body finger print recognition methods based on more quadrants coding
CN200972656Y (en) Device for identifying similar fingerprint of twins
CN103295007A (en) Feature dimension-reduction optimization method for Chinese character recognition

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20091230

Termination date: 20190927