CN1333365C - Finger print identification method - Google Patents
Finger print identification method Download PDFInfo
- Publication number
- CN1333365C CN1333365C CNB031422675A CN03142267A CN1333365C CN 1333365 C CN1333365 C CN 1333365C CN B031422675 A CNB031422675 A CN B031422675A CN 03142267 A CN03142267 A CN 03142267A CN 1333365 C CN1333365 C CN 1333365C
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- comparison
- fingerprints
- singular point
- point
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000010586 diagram Methods 0.000 claims abstract description 18
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 238000000605 extraction Methods 0.000 claims abstract description 9
- 150000001875 compounds Chemical class 0.000 claims abstract description 7
- 239000013598 vector Substances 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 8
- 230000007812 deficiency Effects 0.000 claims description 4
- 230000002708 enhancing effect Effects 0.000 claims description 2
- 238000012512 characterization method Methods 0.000 abstract 4
- 230000000750 progressive effect Effects 0.000 abstract 2
- 235000017166 Bambusa arundinacea Nutrition 0.000 description 7
- 235000017491 Bambusa tulda Nutrition 0.000 description 7
- 241001330002 Bambuseae Species 0.000 description 7
- 235000015334 Phyllostachys viridis Nutrition 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 7
- 239000011425 bamboo Substances 0.000 description 7
- 238000001914 filtration Methods 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The present invention relates to a fingerprint identification method which comprises a fingerprint image processing procedure and a fingerprint characteristic comparison procedure. The fingerprint image processing procedure comprises the steps: initial calculation of original images on the basis of a directional diagram and directed degree and extraction and characterization of singular points, characterization of the foreground of fingerprints, characterization of the class characteristics of the fingerprints and extraction and characterization of the detail characteristics of the fingerprints; in this way, the description of the combined characteristics of the fingerprints is achieved. The fingerprint characteristic comparison procedure adopts a progressive compound comparison mode from large characteristics to small characteristics to detail characteristics, namely that firstly, the class characteristics of two fingerprints are compared; secondarily, singular point characteristics of the two fingerprints are compared; finally, detail characteristics of the two fingerprints are compared; if the fingerprint classes, the fingerprint singular points and the fingerprint details are mutually matched, the two fingerprints are judged as the same fingerprint; if any one kind of characteristics is not matched, the two fingerprints are instantly judged as different fingerprints. The adoption of the progressive compound comparison mode greatly accelerates the speed of fingerprint comparison; besides, the present invention has low forsooth-rejecting ratio (FRR) and low false-admitting ratio (FAR).
Description
Affiliated technical field
The present invention relates to fingerprint identification technology, specifically refer to a kind of fingerprint identification method.
Background technology
Identity recognizing technology has a wide range of applications in many checking and occasions of differentiating litigant's identity of needing, as fields such as national defense safety, police criminal detection, social security and bank, security, insurances.Because people's fingerprint varies, almost do not repeat, so fingerprint recognition is a kind of important method of identification.Fingerprint identification technology is meant gathering the integrated technology that next finger print image information adopts computer technology to handle and discern.
The process of fingerprint recognition roughly is, at first obtain finger print image by fingerprint acquisition device, by computing machine this image is handled then, the eigenwert of fingerprint to be compared is extracted, determine by the comparison eigenwert whether both are consistent, and as the eigenwert unanimity, then decidable two fingerprints are same people's fingerprint again, as inconsistent, judge that then two fingerprints are not same people's fingerprints.
Therefore in the prior art, fingerprint comparison generally is the comparison that is conceived to details in fingerprint, the processing of fingerprint also is conceived to extraction and sign to detail characteristics of fingerprints.Carry out fingerprint comparison in this way, comparison speed and accuracy rate are lower.
Summary of the invention
The objective of the invention is to propose a kind of new fingerprint identification method, in fingerprint comparison, take from big feature to little feature the more compound comparison of going forward one by one of minutia-be first classification comparison, singular point comparison again, the pattern of last details comparison, so just can accelerate the speed compared widely, improve the accuracy rate of comparison.
A kind of fingerprint identification method, being divided into finger print image handles and two processes of fingerprint characteristic comparison, at first obtain finger print image by fingerprint acquisition device, by computing machine this image is handled then, the eigenwert of fingerprint to be compared is extracted, determine by the comparison eigenwert whether both are consistent again.Described finger print image is by following processing procedure, thereby realizes the description to the fingerprint assemblage characteristic: the original fingerprint image is carried out extraction and sign based on directional diagram and oriented degree initial calculation and singular point; Prospect and background to fingerprint are cut apart; The sign of fingerprint classification and category feature; Finger print image enhancing, binaryzation and processing again; The extraction of refinement and minutia and sign; The process of described fingerprint characteristic comparison is the compound comparison pattern of going forward one by one of minutia again from big feature to little feature: it is right promptly earlier two fingerprints to be carried out the fingerprint classification aspect ratio, and it is right to carry out fingerprint singular point aspect ratio again, carries out the detail characteristics of fingerprints comparison at last; All mate as fingerprint classification, fingerprint singular point and details in fingerprint, then be judged to same fingerprint, do not match, then can be judged to different fingerprints immediately as arbitrary feature wherein.
Do the fingerprint classification aspect ratio to the time, be class declaration monoid, like this when two fingerprints can't belong to definite subclass, but its monoid occurs simultaneously and is not equal to empty set, the match is successful then still to confirm its classification; Do fingerprint singular point aspect ratio to the time, single-point comparison angle, multiple spot comparison vector, as determine to have reference point is analyzed unnecessary singular point, and is got rid of invalid singular point, only after the invalid singular point of expectation was excluded, the singular point comparison just was counted as merit; When doing the minutia comparison: see if there is reference point earlier,, then reference point is done in minutiae point samsara choosing, and called simple comparer program and compare as no reference point; If any reference point, further see if there is sign of rotation, as indicating without spin, then call simple comparer program and compare; If any sign of rotation, then be rotated the angle stepping, and call simple comparer program and compare; Have only compare successfully by simple comparer program after, the fingerprint that two quilts are compared can be judged to same fingerprint.
When finger print image is handled, take full figure to resemble the pattern of 2 * 2 pixel piecemeals, 32 directions of each piece.
The invention has the advantages that: (1) has accelerated the speed of fingerprint comparison greatly owing to partly take first classification, the compound comparison pattern of going forward one by one of singular point, back details again at fingerprint comparison.And algorithm is pervasive, quick, healthy and strong, has low refuse sincere (FRR) and low accuracy of system identification (FAR); (2) because the fingerprint processing section takes full figure to resemble the pattern of 2*2 pixel piecemeal, 32 directions of each piece, from position and the direction finger print image directional diagram that become more meticulous, for determining singular position and singular point direction and trend pass filtering image enhancement later on and detecting details direction etc. and lay a good foundation.
Description of drawings
Fig. 1 is the structured flowchart of fingerprint acquisition device;
Fig. 2 is a fingerprint process flow block diagram of the present invention;
Fig. 3 is a fingerprint comparison FB(flow block) of the present invention;
Fig. 4 is a 9*9 direction template synoptic diagram;
Fig. 5 extracts process flow diagram for minutia;
Fig. 6 is minutia comparison process flow diagram;
Fig. 7 is simple comparison process flow diagram.
Embodiment
Now the invention will be further described in conjunction with an embodiment and accompanying drawing thereof.
Fingerprint collecting can in all sorts of ways, as stamp, optically detecting or chip collection etc.Present embodiment adopts the off-line fingerprint harvester of being made up of fingerprint collecting chip and single-chip microcomputer.Wherein 101 is the fingerprint collecting chip, adopt AF-S2 live body acquisition chip, 102 for CPU adopts the DSP special chips of TI company, external bus articulate FLASH103, SRAM104 and with the communication interface circuit 106 (as the structured flowchart of Fig. 1 fingerprint acquisition device) of PC.SRAM is used for the expanding internal storer, and FLASH is used for stored programme and finger print data, and total line traffic control is finished by programmable logic array CPLD105.Can control this off-line fingerprint harvester by agreement by host computer, carry out the fingerprint collecting, comparison of multiple mode and, also can make simple fingerprint comparison independently and operations such as the interpolation of fingerprint database, deletions to the management of fingerprint base in the module.
Fingerprint recognition comprises fingerprint processing and fingerprint comparison two big processes.Narrated respectively below.
One. fingerprint is handled:
It is by the processing to finger print image that fingerprint is handled, and realizes the description to the fingerprint assemblage characteristic.Its FB(flow block) is seen Fig. 2 fingerprint process flow block diagram.
Wherein 201 frames are raw image, 202 frames are that the initial calculation of raw image travel direction figure and oriented degree and singular point are extracted and the singular point angle calculation, the detecting and deleting of pseudo-singular point, 203 frames are foreground segmentation, 204 frames are that prospect characterizes and the polygonal calculating of prospect, 205 frames are fingerprint classification, 206 frames characterize for the classification singular point, 207 frames are that finger print image strengthens, and 208 frames are to image binaryzation after strengthening and processing again, and 209 frames are that binary image refinement and minutia are extracted, 210 frames are that details characterizes, like this, by to foreground features, classification and singular point feature, the sign of minutia reaches the comprehensive description of 211 frames to finger print image.
Existing division is as follows:
202 frames: the initial calculation of raw image travel direction figure and oriented degree and singular point are extracted and the singular point angle calculation the detecting and deleting of pseudo-singular point
This is to carry out rough handling to gathering the finger print image that comes, and fundamental purpose is to obtain the directional diagram and the oriented degree of fingerprint, and obtains singular point.Its process is:
1) piecemeal, obtain directional diagram and oriented degree:
At first carry out pixel level and calculate, carry out the big window of circle then and smoothly calculate.
Present embodiment is taked piecemeal yardstick 2 * 2 pixels to the raw image piecemeal, and the piece direction is got the fine division pattern of 32 directions; This asks for the positional precision of singular point and the speed of directional precision to raising, is used for the precision of trend pass filtering image enhancement, and the details direction detect simply, all very useful fast.
Directional diagram is intended to describe the trend of fingerprint, can take two kinds of computation schemas, promptly based on 9*9 direction template (referring to Fig. 4) or optimize the pixel direction calculating of fitting process based on the least square of pixel gradient, carry out in the piece average then, thereby obtain the predominant direction of piece, i.e. the trend of local fingerprint streakline.
Wherein:
The pixel gradient adopts the Sobel gradient template to calculate, and is designated as: and Gx (i, j), Gy (i, j);
The piece orientation angle is designated as θ, θ=1/2 * atan (the ∑ ∑ (2 * Gx (i, j) * Gy (i, j))/∑ ∑ (Gx (i, j) * Gx (i, j)-Gy (i, j) * Gy (i, j)));
Oriented degree is an index of describing the directive degree of finger print image, and the requirement of cutting apart that is based on multiple image produces, and is intended to cutting apart the finger print image prospect background.
The oriented degree of pixel: the direction relative mistake based on 9 * 9 direction templates calculates, and formula is: (X-Y)/(X+Y), X=usSum[ucMaxDrc wherein], Y=usSum[ucMinDrc].
The oriented degree of piece: the oriented degree arithmetic mean of markization of pixel in the piece.
2) the level and smooth and oriented degree that stretches: the oriented degree of the big window smooth block of full figure, the oriented degree of minimax piece with full figure is a bound then, carries out the stretching of histogram 0-255 markization, and carries out the image foreground background segment in view of the above.
3) level and smooth directional diagram, obtain singular point;
A) level and smooth directional diagram: it is average to take a times angle amount arithmetic mean method to carry out interblock, to obtain predominant direction.Its purpose is that the local fingerprint streakline that has smoothly obtained moves towards, and eliminates disturbance, prevents to produce pseudo-singular point.
B) extraction of singular point: singular point comprises two kinds of arc point and trigpoints.Company rotates counterclockwise with the four directions, and totally unidirectional neighbours are poor to press mould 16.Equal 32 as neighbour's difference, then be judged to be arc point; Equal-32 as neighbour's difference, then be judged to be trigpoint.
C) calculating of singular point angle: the singular point angle is the important indicator of the big local feature of reflection fingerprint, plays key effect in follow-up classification and comparison.Here adopt oeil de boeuf constraint block to flow to and the azimuthal consistance statistical calculation method of this piece based on block directed graph.Singular point angle calculation flow process is: compare each block directed graph angle and azimuthal consistent degree, carry out the minimax statistics of consistent degree under the oeil de boeuf constraint, calculate the direction of arc point and trigpoint with this.
D) the detecting and deleting of pseudo-singular point: because the quality of finger print image own may cause pattern distortion, thereby produce pseudo-singular point, therefore need detect and delete.It is as follows to detect flow process: because common pseudo-singular point all appears at the prospect background intersection, or the stained place of prospect, therefore, confirming pseudo-singular point to have occurred, and after further finding out pseudo-singular point, deleted.
203,204 frames: foreground segmentation and prospect characterize and the polygonal calculating of prospect
1) foreground segmentation is to cut apart with the fixedly threshold values under the oriented scaleization.
2) calculating of prospect polygon is what to carry out on the basis of foreground segmentation.The prospect polygon calculates and takes up and down from the squeeze mode of visual frame to image center, at first determine the border up and down of prospect, and then definite prospect four angles of frame, i.e. four borders at the preceding scenic spot in the lower left corner, the lower right corner, the upper right corner, the upper left corner up and down.
205 frames: fingerprint classification
The division of fingerprint classification: be divided into arch, square-bottomed bamboo basket shape, left dustpan, right dustpan, suitable two dustpans, contrary two dustpans, cusped arch and foreign peoples, and their various combinations.Classification foundation is the distribution according to singular position and direction, but is fuzzy classification sometimes, promptly is included in the monoid of a few classes, sees Table 1:
Table 1 is based on the fingerprint classification table of singular point
| 0 arc point | 1 arc point | 2 arc points | |
| 0 trigpoint | All kinds is all possible. | Square-bottomed bamboo basket, dustpan, cusped arch are all possible, but are not arches. | Square-bottomed bamboo basket shape or two dustpan are not arch, cusped arch, dustpan shape. |
| 1 trigpoint | All kinds is all possible. | May be cusped arch, cusped arch left side dustpan or right dustpan or contrary two dustpans of left dustpan or the suitable two dustpans of right dustpan. | Square-bottomed bamboo basket shape or two dustpan can not be arch, cusped arch, dustpan shape. |
| 2 trigpoints | May be square-bottomed bamboo basket shape or two dustpan, be left dustpan, right dustpan, cusped arch or arch. | May be square-bottomed bamboo basket shape or two dustpan, can not be arch or Gothic arch. | Square-bottomed bamboo basket shape or two dustpan can not be arch, cusped arch, dustpan shape. |
206 frames: classification and singular point characterize
Here be meant that big feature-classification and singular point to fingerprint characterize.The sign of classification relates to class and monoid, and the sign of singular point relates to the classification of singular point, position and direction.
207 frames: finger print image strengthens
The purpose of image enhancement is to strengthen fingerprint ridge, weakens so that the non-fingerprint ridge of eliminating in the image disturbs.
Present embodiment takes the adjustable Gabor wave filter of direction parameter that the original fingerprint image is carried out filtering, realizes image enhancement.
Filter function is described formula:
δ Φ is a rotational angle in the formula, and δ T is cycle of fluctuation, and δ X is a length parameter, and δ Y is a width parameter.
208 frames: strengthen the back image binarization and handle again:
1) binaryzation: adopt adaptive threshold method binaryzation method to carry out image binaryzation, its essence is that getting the window average carries out local auto-adaptive.
2) handle again: need handle again imperfect image after the binaryzation, to satisfy the needs that minutia is extracted, such as the hole of driving away on the streakline.The present invention adopts follow-on median filtering algorithm to handle again.
209: binary image refinement and minutia are extracted
1) refinement of binary image can be adopted existing thinning algorithm.
2) minutia of fingerprint is meant the jumping phenomenon in the details of fingerprint ridge, mainly be bifurcated and end slightly.The Feature Extraction process is extracted process flow diagram referring to the minutia of Fig. 5.Among the figure: 502 frames are to make microlocal analysis through the image of refinement; 503 frames are for getting rid of the constraint of singular point neighborhood; 504 frames for judge be the end slightly or bifurcated? 505 frames are then moved slightly in the end in this way, do the end and put big partial analysis slightly; Bifurcated then moves 506 frames in this way, does the big partial analysis of bifurcation, with further affirmation.
Wherein:
502 frame microlocal analysis: comprise that stain constraint goes forward one by one to six or two constraints, as bifurcated and end necessary condition slightly occurring, simultaneously, when detecting six or two constraints, remember peripheral black and white pixel distribution, as the usefulness of follow-up big partial analysis with this.
Big partial analysis flow process is put at 505 frame ends slightly: follow the tracks of the limited step, see short-term whether occurs, side is propped up or rupture, and as not being that short-term or side are propped up, does not also rupture, and then confirms as the end slightly; As in propping up, rupturing any of short-term, side appears, then negate the end slightly.
The big partial analysis flow process of 506 frame bifurcations is: three branches to bifurcated all follow the tracks of the limited step, see short-term or side fork whether occur, as not being short-term, neither side pitch, and then confirm as bifurcated; As short-term or side fork appear, then negate bifurcated.
210: details characterizes
Details characterizes mainly to the end slightly or the sign of the position of bifurcated and direction.
211: assemblage characteristic is described:
By comprehensive foreground features sign, category feature and singular point characteristic present, minutia are characterized, reach comprehensive description to the finger print image assemblage characteristic.
Two. fingerprint comparison:
The fingerprint comparison of present embodiment is taked the compound comparison pattern of going forward one by one from big feature to little feature, first classification comparison, singular point comparison again, back details comparison, and also the comparison of classification and big feature is relatively more tolerant, thus accelerated comparison speed greatly.Its flow process such as Fig. 3 fingerprint comparison FB(flow block).Existing division is as follows:
301,302 frames: category feature comparison
On the basis of fingerprint classification, it is fairly simple to do the category feature comparison.When present embodiment is compared at category feature, is class declaration monoid, work as two fingerprint subclasses ownership so more at need, can get rid of some class and contain some class, therefore, occuring simultaneously as the monoid of two fingerprints is not equal to empty set, the match is successful then to confirm its classification, change 303 frames and continue comparison, otherwise change 308 frames, be judged to different fingerprints.
303,304 frames: the singular point aspect ratio is right
The purpose that the singular point aspect ratio is right is to determine whether there is common reference points between two fingerprints of comparing, and determines it is which type of reference point.The existing requirement that is similar to the details comparison of singular point comparison also relates to the eliminating of the corresponding preceding scenic spot of the other side's fingerprint predicted position of unnecessary singular point.The singular point aspect ratio is vertically placed time hypothesis finger print image, allows to swing in ± 45 ° of scopes and be inverted.Single-point comparison angle, multiple spot comparison vector.
As determine to have reference point, and more unnecessary singular point is analyzed, and got rid of invalid singular point, only after the invalid singular point of expectation was excluded, the singular point comparison just was counted as merit, entered 305 frame details comparison program; Otherwise, entering 308 frames, it fails to match for singular point, is judged to be different fingerprints.
306,307 frames: minutia comparison
Minutia comparison idiographic flow is referring to Fig. 6 minutia comparison process flow diagram.The technology path of vectorization comparison is still taked in the minutia comparison.Different is: the vector ordering is optional for the angle/length of vector; In addition, adopted and calculated the space crossover region of expressing in advance based on the prospect polygon, promptly under the prerequisite of reference point correspondence, investigate which side that certain fingerprint minutiae is positioned at another polygonal limit of fingerprint prospect, and successful details number and the successful details number and total ratio of comparison of details number, the crossover region comparison of judging crossover region, as number deficiency or ratio deficiency, still think this reference point and with reference to the comparison of angle failure.
602 frames see if there is reference point earlier by Fig. 6? as do not have reference point, and then carry out 606 frames, reference point is done in minutiae point samsara choosing, and carried out 604 frames and call simple comparer program; If any reference point, carry out 603 frames, further see if there is sign of rotation? as indicating without spin, then carry out 604 frames and call simple comparer program; If any sign of rotation, then carry out 608 frames, be rotated the angle stepping, and carry out 604 frames and call simple comparer program; By calling simple comparer program, the comparison of finally determining two fingerprints is success or failure.
The simple comparer program circuit of 604 frames is compared process flow diagram merely referring to Fig. 7: at first carries out 702,703 frames and carries out space overlapping calculating, see whether the overlapping details enough? as inenough, then change this comparison failure of 710 frames; As enough 704,705,706 frames of then carrying out, vector generates, vector sorts, the vector comparison, successively carries out 707 frames again, see whether the successful details of comparison enough? whether 708 frame overlapping ratios enough big? if any one not enough, then this comparison failure; As all enough, then this is compared successfully.
So far, the fingerprint comparison process is finished.If reference point has been obtained in the singular point comparison, then the details comparison speed here is very fast usually.
Claims (5)
1. fingerprint identification method, being divided into finger print image handles and two processes of fingerprint characteristic comparison, at first obtain finger print image by fingerprint acquisition device, by computing machine this image is handled then, the eigenwert of fingerprint to be compared is extracted, determine that by the comparison eigenwert whether both are consistent, is characterized in that again:
Described finger print image is by following processing procedure, thereby realizes the description to the fingerprint assemblage characteristic: the original fingerprint image is carried out extraction and sign based on directional diagram and oriented degree initial calculation and singular point; Prospect and background to fingerprint are cut apart; The sign of fingerprint classification and category feature; Finger print image enhancing, binaryzation and processing again; The extraction of refinement and minutia and sign;
The process of described fingerprint characteristic comparison is the compound comparison pattern of going forward one by one of minutia again from big feature to little feature: it is right promptly earlier two fingerprints to be carried out the fingerprint classification aspect ratio, and it is right to carry out fingerprint singular point aspect ratio again, carries out the detail characteristics of fingerprints comparison at last; All mate as fingerprint classification, fingerprint singular point and details in fingerprint, then be judged to same fingerprint, do not match, then can be judged to different fingerprints immediately as arbitrary feature wherein;
Do the fingerprint classification aspect ratio to the time, be class declaration monoid, like this when two fingerprints can't belong to definite subclass, but its monoid occurs simultaneously and is not equal to empty set, the match is successful then still to confirm its classification;
Do fingerprint singular point aspect ratio to the time, single-point comparison angle, multiple spot comparison vector, as determine to have reference point is analyzed unnecessary singular point, and is got rid of invalid singular point, only after the invalid singular point of expectation was excluded, the singular point comparison just was counted as merit;
When doing the minutia comparison: see if there is reference point earlier,, then reference point is done in minutiae point samsara choosing, and called simple comparer program and compare as no reference point; If any reference point, further see if there is sign of rotation, as indicating without spin, then call simple comparer program and compare; If any sign of rotation, then be rotated the angle stepping, and call simple comparer program and compare; Have only compare successfully by simple comparer program after, the fingerprint that two quilts are compared can be judged to same fingerprint.
2. fingerprint identification method according to claim 1 is characterized in that when finger print image is handled, and takes full figure to resemble the pattern of 2 * 2 pixel piecemeals, 32 directions of each piece.
3. fingerprint identification method according to claim 1, it is characterized in that doing the singular point aspect ratio to the time set finger print image and vertically place, allow in ± 45 ° of scopes, to swing and be inverted.
4. fingerprint identification method according to claim 1 when it is characterized in that the minutia comparison, is taked the technology path of vectorization comparison, and the vector ordering is optional for the angle/length of vector; In addition, adopted and calculated the space crossover region of expressing in advance based on the prospect polygon, promptly under the prerequisite of reference point correspondence, investigate which side that certain fingerprint minutiae is positioned at another polygonal limit of fingerprint prospect, and successful details number and the successful details number and total ratio of comparison of details number, the crossover region comparison of judging crossover region, as number deficiency or ratio deficiency, then declare this reference point and with reference to the comparison of angle failure.
5. fingerprint identification method according to claim 1 is characterized in that the step of described simple comparer program is: at first carries out the space overlapping and calculates, see whether the overlapping details is enough, and as inenough, then this comparison failure; As enough then vectors generate, vector ordering and vector comparison, successively see again whether the successful details of comparison enough, whether the overlapping ratio enough big, if any one not enough, then this comparison failure; As all enough, then this is compared successfully.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031422675A CN1333365C (en) | 2003-08-14 | 2003-08-14 | Finger print identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031422675A CN1333365C (en) | 2003-08-14 | 2003-08-14 | Finger print identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1581206A CN1581206A (en) | 2005-02-16 |
| CN1333365C true CN1333365C (en) | 2007-08-22 |
Family
ID=34579441
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB031422675A Expired - Fee Related CN1333365C (en) | 2003-08-14 | 2003-08-14 | Finger print identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1333365C (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101276411B (en) * | 2008-05-12 | 2010-06-02 | 北京理工大学 | Fingerprint identification method |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1322465C (en) * | 2005-08-15 | 2007-06-20 | 阜阳师范学院 | Image segmentation and fingerprint line distance getting technique in automatic fingerprint identification method |
| CN100447806C (en) * | 2006-06-14 | 2008-12-31 | 北京握奇数据系统有限公司 | Method, device and use for matching two-stage mixed-fingerprint characteristics |
| CN101901332A (en) * | 2009-05-31 | 2010-12-01 | 上海点佰趣信息科技有限公司 | Fingerprint identification system and method |
| CN102184427B (en) * | 2011-04-27 | 2013-06-19 | 杭州晟元芯片技术有限公司 | Method for reducing false accept rate of fingerprints |
| CN103258158A (en) * | 2013-05-08 | 2013-08-21 | 大连民族学院 | Fingerprint authentication electronic commerce scrambler |
| CN108182375B (en) * | 2016-12-08 | 2020-11-06 | 广东精点数据科技股份有限公司 | Fingerprint identification system based on mobile phone payment |
| CN107193393B (en) * | 2017-04-28 | 2021-05-04 | 北京小米移动软件有限公司 | Input method switching method and device |
| CN107958217A (en) * | 2017-11-28 | 2018-04-24 | 广州麦仑信息科技有限公司 | A kind of fingerprint classification identifying system and method based on deep learning |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1136848A (en) * | 1994-10-12 | 1996-11-27 | 洛克希德马丁公司 | Fingerprint matching system |
| KR20030040825A (en) * | 2001-11-16 | 2003-05-23 | 에버미디어 주식회사 | The method for improving fingerprint idenfication using adptive thresholding and minutiae restoration |
-
2003
- 2003-08-14 CN CNB031422675A patent/CN1333365C/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1136848A (en) * | 1994-10-12 | 1996-11-27 | 洛克希德马丁公司 | Fingerprint matching system |
| KR20030040825A (en) * | 2001-11-16 | 2003-05-23 | 에버미디어 주식회사 | The method for improving fingerprint idenfication using adptive thresholding and minutiae restoration |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101276411B (en) * | 2008-05-12 | 2010-06-02 | 北京理工大学 | Fingerprint identification method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1581206A (en) | 2005-02-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10930105B2 (en) | Banknote management method and system | |
| CN101599124B (en) | Method and device for segmenting characters from video image | |
| CN110309687A (en) | A kind of bearing calibration of image in 2 D code and means for correcting | |
| CN113538486B (en) | Method for improving identification and positioning accuracy of automobile sheet metal workpiece | |
| Wang et al. | Traffic sign detection using a cascade method with fast feature extraction and saliency test | |
| CN102567993B (en) | Fingerprint image quality evaluation method based on main component analysis | |
| CN101144716A (en) | Multiple angle movement target detection, positioning and aligning method | |
| CN104680550A (en) | Method for detecting defect on surface of bearing by image feature points | |
| CN1333365C (en) | Finger print identification method | |
| CN103020626A (en) | Method and system for detecting elliptical target in image | |
| CN101114335A (en) | Full angle rapid fingerprint identification method | |
| CN111738271A (en) | Method for identifying shielded fruits in natural environment | |
| CN107016394B (en) | Cross fiber feature point matching method | |
| CN109101854A (en) | A kind of multiple barcode localization method | |
| Bellon et al. | Edge detection to guide range image segmentation by clustering techniques | |
| CN110852388A (en) | Improved SMOTE algorithm based on K-means | |
| CN117689655B (en) | Metal button surface defect detection method based on computer vision | |
| CN114120378A (en) | Three-level classification fingerprint identification method | |
| CN112149548A (en) | CAD drawing intelligent input and identification method and device suitable for terminal row | |
| CN110378337B (en) | Visual input method and system for drawing identification information of metal cutting tool | |
| Sasakawa et al. | Personal verification system with high tolerance of poor-quality fingerprints | |
| WO2021258498A1 (en) | Ellipse image detection algorithm and system based on contour integration | |
| Li et al. | AdaHC: Adaptive hedge horizontal cross-section center detection algorithm | |
| Ashbrook et al. | Tutorial: Algorithms for 2-dimensional object recognition | |
| CN102332088B (en) | Vote symbolic machine visual identification method based on run length feature |
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 | ||
| C56 | Change in the name or address of the patentee |
Owner name: BEIJING FANGZHENG AODE COMPUTER SYSTEM CO., LTD.; Free format text: FORMER NAME OR ADDRESS: YIWEI SCIENCE-TECHNOLOGY CO., LTD., SHANGHAI |
|
| CP03 | Change of name, title or address |
Address after: 6 floor, fangzheng building, 298 Cheng Fu Road, Haidian District, Beijing, zip code: 100871 Co-patentee after: Peking Founder Group Co., Ltd. Patentee after: Beijing Fangzheng ODA computer system Co., Ltd. Address before: Room 80, No. 210, Nandan Road, Shanghai, China Patentee before: Yiwei Science-Technology Co., Ltd., Shanghai |
|
| ASS | Succession or assignment of patent right |
Owner name: FOUNDER INTERNATIONAL CO., LTD. Free format text: FORMER OWNER: BEIJING FANGZHENG AODE COMPUTER SYSTEM CO., LTD. Effective date: 20110519 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 100871 6/F, FOUNDER BUILDING, NO. 298, CHENGFU ROAD, HAIDIAN DISTRICT, BEIJING TO: 100080 19/F, ZHONGXIN BUILDING, NO. 52, NORTH 4TH RING ROAD WEST, HAIDIAN DISTRICT |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20110519 Address after: 100080, No. 52 West Fourth Ring Road, Haidian District central building, 19 floor Co-patentee after: Peking Founder Group Co., Ltd. Patentee after: Founder international software (Beijing) Co., Ltd. Address before: 100871, Beijing, Haidian District Cheng Fu Road 298, founder building, 6 floor Co-patentee before: Peking Founder Group Co., Ltd. Patentee before: Beijing Fangzheng Aode Computer System Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20070822 Termination date: 20170814 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |