CN104933411A - Fingerprint identification processing method and system - Google Patents
Fingerprint identification processing method and system Download PDFInfo
- Publication number
- CN104933411A CN104933411A CN201510332833.2A CN201510332833A CN104933411A CN 104933411 A CN104933411 A CN 104933411A CN 201510332833 A CN201510332833 A CN 201510332833A CN 104933411 A CN104933411 A CN 104933411A
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- fingerprint template
- compared
- characteristic
- result
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/13—Sensors therefor
- G06V40/1306—Sensors therefor non-optical, e.g. ultrasonic or capacitive sensing
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a fingerprint identification processing method and system. The method comprises the following steps that: a programmable device performs parallel calculation on the feature vector of a fingerprint template in a finger template database after reception of a fingerprint authentication request sent by a processor; the programmable device compares a parallel calculation result with the feature vector calculation result of a fingerprint to be compared, and screens fingerprint template data which is accordant with a preset rule; and a processor performs feature similarity score calculation on fingerprint data to be compared and the screened fingerprint template data, and acquires a fingerprint identification operation result from a calculation result. Through adoption of the method and system, time consumed by a processor processing in a fingerprint identification algorithm can be effectively shortened, and the operation efficiency is increased.
Description
Technical field
The present invention relates to computer realm, particularly relate to a kind of fingerprint recognition disposal route and system.
Background technology
Fingerprint is the important biomolecule feature of human body.From nineteen sixties, some countries such as the U.S., Britain, France etc. start the development to automatic system of fingerprint recognition.Along with going deep into of application, the capacity in finger print information storehouse also constantly rises.Such as China starts employing fingerprint automatic recognition system from nineteen ninety, has stamp finger print data more than 3,000 ten thousand parts to the whole nation in 2005, every year with more than 20% speed increment, 2010, national stamp finger print data reached more than 7,000 ten thousand parts.The fingerprint storage capacity of quick growth brings new challenge to the quick calculating of fingerprint.
Common algorithm for recognizing fingerprint is all the recognition principle of feature based Point matching, judge whether local message in two width fingerprint images coincide to gauge rod difference by comparing between any two points, because each finger all has 20-128 characteristic, the data that such needs compare are just close to the factorial of characteristic, this calculated amount obviously cannot practical application, for this reason, fingerprint worker have employed a large amount of mathematical algorithm and reduces the complexity calculated, the time complexity of existing algorithm for recognizing fingerprint can be reduced to characteristic number square.
In actual applications, at present by general processor as arm processor carries out fingerprint recognition.But, when adopting existing arm processor to process algorithm for recognizing fingerprint, exist consuming time longer, the problem that operation efficiency is low.
Summary of the invention
The problem that the embodiment of the present invention solves is the time that reduction arm processor expends at process algorithm for recognizing fingerprint, improves operation efficiency.
For solving the problem, the embodiment of the present invention provides a kind of fingerprint recognition disposal route, comprising:
Programming device, after the finger print identifying request receiving processor transmission, carries out parallel computation to the proper vector of fingerprint template in fingerprint template database; The proper vector result of calculation of parallel computation result and fingerprint to be compared is compared, filters out the fingerprint template data meeting preset rules;
Described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value by described processor, obtain fingerprint recognition operation result from result of calculation.
Optionally, described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value, comprising: by multithreading, the characteristic similarity score value that described finger print data to be compared and the described fingerprint template data filtered out carry out walking abreast is calculated.
Optionally, described the proper vector of fingerprint template in fingerprint template database is carried out parallel computation, comprise: from described fingerprint template database, choose fingerprint template concurrently, calculate length and the angle of the proper vector bar in each fingerprint template characteristic of correspondence point set between any two unique points, obtain parallel computation result, described parallel computation result comprises: the length of each fingerprint template characteristic of correspondence to gauge rod and the set of angle.
Optionally, the proper vector result of calculation of described fingerprint to be compared is: in the unique point set of described fingerprint to be compared, the length of proper vector bar between any two unique points and the set of angle.
Optionally, describedly filter out the fingerprint template data meeting preset rules, comprise: by fingerprint template characteristic of correspondence to the element in the length of gauge rod and the set of angle, contrast to the element in the length of gauge rod and the set of angle with fingerprint characteristic of correspondence to be compared, filter out the fingerprint template that set similarity is greater than preset value.
Optionally, describedly from result of calculation, obtain fingerprint recognition operation result, comprise: characteristic similarity score value will be calculated and predetermined threshold value compares, when described characteristic similarity score value is greater than described predetermined threshold value, judge that corresponding fingerprint and described fingerprint to be compared are as same fingerprint.
Optionally, described programming device is FPGA, and described processor is arm processor.
Optionally, the fingerprint template data in described default fingerprint template database are stored in multiple storer respectively.
Optionally, described storer is DRAM.
The embodiment of the present invention additionally provides a kind of fingerprint recognition disposal system, comprising: programming device and processor, wherein:
Described programming device, is suitable for, after the finger print identifying request receiving processor transmission, carrying out parallel computation to the proper vector of fingerprint template in fingerprint template database; The proper vector result of calculation of parallel computation result and fingerprint to be compared is compared, filters out the fingerprint template data meeting preset rules;
Described processor, be suitable for receiving the fingerprint template data meeting preset rules described in the transmission of described programming device, and described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value, from result of calculation, obtain fingerprint recognition operation result.
Optionally, described programming device, is suitable for being calculated by the characteristic similarity score value that described finger print data to be compared and the described fingerprint template data filtered out carry out walking abreast by multithreading.
Optionally, described programming device, be suitable for choosing fingerprint template concurrently from described fingerprint template database, calculate length and the angle of the proper vector bar in each fingerprint template characteristic of correspondence point set between any two unique points, obtain parallel computation result, described parallel computation result comprises: the length of each fingerprint template characteristic of correspondence to gauge rod and the set of angle.
Optionally, the proper vector result of calculation of the fingerprint to be compared that described programming device gets is: in the unique point set of described fingerprint to be compared, the length of proper vector bar between any two unique points and the set of angle.
Optionally, described programming device, be suitable for fingerprint template characteristic of correspondence to the element in the length of gauge rod and the set of angle, contrast to the element in the length of gauge rod and the set of angle with fingerprint characteristic of correspondence to be compared, filter out the fingerprint template that set similarity is greater than preset value.
Optionally, described processor, is suitable for calculating characteristic similarity score value and predetermined threshold value compares, and when described characteristic similarity score value is greater than described predetermined threshold value, judges that corresponding fingerprint and described fingerprint to be compared are as same fingerprint.
Optionally, described programming device is FPGA, and described processor is arm processor.
Optionally, described fingerprint recognition disposal system comprises: multiple storer, is suitable for storing the fingerprint template data in described default fingerprint template database.
Optionally, described storer is DRAM.
Compared with prior art, the technical scheme of the embodiment of the present invention has the following advantages:
By programming device, the proper vector of the proper vector of fingerprint to be compared and fingerprint template is carried out parallel computation, from fingerprint template database, filter out legal fingerprint template data.Processor only carries out the calculating of characteristic similarity score value to the fingerprint template data filtered out and fingerprint to be compared, thus greatly can reduce the quantity of the fingerprint template carrying out the calculating of characteristic similarity score value, therefore greatly can reduce the time that algorithm for recognizing fingerprint expends, improve operation efficiency.
Further, the fingerprint template data scatter in fingerprint template database is stored in multiple storer, the structural memory source of existing hardware can be effectively utilized, reduce the acquisition cost of large internal memory.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of a kind of fingerprint recognition disposal route in the embodiment of the present invention;
Fig. 2 is the structural representation of a kind of fingerprint recognition disposal system in the embodiment of the present invention;
Fig. 3 is the structural representation of the another kind of fingerprint recognition disposal system in the embodiment of the present invention.
Embodiment
In actual applications, at present by general processor as arm processor carries out fingerprint recognition.But, when adopting existing arm processor to process algorithm for recognizing fingerprint, exist consuming time longer, the problem that operation efficiency is low.
In embodiments of the present invention, by programming device, the proper vector of the proper vector of fingerprint to be compared and fingerprint template is carried out parallel computation, from fingerprint template database, filter out legal fingerprint template data.Processor only carries out the calculating of characteristic similarity score value to the fingerprint template data filtered out and fingerprint to be compared, thus greatly can reduce the quantity of the fingerprint template carrying out the calculating of characteristic similarity score value, therefore greatly can reduce the time that algorithm for recognizing fingerprint expends, improve operation efficiency.
For enabling above-mentioned purpose, the feature and advantage of the embodiment of the present invention more become apparent, below in conjunction with accompanying drawing, specific embodiments of the invention are described in detail.
With reference to Fig. 1, embodiments provide a kind of fingerprint recognition disposal route, be described in detail below by way of concrete steps.
Step S101, programming device, after the finger print identifying request receiving processor transmission, carries out parallel computation to the proper vector of fingerprint template in fingerprint template database.
In concrete enforcement, programming device is field programmable gate array (Field Programmable GateArray, FPGA).FPGA is the product further developed on the basis of the programming devices such as PAL, GAL, EPLD, for carrying out concurrent operation.
In concrete enforcement, when carrying out fingerprint recognition arithmetic operation, processor can send finger print identifying request to programming device.Programming device, after the finger print identifying request receiving processor transmission, reads fingerprint template data in the fingerprint template database preset, calculate concurrently, to obtain parallel computation result to the proper vector of the fingerprint template read out from storer.
In embodiments of the present invention, can there is multiple computing unit in programming device, when carrying out parallel computation, each computing unit can calculate each self-corresponding fingerprint template data.When distributing fingerprint template data, the fingerprint template data that each computing unit is corresponding are different, i.e. fingerprint template data not duplicate allocation.
In embodiments of the present invention, the fingerprint template data in fingerprint template database can be stored in advance in dynamic RAM (Dynamic Random Access Memory, DRAM).Programming device, after the finger print identifying request receiving processor transmission, continuously reads fingerprint template data and is loaded into, calculating concurrently to the proper vector of the fingerprint template read out, having obtained parallel computation result from DRAM.
After programming device reads fingerprint template data, the calculating of proper vector bar can be carried out to each fingerprint template, calculate length and the angle of the proper vector bar in each fingerprint template characteristic of correspondence point set between any two unique points, all fingerprint template characteristics of correspondence are parallel computation result to the combination of gauge rod result of calculation.
Can learn from above-mentioned computation process, when in the set of fingerprint template characteristic of correspondence point, element is more, the element that the proper vector bar result of calculation obtained comprises is more.When being 20 with the element number in unique point set, calculate length and the angle of the proper vector bar between any two unique points, the result obtained is (20 × 19)/2=190, and namely a fingerprint template characteristic of correspondence exists 190 elements in gauge rod result of calculation.Therefore, for reducing calculated amount, when carrying out proper vector bar and calculating, length and the angle of the feature bar vector between adjacent two elements only can be calculated.Also other method can be adopted to reduce the calculated amount of feature bar vector calculation, do not repeat herein.
Step S102, compares the proper vector result of calculation of concurrent operation result and fingerprint to be compared, sifts out the fingerprint template data meeting preset rules.
In concrete enforcement, finger print information to be compared can be stored in programming device.The proper vector result of calculation of fingerprint to be compared can be: the proper vector bar result of calculation of fingerprint to be compared, namely calculate in fingerprint characteristic of correspondence point to be compared set, the length of the proper vector bar between any two unique points and angle, the set obtained is the proper vector bar result of calculation of fingerprint to be compared.
In embodiments of the present invention, calculating fingerprint characteristic of correspondence to be compared after gauge rod result of calculation, itself and fingerprint template characteristic of correspondence can compared to gauge rod result of calculation, to filter out partial fingerprint template.The principle of screening can be: by fingerprint characteristic of correspondence to be compared to gauge rod result of calculation, carry out Similarity Measure with fingerprint template characteristic of correspondence to gauge rod result of calculation, filter out the fingerprint template that similarity is greater than preset value, and be sent to processor.That is: according to fingerprint characteristic of correspondence to be compared to the element in the length of gauge rod and the set of angle, with fingerprint template characteristic of correspondence to the element in the length of gauge rod and the set of angle, Similarity Measure is carried out to the two.
In embodiments of the present invention, gather similar can being expressed as two: the element number in two set is equal, the element one_to_one corresponding in two set, when one of them element in two set, and when the difference one to one between element is in certain limit, judge that two elements are similar; When the number of similar element is greater than certain value in two set, judge that two set are similar.
Such as, fingerprint characteristic of correspondence to be compared is { a0, a1 to the element in gauge rod result of calculation, a2 ..., an}, one of them fingerprint template characteristic of correspondence is { b0, b1, b2 to the element in gauge rod result of calculation, bn}, a0 and b0 one_to_one corresponding, a1 and b1 one_to_one corresponding,, an and bn one_to_one corresponding.The difference of a0 and b0 is in certain limit, judges that a0 and b0 is similar.The difference of a1 and b1 is greater than certain limit, judges that a1 and b1 is dissimilar.According to the number of similar element, the similarity of two set can be obtained, thus judge whether two set are similar.
Step S103, described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value by described processor, obtain fingerprint recognition operation result from result of calculation.
In concrete enforcement, the legal fingerprint template data filtered out, after filtering out legal fingerprint template data, can be sent to processor by programming device.Itself and finger print data to be compared, according to after the fingerprint template data filtered out received, are carried out the calculating of characteristic similarity score value by processor.When the characteristic similarity score value calculated is greater than predetermined threshold value, namely judge that characteristic similarity score value is greater than the fingerprint template of predetermined threshold value and fingerprint to be compared as same fingerprint.
In embodiments of the present invention, processor can be arm processor.The fingerprint template data filtered out and finger print data to be compared can be carried out accurate comparing calculation by arm processor, and to obtain characteristic similarity score value, accurate comparing calculation can comprise the generation of feature pole pair, rotation, translation and Similarity Measure etc.Arm processor can set up precise alignment thread, carries out the calculating of characteristic similarity score value.
Such as, the number that programming device filters out legal fingerprint template data is 20, and 20 fingerprint template data are sent to arm processor.Arm processor carries out precise alignment calculating to the fingerprint template data that finger print data to be compared and 20 filter out, learn that the characteristic similarity score value of fingerprint template A and fingerprint to be compared is 90%, be greater than predetermined threshold value 80%, therefore, judge to there is the fingerprint template A identical with fingerprint to be compared in fingerprint template database.
In actual applications, the data volume of fingerprint template database may be comparatively large, and the storage space taken is more, therefore higher to the requirement of the storage space of DRAM.Under normal circumstances, use the DRAM of larger storage space to store data in fingerprint template database, cost is higher.
And in embodiments of the present invention, fingerprint template data can be stored in multiple DRAM respectively according to certain rule.Can divide according to the order of fingerprint template, also can divide according to the type of fingerprint template, other classified types can also be there is, do not repeat herein.
Such as, divide according to the order of fingerprint template, the fingerprint template data that sequence is designated 1 ~ n are stored in DRAM1, the fingerprint template data that sequence is designated n+1 ~ 2n are stored in DRAM2.Programming device read fingerprint template data time, can successively from DRAM1, DRAM2 ..., read fingerprint template data in DRAMn.
Therefore, adopt the scheme in the embodiment of the present invention, the storage capacity requirement to DRAM can be reduced, thus reduce costs.
In actual applications, also the fingerprint template data in fingerprint template database can be stored in different computing nodes, by dispatch server, fingerprint comparison task be sent to multiple computing node.Each computing node after calculating completes by result feedback to dispatch server, dispatch server according to feedback result, obtain final fingerprint recognition operation result.
General processor, such as arm processor, it is mainly used in solving complicated logical process process, and branch decision computing has advantage.And the logical relation of algorithm for recognizing fingerprint is simple, operand is large, data correlation degree is less, when adopting arm processor to carry out computing, there is longer, problem that operation efficiency is low consuming time.Therefore, in prior art, usually use parallel computation to improve operation efficiency.
Usual employing FPGA carries out computing to algorithm for recognizing fingerprint, and operation efficiency is high, stable performance and low in energy consumption.But, because the complexity of algorithm for recognizing fingerprint is higher, complete algorithm for recognizing fingerprint is all transplanted in FPGA, the requirement of the logic gate number magnitude of FPGA is significantly improved, causes cost increase.
In an embodiment of the present invention, adopt FPGA to carry out the larger calculating of operand, such as, proper vector bar generates matching operation, and adopt arm processor to carry out the less calculating of operand, such as, fingerprint template and fingerprint characteristic similarity score to be compared calculate.By the Synchronous data dispose of FPGA and ARM, the processing speed of algorithm for recognizing fingerprint can be accelerated.
In embodiments of the present invention, when carrying out fingerprint recognition computing, first the fingerprint template in fingerprint template database is screened, filter out the fingerprint template tentatively meeting fingerprint feature information to be compared, the fingerprint template data making processor will carry out processing greatly reduce, and effectively improve operation efficiency.
As can be seen here, in embodiments of the present invention, by programming device, the proper vector of the proper vector of fingerprint to be compared and fingerprint template is carried out parallel computation, from fingerprint template database, filter out legal fingerprint template data.Processor only carries out the calculating of characteristic similarity score value to the fingerprint template data filtered out and fingerprint to be compared, thus greatly can reduce the quantity of the fingerprint template carrying out the calculating of characteristic similarity score value, therefore greatly can reduce the time that algorithm for recognizing fingerprint expends, improve operation efficiency.
With reference to Fig. 2, embodiments provide a kind of fingerprint recognition disposal system 20, comprise programming device 201 and processor 202, wherein:
Described programming device 201, is suitable for, after the finger print identifying request receiving processor transmission, the proper vector of fingerprint template in fingerprint template database being carried out parallel computation; The proper vector result of calculation of parallel computation result and fingerprint to be compared is compared, filters out the fingerprint template data meeting preset rules;
Described processor 202, be suitable for receiving the fingerprint template data meeting preset rules described in the transmission of described programming device 201, and described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value, from result of calculation, obtain fingerprint recognition operation result.
In concrete enforcement, described programming device 201, is suitable for being calculated by the characteristic similarity score value that described finger print data to be compared and the described fingerprint template data filtered out carry out walking abreast by multithreading.
In concrete enforcement, described programming device 201, be suitable for choosing fingerprint template concurrently from described fingerprint template database, calculate length and the angle of the proper vector bar in each fingerprint template characteristic of correspondence point set between any two unique points, obtain parallel computation result, described parallel computation result comprises: the length of each fingerprint template characteristic of correspondence to gauge rod and the set of angle.
In concrete enforcement, the proper vector result of calculation of the fingerprint to be compared that described programming device 201 gets is: in the unique point set of described fingerprint to be compared, the length of proper vector bar between any two unique points and the set of angle.
In concrete enforcement, described programming device 201, be suitable for fingerprint template characteristic of correspondence to the element in the length of gauge rod and the set of angle, contrast to the element in the length of gauge rod and the set of angle with fingerprint characteristic of correspondence to be compared, filter out the fingerprint template that set similarity is greater than preset value.
In concrete enforcement, described processor 202, is suitable for calculating characteristic similarity score value with predetermined threshold value compares, and when described characteristic similarity score value is greater than described predetermined threshold value, judges that the fingerprint of correspondence and described fingerprint to be compared are as same fingerprint.
In concrete enforcement, described programming device is FPGA, and described processor is arm processor.
In concrete enforcement, fingerprint recognition disposal system 20 can comprise: multiple storer, is suitable for storing the fingerprint template data in described default fingerprint template database.
In concrete enforcement, described storer can be DRAM.
With reference to Fig. 3, give the structural representation of a kind of fingerprint recognition disposal system in the embodiment of the present invention, comprising: FPGA hardware 301, arm processor 302 and DRAM storer 303, wherein:
Fingerprint template database is stored in DRAM storer 303;
FPGA hardware 301 comprises: fingerprint template reads in unit 3011, fingerprint template storage unit 3012, fingerprint template allocation units 3013, multiple proper vector bar computing unit 3014, screening unit 3015 and finger print data storage unit 3016 to be compared;
Arm processor 302 comprises: multiple score value computing unit 3021 and finger print identifying request unit 3022.
Below the workflow of the fingerprint recognition disposal system provided in above-described embodiment is described.
Arm processor 302 sends finger print identifying request by finger print identifying request unit 3022 to FPGA hardware 301.
Fingerprint feature information to be compared in finger print data storage unit 3016 to be compared, after receiving finger print identifying request, is sent to multiple proper vector bar computing unit 3014 by FPGA hardware 301.Control fingerprint template to read in unit 3011 from DRAM storer 303, read data in fingerprint template database, and be stored in fingerprint template storage unit 3012.By fingerprint template allocation units 3013, the fingerprint template database data stored in fingerprint template storage unit 3012 is assigned to multiple proper vector bar computing unit 3014.Multiple proper vector bar computing unit 3014 is according to fingerprint feature information to be compared, and fingerprint template database data, walks abreast and carries out the calculating of proper vector bar.
Screened by the result of calculation of screening unit 3015 to multiple proper vector bar computing unit 3014, and the selection result is sent to multiple score value computing unit 3021.Cross multiple score value computing unit 3021, the characteristic similarity score value of the fingerprint template that calculating sifting goes out and template to be compared, thus obtain the result of fingerprint recognition.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is that the hardware that can carry out instruction relevant by program has come, this program can be stored in a computer-readable recording medium, and storage medium can comprise: ROM, RAM, disk or CD etc.
Although the present invention discloses as above, the present invention is not defined in this.Any those skilled in the art, without departing from the spirit and scope of the present invention, all can make various changes or modifications, and therefore protection scope of the present invention should be as the criterion with claim limited range.
Claims (18)
1. a fingerprint recognition disposal route, is characterized in that, comprising:
Programming device, after the finger print identifying request receiving processor transmission, carries out parallel computation to the proper vector of fingerprint template in fingerprint template database; The proper vector result of calculation of parallel computation result and fingerprint to be compared is compared, filters out the fingerprint template data meeting preset rules;
Described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value by described processor, obtain fingerprint recognition operation result from result of calculation.
2. fingerprint recognition disposal route as claimed in claim 1, it is characterized in that, described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value, comprising: by multithreading, the characteristic similarity score value that described finger print data to be compared and the described fingerprint template data filtered out carry out walking abreast is calculated.
3. fingerprint recognition disposal route as claimed in claim 1, it is characterized in that, described the proper vector of fingerprint template in fingerprint template database is carried out parallel computation, comprise: from described fingerprint template database, choose fingerprint template concurrently, calculate length and the angle of the proper vector bar in each fingerprint template characteristic of correspondence point set between any two unique points, obtain parallel computation result, described parallel computation result comprises: the length of each fingerprint template characteristic of correspondence to gauge rod and the set of angle.
4. fingerprint recognition disposal route as claimed in claim 3, it is characterized in that, the proper vector result of calculation of described fingerprint to be compared is: in the unique point set of described fingerprint to be compared, the length of proper vector bar between any two unique points and the set of angle.
5. fingerprint recognition disposal route as claimed in claim 4, it is characterized in that, describedly filter out the fingerprint template data meeting preset rules, comprise: by fingerprint template characteristic of correspondence to the element in the length of gauge rod and the set of angle, contrast to the element in the length of gauge rod and the set of angle with fingerprint characteristic of correspondence to be compared, filter out the fingerprint template that set similarity is greater than preset value.
6. fingerprint recognition disposal route as claimed in claim 1, is characterized in that, describedly from result of calculation, obtains fingerprint recognition operation result, comprising:
Characteristic similarity score value will be calculated and predetermined threshold value compares, when described characteristic similarity score value is greater than described predetermined threshold value, judge that corresponding fingerprint and described fingerprint to be compared are as same fingerprint.
7. fingerprint recognition disposal route as claimed in claim 1, it is characterized in that, described programming device is FPGA, and described processor is arm processor.
8. fingerprint recognition disposal route as claimed in claim 1, it is characterized in that, the fingerprint template data in described default fingerprint template database are stored in multiple storer respectively.
9. fingerprint recognition disposal route as claimed in claim 8, it is characterized in that, described storer is DRAM.
10. a fingerprint recognition disposal system, is characterized in that, comprising: programming device and processor, wherein:
Described programming device, is suitable for, after the finger print identifying request receiving processor transmission, carrying out parallel computation to the proper vector of fingerprint template in fingerprint template database; The proper vector result of calculation of parallel computation result and fingerprint to be compared is compared, filters out the fingerprint template data meeting preset rules;
Described processor, be suitable for receiving the fingerprint template data meeting preset rules described in the transmission of described programming device, and described finger print data to be compared and the described fingerprint template data filtered out are carried out the calculating of characteristic similarity score value, from result of calculation, obtain fingerprint recognition operation result.
11. fingerprint recognition disposal systems as claimed in claim 10, is characterized in that, described programming device, are suitable for being calculated by the characteristic similarity score value that described finger print data to be compared and the described fingerprint template data filtered out carry out walking abreast by multithreading.
12. fingerprint recognition disposal systems as claimed in claim 10, it is characterized in that, described programming device, be suitable for choosing fingerprint template concurrently from described fingerprint template database, calculate length and the angle of the proper vector bar in each fingerprint template characteristic of correspondence point set between any two unique points, obtain parallel computation result, described parallel computation result comprises: the length of each fingerprint template characteristic of correspondence to gauge rod and the set of angle.
13. fingerprint recognition disposal systems as claimed in claim 12, it is characterized in that, the proper vector result of calculation of the fingerprint to be compared that described programming device gets is: in the unique point set of described fingerprint to be compared, the length of proper vector bar between any two unique points and the set of angle.
14. fingerprint recognition disposal systems as claimed in claim 13, it is characterized in that, described programming device, be suitable for fingerprint template characteristic of correspondence to the element in the length of gauge rod and the set of angle, contrast to the element in the length of gauge rod and the set of angle with fingerprint characteristic of correspondence to be compared, filter out the fingerprint template that set similarity is greater than preset value.
15. fingerprint recognition disposal systems as claimed in claim 9, it is characterized in that, described processor, be suitable for calculating characteristic similarity score value and predetermined threshold value compares, when described characteristic similarity score value is greater than described predetermined threshold value, judge that corresponding fingerprint and described fingerprint to be compared are as same fingerprint.
16. fingerprint recognition disposal systems as claimed in claim 9, it is characterized in that, described programming device is FPGA, and described processor is arm processor.
17. fingerprint recognition disposal systems as claimed in claim 9, is characterized in that, comprising: multiple storer, are suitable for storing the fingerprint template data in described default fingerprint template database.
18. fingerprint recognition disposal systems as claimed in claim 17, it is characterized in that, described storer is DRAM.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510332833.2A CN104933411A (en) | 2015-06-16 | 2015-06-16 | Fingerprint identification processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510332833.2A CN104933411A (en) | 2015-06-16 | 2015-06-16 | Fingerprint identification processing method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104933411A true CN104933411A (en) | 2015-09-23 |
Family
ID=54120571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510332833.2A Pending CN104933411A (en) | 2015-06-16 | 2015-06-16 | Fingerprint identification processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104933411A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106055946A (en) * | 2016-05-18 | 2016-10-26 | 成都芯软科技发展有限公司 | System and method for identity recognition |
| CN106096515A (en) * | 2016-05-31 | 2016-11-09 | 广东欧珀移动通信有限公司 | A kind of solution lock control method and terminal unit |
| CN109543736A (en) * | 2018-11-15 | 2019-03-29 | 新华三技术有限公司 | A kind of feature comparison method and device |
| CN110199295A (en) * | 2019-04-04 | 2019-09-03 | 深圳市汇顶科技股份有限公司 | The method, apparatus and electronic equipment of fingerprint recognition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1439997A (en) * | 2002-02-22 | 2003-09-03 | 杭州中正生物认证技术有限公司 | Fingerprint identifying method and system |
| US20100027852A1 (en) * | 2004-11-12 | 2010-02-04 | Ming Hsieh | System and Method for Fast Biometric Pattern Matching |
| CN104112005A (en) * | 2014-07-15 | 2014-10-22 | 电子科技大学 | Distributed mass fingerprint identification method |
| CN204680034U (en) * | 2015-06-16 | 2015-09-30 | 迪安杰科技无锡有限公司 | Fingerprint recognition disposal system |
-
2015
- 2015-06-16 CN CN201510332833.2A patent/CN104933411A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1439997A (en) * | 2002-02-22 | 2003-09-03 | 杭州中正生物认证技术有限公司 | Fingerprint identifying method and system |
| US20100027852A1 (en) * | 2004-11-12 | 2010-02-04 | Ming Hsieh | System and Method for Fast Biometric Pattern Matching |
| CN104112005A (en) * | 2014-07-15 | 2014-10-22 | 电子科技大学 | Distributed mass fingerprint identification method |
| CN204680034U (en) * | 2015-06-16 | 2015-09-30 | 迪安杰科技无锡有限公司 | Fingerprint recognition disposal system |
Non-Patent Citations (1)
| Title |
|---|
| 冯国松: "指纹识别算法及其硬件实现研究", 《中国硕士学位论文全文数据库信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106055946A (en) * | 2016-05-18 | 2016-10-26 | 成都芯软科技发展有限公司 | System and method for identity recognition |
| CN106096515A (en) * | 2016-05-31 | 2016-11-09 | 广东欧珀移动通信有限公司 | A kind of solution lock control method and terminal unit |
| CN109543736A (en) * | 2018-11-15 | 2019-03-29 | 新华三技术有限公司 | A kind of feature comparison method and device |
| CN109543736B (en) * | 2018-11-15 | 2020-11-27 | 新华三技术有限公司 | Feature comparison method and device |
| CN110199295A (en) * | 2019-04-04 | 2019-09-03 | 深圳市汇顶科技股份有限公司 | The method, apparatus and electronic equipment of fingerprint recognition |
| WO2020199178A1 (en) * | 2019-04-04 | 2020-10-08 | 深圳市汇顶科技股份有限公司 | Fingerprint recognition method and apparatus, and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108595157B (en) | Block chain data processing method, device, equipment and storage medium | |
| Park et al. | Highway traffic accident prediction using VDS big data analysis | |
| CN103970851B (en) | The method that magnanimity Credential data directly provides general headquarters of large-size enterprise group financial statement | |
| CN106649346B (en) | Data repeatability checking method and device | |
| CN102915347B (en) | A kind of distributed traffic clustering method and system | |
| US20140207820A1 (en) | Method for parallel mining of temporal relations in large event file | |
| CN104933411A (en) | Fingerprint identification processing method and system | |
| CN106294524B (en) | A kind for the treatment of method and apparatus of relation data | |
| CN104281664A (en) | Data segmenting method and system of distributed graph calculating system | |
| CN104834751A (en) | Data analysis method based on Internet of things | |
| Sukhija et al. | Topic modeling and visualization for big data in social sciences | |
| CN102012936A (en) | Massive data aggregation method and system based on cloud computing platform | |
| CN106326398A (en) | Data consistency comparison method and device | |
| Gomes et al. | SAE2: advances on the social adaptive ensemble classifier for data streams | |
| CN105589908A (en) | Association rule computing method for transaction set | |
| CN108763536A (en) | Data bank access method and device | |
| US10558391B2 (en) | Data processing system and data processing method | |
| CN106201917A (en) | A kind of data handling system and method | |
| CN204680034U (en) | Fingerprint recognition disposal system | |
| CN107292320B (en) | System and index optimization method and device thereof | |
| CN106648839A (en) | Method and device for processing data | |
| CN103455491A (en) | Method and device for classifying search terms | |
| CN106648891A (en) | MapReduce model-based task execution method and apparatus | |
| CN107562703A (en) | Dictionary tree reconstructing method and system | |
| CN111967671B (en) | Cross-border active user identification method and device based on support vector data domain description |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20160325 Address after: 101300, Shunyi District Beijing Dongfang Sun City Ming Lake Park T2A-7 Applicant after: Li Bin Address before: 214000, Jiangsu Province, Xishan District, Wuxi Province East Pavilion source village 245-602 room Applicant before: DI'ANJIE TECHNOLOGY WUXI CO., LTD. |
|
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150923 |
|
| WD01 | Invention patent application deemed withdrawn after publication |