CN102354367A - Fingerprint identification card and fingerprint identification method running on card - Google Patents
Fingerprint identification card and fingerprint identification method running on card Download PDFInfo
- Publication number
- CN102354367A CN102354367A CN2011102857487A CN201110285748A CN102354367A CN 102354367 A CN102354367 A CN 102354367A CN 2011102857487 A CN2011102857487 A CN 2011102857487A CN 201110285748 A CN201110285748 A CN 201110285748A CN 102354367 A CN102354367 A CN 102354367A
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- card
- minutiae point
- fingerprint identification
- comparing unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The invention relates to a fingerprint identification card and a fingerprint identification method running on the card. The fingerprint identification card mainly comprises a fingerprint acquisition module, a fingerprint verification module and a control module, wherein the fingerprint acquisition module is used for acquiring fingerprint information of a user by a fingerprint scanner on a data card; the fingerprint verification module is used for carrying out comparison on the acquired fingerprint information and a stored fingerprint template, then outputting a comparison result; and the control module is used for carrying out corresponding control on the data card according to the comparison result output by the fingerprint verification module. The method disclosed by the invention is implemented through the following steps of: acquiring fingerprint information by the fingerprint acquisition module of the fingerprint identification card; carrying out comparison on the acquired fingerprint information and a fingerprint characteristic template stored in a database, and then outputting a comparison result by the fingerprint verification module; and carrying out corresponding control on a data card. The card and method provided by the invention have the advantages that the identification effect can be still good when only a small RAM (random-access memory) is used, therefore, the method is suitable for running on various cards such as JAVA cards; the method can be running on chips with an RAM (only 12 KB bytes) and a basic frequency of 72 M; and the method can be running on a CPU (central processing unit) in a card so as to realize the fingerprint identification function of the inside of the card.
Description
Technical field
The present invention relates to the living things feature recognition field, particularly fingerprint identification technology, especially a kind of fingerprint recognition card is gone up the fingerprint identification method of operation with card.
Background technology
Biometrics identification technology is meant a kind of technology of utilizing human body biological characteristics to carry out authentication.Biological recognition system is that biological characteristic is taken a sample, and through the algorithm that extracts characteristic the biological characteristic of taking a sample out is changed into numerical characteristic, and the feature templates that further these characteristics combination is formed, and deposits in the database.When recognition system is carried out authentication; Recognition system is obtained on-the-spot biological characteristic, convert into numerical characteristic and with database in the feature templates deposited compare, calculate the similarity between the two; Determining whether coupling, thereby this people is accepted or refuses in decision.
Fingerprint is a kind of of biological characteristic, has unique, regeneration, non-repudiation, conveniently extracts, is easy to characteristics such as identification.At present fingerprint identification technology is proven technique in the biometrics identification technology, has been accepted and approval by the most of national government in the whole world, has been widely applied to fields such as government, army, bank, social welfare guarantee, ecommerce and safety guard.
Need resource very big in the existing algorithm for recognizing fingerprint, general RAM needs the 64K byte, and dominant frequency needs more than the 100M.Smartcard internal has CPU and RAM, can carry out identification, the means of payment, encrypt/decrypt, information storage function.Use very extensive at present.Because cost and volume restrictions, the cpu chip dominant frequency is all not high in the card, and the RAM capacity is less, only about 10KB ~ 20KB.Because the RAM that needs is excessive, has limited the application of existing algorithm for recognizing fingerprint in card.
Chinese patent is announced CN1217287C number and disclosed a kind of fingerprint identification method: unique point is by the x coordinate of unique point i in the fingerprint characteristic; The y coordinate; The fingerprint lines is in the angle theta of the tangent line and the x axle at unique point i place, and characterization point i is the lines crunode or the attribute representation of end points.When characteristic had m unique point, fingerprint on site characteristic chained list was made up of m*m element.The data structure of each element Lij is by the length of this line; The fingerprint lines is in the angle theta 1 of the tangent line and the line Lij at unique point i place; The fingerprint lines is in the angle theta 2 of the tangent line and the line Lij at unique point j place, and the synthesized attribute of attribute is represented between explanation line end points i and the j.As everyone knows: the expression scope of 1 byte is 0 ~ 255; And the scope of angle is 0 ~ 359 degree; Be the image of 400 pixels for secondary length and width; Distance range between 2 points is: 0 ~ 566 pixel; Length is by 2 byte representations; θ 1 is by 2 byte representations, and θ 2 is by 2 byte representations, and attribute is by 1 byte representation.1 element is made up of 7 bytes.If unique point number m is 50 o'clock, unit have 2500, takies 17500 bytes altogether.2 characteristics are compared each other and are then needed 2*m*m element, totally 35000 bytes.
Chinese patent announces CN100412883C number and also disclose a kind of fingerprint identification method: matching process also is based on the minutiae point line.Consider 2 minutiae point mi, the wire length of mj (2 byte representation): dij in the time of coupling; The angle ai of line and minutiae point direction (2 byte representation), bi (2 byte representation); The type t of two minutiae point (2 byte representation); Curvature c (1 byte representation); Ridge density g (1 byte representation); Such 1 line needs 10 byte representations.If 2 characteristics respectively have 50 minutiae point, then need 2*50*50*10, totally 50000 bytes.
The RAM that can see above algorithm for recognizing fingerprint needs is all very big, needs 30K at least, has limited the application of algorithm for recognizing fingerprint at embedded chip.In embedded chip, embedded RAM accounts for chip area very much, so big RAM just means expensive.On all kinds of cards, on the JAVA card, the RAM of inside chip only has 10KB ~ 20KB, upward compares in the chip so existing algorithm for recognizing fingerprint is difficult to be implemented in card.
Therefore, be necessary to carry out technological improvement to above-mentioned defective.
Summary of the invention
The object of the present invention is to provide a kind of cheaply at the inner fingerprint identification method of realizing fingerprint comparison of finger-print card and based on the fingerprint recognition card of this kind method.
The present invention solves the technical scheme that its technical matters adopts: this fingerprint recognition card mainly comprises: finger print acquisition module, be used for through the fingerprint scanner on the data card, and gather user's finger print information; Fingerprint authentication module, the finger print information that is used for gathering is compared with the fingerprint template of storage, exports comparison result; Control module is used for the comparison result according to fingerprint authentication module output, and data are sticked into capable control corresponding.
This fingerprint identification method; This method is gathered finger print information through the finger print acquisition module of fingerprint recognition card; Through fingerprint authentication module the finger print information and the database of fingerprint feature templates that collect are compared; Adopt the method for minutiae point comparison; Per two details Mi; The line of Mj; Be called 1 comparing unit; Comparing unit is by wire length d; The fingerprint lines is in the angle theta 1 of the tangent line and the line Lij at minutiae point i place; The differential seat angle θ 2 of minutiae point i and minutiae point j forms, and with comparison result output, data is sticked into capable control corresponding.
The length d of said line is d < dMAX (dMAX≤255 pixels).
Said angle theta 1 is 0 ~ 255 unit degree with differential seat angle θ 2.
The quantity of said comparing unit is defined as maximum M minutiae point and participates in comparison, and each minutiae point generates N comparing unit, and 1 fingerprint characteristic needs M*N comparing unit at most.
Said comparing unit does not comprise the minutiae point type information.
The present invention compared with prior art, useful effect is: can only use very little RAM, and recognition effect is still fine, is adapted on all kinds of cards, as moving on the JAVA card.Can move on the chip of 72M dominant frequency at 12KB byte RAM only.Can in card, move on the CPU, realize the fingerprint identification function in the card.
Description of drawings
Fig. 1 is the synoptic diagram of crunode for minutiae point;
Fig. 2 is the synoptic diagram of end points for minutiae point;
Fig. 3 is the comparing unit structural representation;
Fig. 4 was fixed as 50 o'clock for N, and different M is to the synoptic diagram of the influence of EER;
Fig. 5 was fixed as 50 o'clock for M, and different N is to the synoptic diagram of the influence of EER.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is described further:
This fingerprint recognition card of the present invention mainly comprises: finger print acquisition module, be used for through the fingerprint scanner on the data card, and gather user's finger print information; Fingerprint authentication module, the finger print information that is used for gathering is compared with the fingerprint template of storage, exports comparison result; Control module is used for the comparison result according to fingerprint authentication module output, and data are sticked into capable control corresponding.
This fingerprint identification method; Finger print acquisition module through the fingerprint recognition card is gathered finger print information; Through fingerprint authentication module the finger print information and the database of fingerprint feature templates that collect are compared; Adopt the method for minutiae point comparison; Per two details Mi; The line of Mj; Be called 1 comparing unit; Comparing unit is by wire length d; The fingerprint lines is in the angle theta 1 of the tangent line and the line Lij at minutiae point i place; The differential seat angle θ 2 of minutiae point i and minutiae point j forms, and with comparison result output, data is sticked into capable control corresponding.
Owing to always have noise in the fingerprint image acquisition, after through the fingerprint Processing Algorithm, the situation of fingerprint minutiae type error often appears, that is: be originally to be identified as crunode by end points, perhaps be originally to be identified as end points by crunode.Like Fig. 1 is 101_6 image among the DB1_B in the FVC2000 image library, and Fig. 2 is a 101_7 image among the DB1_B in the FVC2000 image library.Fig. 1 and Fig. 2 are same piece of fingers, and what the centre was irised out is the minutiae point of a correspondence, and the minutiae point of irising out looks it is a crunode in Fig. 1, and in Fig. 2, looks it is an end points.Because the type of error of minutiae point is bigger, so when the structure comparing unit, do not comprise the minutiae point type information.
Because finger is soft, the fingerprint image that collects has deformation, the part that distance is far away more, and deformation is big more, so wire length d is being limited, < dMAX (dMAX≤255), d only needs 1 byte to represent like this to limit d.Comprise wire length d information in the comparing unit.
Common angular range is 0 ~ 359 degree, needs 2 bytes to represent, will calculate angle in this method and standardize to 0 ~ 255 unit degree.That is, 360 degree are equally divided into 256 equal portions, 1 unit kilsyth basalt shows 1.40625 degree.Angle only needs 1 byte can represent that promptly angle theta 1 is 0 ~ 255 unit degree with differential seat angle θ 2 equal alligatoring like this.
As shown in Figure 3; 1 comparing unit is by wire length d (1 byte); The fingerprint lines is in the angle theta 1 (1 byte) of the tangent line and the line Lij at minutiae point i place, and the differential seat angle θ 2 (1 byte) that comparing unit also comprises minutiae point i and minutiae point j forms, and 1 comparing unit takies 3 bytes.
The quantity of restriction comparing unit is set maximum M minutiae point and is participated in comparison, and each minutiae point generates N comparing unit, and such 1 fingerprint characteristic needs M*N comparing unit at most.If M is 40, N is 15, and then 1 fingerprint characteristic needs i.e. 1800 bytes of 40*15*3.2 fingerprint characteristic comparisons need 3600 bytes.The comparing unit information that in the time of comparison, needs only takies 3600 bytes like this, because the quantity of comparing unit has descended, the operand that needs also descends thereupon, promptly can use lower dominant frequency.
Experimental technique explanation: select a fingerprint base, comprise 200 fingers { F0, F1, F2 ... F199}, wherein each finger has 10 characteristics { T0, T1, T2 ... T9}.
When refusing really to test: select 2 different characteristic: Ta, the Tb of same finger F x to compare, record comparison score.Compare 200*10*9=18000 time altogether.
When recognizing false the test: select 2 characteristic: Ta, the Tb of different finger F x, Fy to compare, record comparison score.Compare 200*199*10*10=3980000 time altogether.
Join shown in Figure 4ly, N was fixed as 50 o'clock, and different M is to the influence of EER:
|
15 | 20 | 25 | 30 | 35 | 40 | 45 | 50 |
EER (% of unit) | 2.364 | 1.3 | 1.096 | 1.056 | 1.028 | 0.997 | 1.007 | 1.007 |
Join shown in Figure 5ly, M was fixed as 50 o'clock, and different N is to the influence of EER:
|
15 | 20 | 25 | 30 | 35 | 40 | 45 | 50 |
EER (% of unit) | 0.997 | 1.008 | 0.998 | 1.008 | 1.007 | 1.007 | 1.007 | 1.007 |
And experiment further draws, and different M, N are to the influence of EER, FRR:
|
50 | 30 | 50 | 40 |
|
50 | 30 | 15 | 15 |
EER (% of unit) | 1.007 | 1.056 | 0.997 | 0.998 |
FRR (% of unit) | 4.0667 | 4.2 | 4.25 | 4.26 |
Can be known by last table: when M, N are reduced to 40,15 by 50,50 respectively, test and actually refuse sincere (ERR) and only bring up to 4.260% from 4.0667%, promptly usability only descends 4.47%, and speed improves more than 30%, and the RAM that needs is reduced to former about 1/10.Actual this algorithm for recognizing fingerprint of test can the embedded chip in the card (the 72M dominant frequency, 12KBRAM) in the realization fingerprint identification function, average specific to 2 characteristic times in 30mS.
Through analyzing, rationally reduce the right information of minutiae point of participating in comparing, reach and take the RAM size when reducing to compare, and performance does not have the purpose of greater loss.Rationally reduce the right number of minutiae point of participating in comparing, reach and take the RAM size when reducing to compare, and performance does not have the purpose of greater loss.
Terminological interpretation:
1, EER: be the abbreviation that equates error rate (Equal Error Rate), it is an accuracy of system identification (FAR) and refuse sincere (FRR) error rate when equal.
2, FAR: accuracy of system identification.
3, FRR: refuse sincere.
Except that the foregoing description, the present invention can also have other embodiments.All employings are equal to the technical scheme of replacement or equivalent transformation formation, all drop on the protection domain of requirement of the present invention.
Claims (6)
1. fingerprint recognition card is characterized in that: mainly comprise:
Finger print acquisition module is used for through the fingerprint scanner on the data card, gathers user's finger print information;
Fingerprint authentication module, the finger print information that is used for gathering is compared with the fingerprint template of storage, exports comparison result;
Control module is used for the comparison result according to fingerprint authentication module output, and data are sticked into capable control corresponding.
2. fingerprint identification method that adopts fingerprint recognition card as claimed in claim 1; It is characterized in that: the finger print acquisition module by the fingerprint recognition card is gathered finger print information; By fingerprint authentication module the finger print information and the database of fingerprint feature templates that collect are compared; Adopt the method for minutiae point comparison; The line of per two details Mi, Mj; Be called 1 comparing unit; Comparing unit is by wire length d; The fingerprint lines is in the angle theta 1 of the tangent line and the line Lij at minutiae point i place; The differential seat angle θ 2 of minutiae point i and minutiae point j forms; With comparison result output, data are sticked into capable control corresponding.
3. fingerprint identification method according to claim 2 is characterized in that: the length d of said line is d < dMAX, dMAX≤255 pixels.
4. fingerprint identification method according to claim 2 is characterized in that: said angle theta 1 is 0 ~ 255 unit degree with differential seat angle θ 2.
5. fingerprint identification method according to claim 2 is characterized in that: the quantity of said comparing unit is defined as maximum M minutiae point and participates in comparison, and each minutiae point generates N comparing unit, and 1 fingerprint characteristic needs M*N comparing unit at most.
6. according to claim 2 or 5 described fingerprint identification methods, it is characterized in that: said comparing unit does not comprise the minutiae point type information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102857487A CN102354367A (en) | 2011-09-23 | 2011-09-23 | Fingerprint identification card and fingerprint identification method running on card |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102857487A CN102354367A (en) | 2011-09-23 | 2011-09-23 | Fingerprint identification card and fingerprint identification method running on card |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102354367A true CN102354367A (en) | 2012-02-15 |
Family
ID=45577929
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2011102857487A Pending CN102354367A (en) | 2011-09-23 | 2011-09-23 | Fingerprint identification card and fingerprint identification method running on card |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102354367A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105447447A (en) * | 2015-11-11 | 2016-03-30 | 广东欧珀移动通信有限公司 | Fingerprint identification method and fingerprint identification system of terminal |
CN105634733A (en) * | 2015-07-24 | 2016-06-01 | 宇龙计算机通信科技(深圳)有限公司 | Encryption method and system, decryption method and system and terminal |
CN106485240A (en) * | 2016-10-28 | 2017-03-08 | 南京信息职业技术学院 | System for monitoring state of resident identification card and monitoring method thereof |
CN108090341A (en) * | 2017-12-15 | 2018-05-29 | 深圳市文鼎创数据科技有限公司 | Java card control method and java card |
CN113240419A (en) * | 2021-05-12 | 2021-08-10 | 前海联大(深圳)技术有限公司 | Method for using digital currency safe storage |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101414351A (en) * | 2008-11-03 | 2009-04-22 | 章毅 | Fingerprint recognition system and control method |
CN101777130A (en) * | 2010-01-22 | 2010-07-14 | 北京大学 | Method for evaluating similarity of fingerprint images |
-
2011
- 2011-09-23 CN CN2011102857487A patent/CN102354367A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101414351A (en) * | 2008-11-03 | 2009-04-22 | 章毅 | Fingerprint recognition system and control method |
CN101777130A (en) * | 2010-01-22 | 2010-07-14 | 北京大学 | Method for evaluating similarity of fingerprint images |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105634733A (en) * | 2015-07-24 | 2016-06-01 | 宇龙计算机通信科技(深圳)有限公司 | Encryption method and system, decryption method and system and terminal |
CN105634733B (en) * | 2015-07-24 | 2019-01-15 | 宇龙计算机通信科技(深圳)有限公司 | Encryption method and system, decryption method and system and terminal |
CN105447447A (en) * | 2015-11-11 | 2016-03-30 | 广东欧珀移动通信有限公司 | Fingerprint identification method and fingerprint identification system of terminal |
CN106485240A (en) * | 2016-10-28 | 2017-03-08 | 南京信息职业技术学院 | System for monitoring state of resident identification card and monitoring method thereof |
CN108090341A (en) * | 2017-12-15 | 2018-05-29 | 深圳市文鼎创数据科技有限公司 | Java card control method and java card |
CN113240419A (en) * | 2021-05-12 | 2021-08-10 | 前海联大(深圳)技术有限公司 | Method for using digital currency safe storage |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108229427B (en) | Identity security verification method and system based on identity document and face recognition | |
Gunasekaran et al. | Deep multimodal biometric recognition using contourlet derivative weighted rank fusion with human face, fingerprint and iris images | |
CN106778457A (en) | The fingerprint identification method and system of fingerprint recognition rate can be improved | |
CN104580143A (en) | Security authentication method based on gesture recognition, terminal, server and system | |
Kaur et al. | Signature identification and verification techniques: state-of-the-art work | |
CN102354367A (en) | Fingerprint identification card and fingerprint identification method running on card | |
CN103077459A (en) | Method for carrying out living body authentication and payment by fusing multi-biometric features of user | |
CN103595538A (en) | Identity verification method based on mobile phone acceleration sensor | |
CN101556713A (en) | Application process of fingerprint identification technology on POS machine, imprinter and cash dispenser | |
CN110163123A (en) | One kind referring to vein fusion identification method based on single width near-infrared finger-image fingerprint | |
Appati et al. | Implementation of a Transform‐Minutiae Fusion‐Based Model for Fingerprint Recognition | |
Pal et al. | Non-English and non-Latin signature verification systems: a survey | |
Zhao et al. | Fingerprint pre-processing and feature engineering to enhance agricultural products categorization | |
Khoh et al. | Score level fusion approach in dynamic signature verification based on hybrid wavelet‐Fourier transform | |
Garhawal et al. | A study on handwritten signature verification approaches | |
Kumar et al. | Genuine and forged offline signature verification using back propagation neural networks | |
Akulwar et al. | Secured multi modal biometric system: a review | |
Liu et al. | Feature guided fingerprint pore matching | |
CN112116012A (en) | Finger vein instant registration and identification method and system based on deep learning | |
Huang et al. | Axially-enhanced local attention network for finger vein recognition | |
Gona et al. | Multimodal biometric reorganization system using deep learning convolutional neural network | |
CN103593660A (en) | Palm print recognition method based on cross gradient encoding of image with stable characteristics | |
CN106339678A (en) | Fingerprint image representation method based on a variety of feature points | |
CN1825338A (en) | Embedded sliding fingerprint acquiring and identifying module | |
Adamski et al. | Signature verification by only single genuine sample in offline and online systems |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C53 | Correction of patent of invention or patent application | ||
CB02 | Change of applicant information |
Address after: The city of Hangzhou in West Zhejiang province 311121 No. 998 Building 9 East Sea Park Applicant after: Hangzhou Shengyuan Chip Technique Co., Ltd. Address before: 310012, room 17, building 176, 203 Tianmu Mountain Road, Hangzhou, Zhejiang, Xihu District Applicant before: Hangzhou Shengyuan Chip Technique Co., Ltd. |
|
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120215 |