CN101408939B - Method and system for extracting fingerprint thin node in fingerprint image - Google Patents
Method and system for extracting fingerprint thin node in fingerprint image Download PDFInfo
- Publication number
- CN101408939B CN101408939B CN2008102246142A CN200810224614A CN101408939B CN 101408939 B CN101408939 B CN 101408939B CN 2008102246142 A CN2008102246142 A CN 2008102246142A CN 200810224614 A CN200810224614 A CN 200810224614A CN 101408939 B CN101408939 B CN 101408939B
- Authority
- CN
- China
- Prior art keywords
- fingerprint image
- model
- crestal line
- module
- discrete
- 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 26
- 238000000605 extraction Methods 0.000 claims abstract description 35
- 238000001914 filtration Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 5
- 230000002708 enhancing effect Effects 0.000 abstract description 2
- 239000000284 extract Substances 0.000 description 14
- 230000006870 function Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 3
- 239000007787 solid Substances 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 230000001788 irregular Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000026676 system process Effects 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000002386 leaching Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Abstract
The invention relates to a method for extracting fingerprint minutiae of a fingerprint image and a system thereof. The system comprises a ridge line orientation field generating module which is used for generating the ridge line orientation field of the fingerprint image; an AM-FM model generating module which is used for generating a discrete AM-FM model according to the fingerprint image and the ridge line orientation field of the fingerprint image; and a minutiae extraction module which is used for extracting the minutiae according to the discrete AM-FM model and the ridge line orientation field of the fingerprint image. The method and the system improve the extraction accuracy of the fingerprint minutiae, thus enhancing performance of a fingerprint identification algorithm.
Description
Technical field
The present invention relates to fingerprint image and handle, relate in particular to a kind of fingerprint thin node in fingerprint image extracting method and system.
Background technology
Bio-identification is as Secure Application, because its ease for use and unforgeable to a certain extent have vast market prospect.Fingerprint collecting is convenient, and what can be used for discerning contains much information, and identification accurately has great using value.But it is to be difficult for gathering that the human finger has quite a few, because a variety of causes as common dried, wet, grey, wet goods transient cause, even because the problem of finger skin itself causes the images acquired signal to noise ratio (S/N ratio) low, is difficult to handle.Therefore, on certain sensor acquisition level, further improving the accuracy rate that finger print information extracts, is the key that makes fingerprint recognition system practicability.Tradition is the two-value streakline to the understanding of fingerprint, and has defined the minutiae point definition based on topological characteristic: the end points of crestal line or bifurcation.This definition requires to extract minutiae point by scanning and identification in the processing procedure, and step is more complicated, and it is inaccurate that the not good image of acquisition quality is extracted minutiae point.
The AM-FM model is a kind of complex signal model of fingerprint image.Under this model, the regular domain of fingerprint, i.e. fingerprint ridge parallel zone, irregular area with fingerprint, be fingerprint minutiae, more intuitive differentiation is arranged: the mould of regular domain AM-FM model value is proportional to the image local contrast, and irregular area AM-FM model value is through 0.This definition is very succinct, and has provided extremely simply, directly extracts the method for minutiae point on the AM-FM model: the mould of AM-FM model is near zero zone, if around the mould of AM-FM model all bigger, then minutiae point is contained in this zone.Set up the AM-FM model and directly extract minutiae point from the AM-FM model from original graph one step, do not have miscellaneous image transitions and scanning, leaching process is more succinctly also more accurate.Especially to low-quality image, can extract minutiae point with accuracy far above traditional minutiae point extracting mode.
But, sharply descend based on the easy performance that causes affected by noise of the extracting mode of AM-FM model.This is because in the minutiae point decision rule, it is 0 that minutiae point center AM-FM model value gets mould, this means easily to be covered by noise, also produces fake minutiae because of noise easily.So to ropy image, need carry out smoothly, to reduce The noise to the AM-FM model.
That uses among the present invention is necessary that the term definition that describes is as follows:
The AM-FM model: a complex signal model, think that signal is made of the function that one or several meet as giving a definition, only refer to the AM-FM model of single composition (single component) in this article:
h(x)＝A(x)exp(Ω(x))
Wherein x is a vector, for image, is the coordinate in the two-dimensional space.Wherein A (x) is an Amplitude Modular amplitude modulation part, and exp (Ω (x)) is a Frequency Modular frequency modulation part, and Ω (x) represents phase place.
Discrete grid block: some points and the connection between them in the continuous space.The two-dimentional uniform grid that the discrete grid block of indication is made up of rectangle among the present invention stores with matrix-style.Image itself promptly is a kind of function that stores with discrete grid block.
Coordinate: refer to the coordinate in space, also refer to the subscript of discrete grid block node.Because employed discrete grid block is uniformly and saves as matrix, volume coordinate and discrete grid block node are corresponding one by one.
Resolution: the inverse of the space length of discrete grid block adjacent node correspondence, the i.e. density of discrete grid block.
Discrete AM-FM model: the discrete function that the sampling of AM-FM model on the discrete grid block node formed.Since the value in the digital processing be the discrete sampling value, so AM-FM model value and discrete AM-FM model value are not made any distinction between.
Multiple Gabor wave filter: with real Gabor wave filter masterplate two-dimentional Hilbert conversion as imaginary part.It is defined as:
σ wherein
xControlled frequency bandwidth σ
yControlling party is to bandwidth,
Be frequency direction, f is the wave filter intermediate frequency.
Multiple Gabor bank of filters: it is half annulus at center that one group of multiple Gabor wave filter that the intermediate frequency equal direction is different, its frequency response cover on the frequency domain at least with the initial point.
Fingerprint image: the gray scale image of reflection finger print texture information.
Local frequency spectrum: the frequency spectrum that fingerprint image obtains through the windowing spectrum analysis.
Fingerprint ridge line: the part of expression fingerprint projection on the fingerprint image, corresponding recessed part is a valley line.
Crestal line direction: crestal line somewhere tangential.Because direction that tangentially can corresponding two phase difference of pi angles is so think a crestal line direction with them.The interval of crestal line direction be [0, π) rather than [0,2 π).The crestal line interval has only semi-circumference, answers the reason that the frequency response of Gabor bank of filters only need cover semicircular ring just.
The crestal line normal direction: with the crestal line direction to vertical direction.Accordingly, the interval also be [0, π).
The crestal line field of direction: because fingerprint image each several part crestal line direction is different, to each partial fingerprint image, estimate the crestal line direction, then coordinate is exactly the crestal line field of direction to the two-dimensional discrete function of crestal line direction.A value is exactly corresponding local crestal line direction in the crestal line field of direction.
Direction degree of confidence: the credibility of certain local crestal line direction judgement.
Direction degree of confidence field: corresponding with the crestal line field of direction, a value in the direction degree of confidence field is the credibility of the crestal line direction of respective coordinates in the crestal line field of direction.
Fingerprint minutiae: the bifurcation of crestal line and end points.Because the conjugacy of crestal line and valley line, the corresponding respectively end points and the bifurcation of valley line.Under the AM-FM model, show as mistake 0 point of AM-FM model value.
Summary of the invention
In order to solve above-mentioned technical matters, a kind of fingerprint thin node in fingerprint image extracting method and system based on the AM-FM model is provided, its purpose is, is embodied as the more excellent fingerprint image characteristics extraction algorithm of performance, to improve the Fingerprint Image Recognition Algorithms performance.With respect to traditional extraction process, the method especially has minutiae point to extract the advantage of accuracy rate to the inferior quality fingerprint image.
The invention provides a kind of fingerprint thin node in fingerprint image extraction system, comprising:
Crestal line field of direction generation module comprises crestal line direction extraction module, is used for carrying out according to fingerprint image the crestal line field of direction extraction of fingerprint image;
AM-FM model generation module comprises the AM-FM extraction module, is used for extracting the first discrete AM-FM model according to the crestal line field of direction and the fingerprint image of fingerprint image; Described AM-FM extraction module comprises multiple Gabor wave filter generation module and filtering correcting module as a result, described multiple Gabor wave filter generation module, be used to generate the multiple Gabor wave filter of filter direction and local fingerprint image crestal line direction quadrature, and local fingerprint image is carried out filtering obtain the filtering result; Described filtering is correcting module as a result, to multiply by described filtering result again after the increasing function mapping of direction degree of confidence with any process initial point, the value that obtains is as the AM-FM model value of this local fingerprint image center, and this AM-FM model value is used to make up the first discrete AM-FM model;
The minutiae point extraction module comprises minutiae point location determination module and minutiae point direction determining module, is used for extracting minutiae point according to the crestal line field of direction of the first discrete AM-FM model and fingerprint image; Described minutiae point location determination module, after being used for the AM-FM model value of all coordinates asked mould, if: mould less than threshold value, mould less than the direction difference of neighborhood and mould in all directions reckling greater than threshold value, then there is minutiae point in the judgement of minutiae point location determination module, and output minutiae point coordinate sequence; Described minutiae point direction determining module is used for the coordinate according to the minutiae point coordinate sequence, and the crestal line direction of taking out this coordinate from the field of direction is exported the minutiae point direction sequence as the minutiae point direction.
Crestal line field of direction generation module also comprises the level and smooth module of the field of direction, be used for the crestal line field of direction of fingerprint image is carried out smoothly, and the crestal line field of direction of the fingerprint image after will be level and smooth exports the AM-FM extraction module to as the crestal line field of direction of fingerprint image;
AM-FM model generation module also comprises level and smooth module of AM-FM and/or geometric transformation reconstructed module, be used for the first discrete AM-FM model of AM-FM extraction module output is carried out smoothly, and/or the first discrete AM-FM model of AM-FM extraction module output carried out interpolation, the output second discrete AM-FM model;
The minutiae point extraction module also comprises the resolution reconstructed module, is used for realizing resolution reconstruct by interpolation.
Also comprise central point extraction module and/or image reconstruction module;
The central point extraction module is used for extracting center point coordinate according to the crestal line field of direction of fingerprint image;
The image reconstruction module is used for according to the first discrete AM-FM or the second discrete AM-FM model output overall situation or the local fingerprint image.
The invention provides a kind of fingerprint thin node in fingerprint image extracting method, comprising:
Step 1 is carried out the crestal line field of direction of fingerprint image and is extracted according to fingerprint image, generate the crestal line field of direction of fingerprint image;
Step 2 generates the multiple Gabor wave filter of filter direction and local fingerprint image crestal line direction quadrature, and local fingerprint image is carried out filtering obtains the filtering result; To multiply by described filtering result again after the increasing function mapping of direction degree of confidence with any process initial point, the value that obtains utilizes this AM-FM model value to make up the first discrete AM-FM model as the AM-FM model value of this local fingerprint image center;
Step 3, the AM-FM model value of all coordinates asked mould after, if: mould is then judged to have minutiae point greater than threshold value less than direction difference reckling in all directions of neighborhood and mould less than threshold value, mould, and output minutiae point coordinate sequence; According to the coordinate in the minutiae point coordinate sequence, the crestal line direction of taking out this coordinate from the field of direction is exported the minutiae point direction sequence as the minutiae point direction.
In the step 1, also the crestal line field of direction of fingerprint image is carried out smoothly, and the crestal line field of direction of the fingerprint image after level and smooth is exported as the crestal line field of direction of fingerprint image;
In the step 2, also the first discrete AM-FM model is carried out smoothly, and/or the first discrete AM-FM model is carried out interpolation, the output second discrete AM-FM model;
In the step 3, realize resolution reconstruct by interpolation.
Also comprise:
Step 4 is extracted center point coordinate according to the crestal line field of direction of fingerprint image; And/or step 5, according to the first discrete AM-FM or the second discrete AM-FM model output overall situation or the local fingerprint image.
The present invention compares with the conventional fingerprint feature extraction algorithm, has taked diverse system, and is simple and accurate.The present invention extracts the accuracy except improving minutiae point, also can accurately extract central point, reconstitutes the fingerprint image of any geometry deformation.
Description of drawings
Fig. 1 is a fingerprint image minutiae point extraction system process flow diagram provided by the invention;
Fig. 2 is that AM-FM provided by the invention extracts process flow diagram;
Fig. 3 is that process flow diagram is extracted in minutiae point provided by the invention position;
Fig. 4 is that minutiae point direction provided by the invention is extracted process flow diagram;
Fig. 5 is the fingerprint image of gathering;
Fig. 6 is the mould of the fingerprint image AM-FM model among Fig. 5;
Fig. 7 is the enhancing figure and the minutiae point mark of reconstruct;
Fig. 8 a and Fig. 8 b are direction field pattern and direction degree of confidence field pattern.
Embodiment
Fingerprint image minutiae point extraction system process flow diagram provided by the invention, as shown in Figure 1.Three major parts are arranged: the crestal line field of direction generates, the AM-FM model generates, minutiae point is extracted.Fingerprint image is the data inputs, three major parts of solid wire frame representation top layer, and the fine line frame table in the solid existing frame shows necessary module, and the fine dotted line frame table in the solid wire frame shows optional module, and rightmost frame of broken lines is represented expansion module.Fingerprint image produces the crestal line field of direction through crestal line field of direction generating portion, the crestal line direction is extracted as necessary module, if directly extract the level and smooth degree of the fingerprint image crestal line field of direction not enough (causing) by the noise on the fingerprint image, then need level and smooth module travel direction field, service orientation field level and smooth.The fingerprint image and the field of direction are extracted the input and the discrete AM-FM model of output of part as the AM-FM model, AM-FM is extracted as necessary module, it is output as discrete AM-FM model, when the discrete AM-FM model of direct extraction inadequately level and smooth (distortion by noise on the fingerprint image and finger print information causes) then use the level and smooth module of AM-FM to carry out smoothly, when input picture exists affine or distortion of projection need proofread and correct or use geometric transformation reconstructed module need carry out other geometric transformations the time.The AM-FM model and the field of direction are the inputs that minutiae point is extracted part, are output as the minutiae point sequence, comprise minutiae point coordinate and minutiae point direction.Wherein minutiae point position and direction determining are necessary module, and each carries out its corresponding function, the resolution reconstruct of the AM-FM model that disperses when the resolution of discrete AM-FM model is lower than desired minutiae point coordinate precision.It is input and output center point coordinate that one of expansion module, central point extract with the field of direction; Two of expansion module, image reconstruction serves as input and the output overall situation or local fingerprint image with the AM-FM model.
The extraction of the crestal line field of direction and the crestal line field of direction are level and smooth, exist Several Methods to use.What need statement is that the AM-FM model extracts the field of direction is extracted the error sensitivity, so must the superior field of direction of usability extract and smoothing method in the native system.The crestal line direction generating algorithm of outbound course degree of confidence is used in suggestion, if selected algorithm is not exported this numerical value, then regarding the output perseverance as is 1.Total outbound course field and direction degree of confidence field simultaneously of crestal line field of direction generation module like this.
The crestal line field of direction that is generated by fingerprint image shown in Figure 5 and direction degree of confidence field are shown in Fig. 8 a and Fig. 8 b.Field of direction gray scale 0-255 represents crestal line direction 0-π, and direction degree of confidence field gray scale 0-255 represents direction degree of confidence 0-1.Can observe, the direction of fingerprint field is finger-print region very level and smooth (0 and 255 is adjacent directions), has local steepness at the fingerprint central point, these most of just central point extracting method based on field of direction characteristic; Direction degree of confidence field is comparatively stable at finger-print region, and the crestal line directional curvature is big smaller so central point is neighbouring, and the background degree of confidence is 0.
AM-FM provided by the invention extracts flow process as shown in Figure 2.If certain partial fingerprint image crestal line direction is i, then from multiple Gabor bank of filters, take out the multiple Gabor wave filter of counterparty to i (crestal line direction and filter direction quadrature), this local fingerprint image is carried out filtering.Taking out a wave filter from precalculated bank of filters, is for the consideration that reduces complexity operation time, under insensitive condition to time complexity, also can press the multiple Gabor wave filter of constructing definitions according to direction.The filtering result is through the correction of direction degree of confidence, as the AM-FM model value of this local center point.All parts of whole fingerprint image just obtain discrete AM-FM model through above-mentioned processing.
Described being modified to will be multiply by the described output of Gabor wave filter again again after the increasing function mapping of direction degree of confidence with any process initial point.The simplest increasing function is an identity function.
Optional modules A M-FM smoothly based on: when amplitude function A (x) and both spectrum distribution non-overlapping copies of phase function Ω (x), can think that AM and FM are irrelevant.Fingerprint image meets this condition, so AM-FM's is level and smooth, because A (x) variation is small, is exactly the level and smooth of phase function Ω (x) in non-minutiae point zone; Then directly carry out smoothly in complex field in the minutiae point zone because A (x) changes big and crosses 0, Ω (x) this cross 0 discontinuous, but the AM-FM model this is continuous in complex field.
The principle of optional modules A M-FM geometric transformation reconstruct is, on the pairing discrete point coordinate of target geometric transformation, and the interpolation AM-FM model value that goes out to disperse.With level and smooth the same, utilize non-minutiae point zone A (x) constant Ω (x) linear change and minutiae point zone AM-FM model originally in the continuous characteristics of complex field, but the target of this module is an interpolation reconstruction.
Optional module resolution reconstruct also realizes by interpolation.In fact resolution reconstruct itself just is equivalent to a kind of geometric transformation.
The location determination module of minutiae point provided by the invention as shown in Figure 3.Discrete AM-FM model value is asked mould, and the result who asks mould as shown in Figure 6.There is minutiae point in coordinate by following 3 rules: 1, mould is less than threshold value, 2, mould is less than neighborhood, 3, mould in the direction difference (because the 2nd point, the direction difference just is all) of all directions all greater than threshold value.All coordinates are imported this determination module, deposit in the minutiae point coordinate sequence by the coordinate of judging, the sequential element value is a coordinate.Threshold value by fingerprint base to the test of algorithm to determine.Neighborhood is the pocket around the point, and it is inner for the circle at center generally to get this point on the mathematics, often gets this point and be the rectangle inside at center under discrete case.
Minutiae point direction determining module provided by the invention as shown in Figure 4.According to the coordinate in the minutiae point coordinate sequence, from the field of direction, take out the crestal line direction of this coordinate position, as the minutiae point direction sequence.Two sequences are merged into the minutiae point sequence, have indicated the position and the direction of each minutiae point.
The minutiae point that is labeled in the restructuring graph is seen Fig. 7, and wherein high bright spot is a minutiae point, and high bright spot is the direction of minutiae point to the direction of time bright spot.
Carrying out with the field of direction that central point extracts is the classic algorithm that fingerprint image is handled, and image reconstruction is exactly that AM-FM gets real part after with certain resolution reconstruct.These two modules point out that native system when extracting minutiae point, also can realize the out of Memory that traditional system can extract.
Those skilled in the art can also carry out various modifications to above content under the condition that does not break away from the definite the spirit and scope of the present invention of claims.Therefore scope of the present invention is not limited in above explanation, but determine by the scope of claims.
Claims (6)
1. a fingerprint thin node in fingerprint image extraction system is characterized in that, comprising:
Crestal line field of direction generation module comprises crestal line direction extraction module, is used for carrying out according to fingerprint image the crestal line field of direction extraction of fingerprint image;
AM-FM model generation module comprises the AM-FM extraction module, is used for extracting the first discrete AM-FM model according to the crestal line field of direction and the fingerprint image of fingerprint image; Described AM-FM extraction module comprises multiple Gabor wave filter generation module and filtering correcting module as a result, described multiple Gabor wave filter generation module, be used to generate the multiple Gabor wave filter of filter direction and local fingerprint image crestal line direction quadrature, and local fingerprint image is carried out filtering obtain the filtering result; Described filtering is correcting module as a result, to multiply by described filtering result again after the increasing function mapping of direction degree of confidence with any process initial point, the value that obtains is as the AM-FM model value of this local fingerprint image center, and this AM-FM model value is used to make up the first discrete AM-FM model;
The minutiae point extraction module comprises minutiae point location determination module and minutiae point direction determining module, is used for extracting minutiae point according to the crestal line field of direction of the first discrete AM-FM model and fingerprint image; Described minutiae point location determination module, after being used for the AM-FM model value of all coordinates asked mould, if: mould less than threshold value, mould less than the direction difference of neighborhood and mould in all directions reckling greater than threshold value, then there is minutiae point in the judgement of minutiae point location determination module, and output minutiae point coordinate sequence; Described minutiae point direction determining module is used for the coordinate according to the minutiae point coordinate sequence, and the crestal line direction of taking out this coordinate from the field of direction is exported the minutiae point direction sequence as the minutiae point direction.
2. fingerprint thin node in fingerprint image extraction system as claimed in claim 1 is characterized in that,
Crestal line field of direction generation module also comprises the level and smooth module of the field of direction, be used for the crestal line field of direction of fingerprint image is carried out smoothly, and the crestal line field of direction of the fingerprint image after will be level and smooth exports the AM-FM extraction module to as the crestal line field of direction of fingerprint image;
AM-FM model generation module also comprises level and smooth module of AM-FM and/or geometric transformation reconstructed module, be used for the first discrete AM-FM model of AM-FM extraction module output is carried out smoothly, and/or the first discrete AM-FM model of AM-FM extraction module output carried out interpolation, the output second discrete AM-FM model;
The minutiae point extraction module also comprises the resolution reconstructed module, is used for realizing resolution reconstruct by interpolation.
3. fingerprint thin node in fingerprint image extraction system as claimed in claim 2 is characterized in that, also comprises central point extraction module and/or image reconstruction module;
The central point extraction module is used for extracting center point coordinate according to the crestal line field of direction of fingerprint image;
The image reconstruction module is used for according to the first discrete AM-FM or the second discrete AM-FM model output overall situation or the local fingerprint image.
4. a fingerprint thin node in fingerprint image extracting method is characterized in that, comprising:
Step 1 is carried out the crestal line field of direction of fingerprint image and is extracted according to fingerprint image, generate the crestal line field of direction of fingerprint image;
Step 2 generates the multiple Gabor wave filter of filter direction and local fingerprint image crestal line direction quadrature, and local fingerprint image is carried out filtering obtains the filtering result; To multiply by described filtering result again after the increasing function mapping of direction degree of confidence with any process initial point, the value that obtains utilizes this AM-FM model value to make up the first discrete AM-FM model as the AM-FM model value of this local fingerprint image center;
Step 3, the AM-FM model value of all coordinates asked mould after, if: mould is then judged to have minutiae point greater than threshold value less than direction difference reckling in all directions of neighborhood and mould less than threshold value, mould, and output minutiae point coordinate sequence; According to the coordinate in the minutiae point coordinate sequence, the crestal line direction of taking out this coordinate from the field of direction is exported the minutiae point direction sequence as the minutiae point direction.
5. fingerprint thin node in fingerprint image extracting method as claimed in claim 4 is characterized in that,
In the step 1, also the crestal line field of direction of fingerprint image is carried out smoothly, and the crestal line field of direction of the fingerprint image after level and smooth is exported as the crestal line field of direction of fingerprint image;
In the step 2, also the first discrete AM-FM model is carried out smoothly, and/or the first discrete AM-FM model is carried out interpolation, the output second discrete AM-FM model;
In the step 3., realize resolution reconstruct by interpolation.
6. fingerprint thin node in fingerprint image extracting method as claimed in claim 4 is characterized in that, also comprises:
Step 4 is extracted center point coordinate according to the crestal line field of direction of fingerprint image; And/or
Step 5 is according to the first discrete AM-FM or the second discrete AM-FM model output overall situation or the local fingerprint image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102246142A CN101408939B (en) | 2008-10-21 | 2008-10-21 | Method and system for extracting fingerprint thin node in fingerprint image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102246142A CN101408939B (en) | 2008-10-21 | 2008-10-21 | Method and system for extracting fingerprint thin node in fingerprint image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101408939A CN101408939A (en) | 2009-04-15 |
| CN101408939B true CN101408939B (en) | 2011-04-06 |
Family
ID=40571947
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008102246142A Expired - Fee Related CN101408939B (en) | 2008-10-21 | 2008-10-21 | Method and system for extracting fingerprint thin node in fingerprint image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101408939B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853382B (en) * | 2010-05-18 | 2012-08-15 | 清华大学 | Method and device for acquiring direction of fingerprint |
| JP5669572B2 (en) * | 2010-12-28 | 2015-02-12 | ラピスセミコンダクタ株式会社 | Fingerprint authentication device |
| CN103824060A (en) * | 2014-02-28 | 2014-05-28 | 清华大学 | Method for extracting fingerprint detail points |
| CN103886296B (en) * | 2014-03-25 | 2017-02-15 | 清华大学 | Fingerprint identification method and device based on feedback |
| CN107392862A (en) * | 2017-06-29 | 2017-11-24 | 天津大学 | Based on Hilbert L2The enhancement method of fingerprint image of model |
| CN107679494B (en) * | 2017-09-30 | 2021-04-02 | 西安电子科技大学 | Fingerprint image matching method based on selective extension |
| CN107977653B (en) * | 2017-12-22 | 2022-08-09 | 云南大学 | Multi-scale singular point detection method combining complex filter and PI (proportional integral) value |
| CN109145810A (en) * | 2018-08-17 | 2019-01-04 | 中控智慧科技股份有限公司 | Details in fingerprint point detecting method, device, equipment, system and storage medium |
| CN112070032A (en) * | 2020-09-10 | 2020-12-11 | 哈尔滨理工大学 | Fingerprint minutiae acquiring method based on fingerprint phase gradient |
| CN113011284B (en) * | 2021-03-01 | 2023-05-26 | 杭州景联文科技有限公司 | Signature fingerprint identification method |
-
2008
- 2008-10-21 CN CN2008102246142A patent/CN101408939B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN101408939A (en) | 2009-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101408939B (en) | Method and system for extracting fingerprint thin node in fingerprint image | |
| Han et al. | Personal authentication using palm-print features | |
| US20100046810A1 (en) | Fingerprint image acquiring device, fingerprint authenticating apparatus, fingerprint image acquiring method, and fingerprint authenticating method | |
| Azmi et al. | Biometric signature verification system based on freeman chain code and k-nearest neighbor | |
| CN109102037A (en) | Chinese model training, Chinese image-recognizing method, device, equipment and medium | |
| US20100232659A1 (en) | Method for fingerprint template synthesis and fingerprint mosaicing using a point matching algorithm | |
| CN107066961B (en) | Fingerprint method for registering and device | |
| CN109977191A (en) | Problem map detection method, device, electronic equipment and medium | |
| CN101114340A (en) | VLSI realizing system and method of histogram equalization image processing | |
| CN106991380A (en) | A kind of preprocess method based on vena metacarpea image | |
| Sudiro et al. | Adaptable fingerprint minutiae extraction algorithm based-on crossing number method for hardware implementation using FPGA device | |
| KR100479338B1 (en) | Apparatus for verifying an online signature using of transform technique and method teherefor | |
| Alonso-Fernandez et al. | Fingerprint recognition | |
| CN116704557A (en) | Low-quality fingerprint matching method based on texture information | |
| Bartolo et al. | Scribbles to vectors: preparation of scribble drawings for CAD interpretation | |
| KR100958962B1 (en) | Apparatus and Method of fingerprint classification | |
| US20080123909A1 (en) | Method of transforming minutiae using taylor series for interoperable fingerprint recognition between disparate fingerprint sensors | |
| CN113409214B (en) | Rapid repairing method for damaged fingerprint image | |
| Shang et al. | Multifractal characteristics of palmprint and its extracted algorithm | |
| Estrozi et al. | 1D and 2D Fourier-based approaches to numeric curvature estimation and their comparative performance assessment | |
| JP4537097B2 (en) | Pattern detection method and apparatus | |
| CN113269159A (en) | Gesture recognition method fusing electromyographic signals and visual images | |
| Mane et al. | Signature matching with automated cheque system | |
| Karthic et al. | Feautre Extraction For Human Identifiction using Finger And Palm Images | |
| KR100675000B1 (en) | Apparatus and method for preprocessing fingerprint image |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20170426 Address after: 610041 high tech Zone, Chengdu province Tianfu Avenue North Section of the new century global center, No. 1700 E3-1-1603 Patentee after: Hefei formula Electronic Technology Co., Ltd. Address before: 610041 Sichuan, Chengdu hi tech Zone Tianfu Avenue South extension of the new high-tech incubator Park, building A-E-3, No. 1 Patentee before: Chengdu Finchos Electron Co., Ltd. |
|
| TR01 | Transfer of patent right | ||
| DD01 | Delivery of document by public notice | ||
| DD01 | Delivery of document by public notice |
Addressee: Liu Xiaodong Document name: payment instructions |
|
| DD01 | Delivery of document by public notice | ||
| DD01 | Delivery of document by public notice |
Addressee: Liu Xiaodong Document name: Notice of termination of patent right |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110406 Termination date: 20201021 |