CN103268473A - Three-dimension finger print image ellipsoid fitting processing method - Google Patents
Three-dimension finger print image ellipsoid fitting processing method Download PDFInfo
- Publication number
- CN103268473A CN103268473A CN2013101429522A CN201310142952A CN103268473A CN 103268473 A CN103268473 A CN 103268473A CN 2013101429522 A CN2013101429522 A CN 2013101429522A CN 201310142952 A CN201310142952 A CN 201310142952A CN 103268473 A CN103268473 A CN 103268473A
- Authority
- CN
- China
- Prior art keywords
- ellipsoid
- fingerprint image
- fingerprint
- finger print
- print image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a three-dimension finger print image ellipsoid fitting processing method which includes the following steps. Firstly, an obtained three-dimension finger print image is arranged on an ellipsoid reference model in a fitting mode and the size and the position of an ellipsoid are confirmed. Secondly, a linear stretching and projecting method is used for projecting the ellipsoid model onto a two-dimension plane. Finally, anamorphose of a projected image is rectified, detailed ridge information is abstracted and a corresponding two-dimension finger print image is finally obtained. As the ellipsoid model is selected as the fitting reference model of the three-dimension finger print image, a fitting zone is larger than a spheric reference model and a columnar reference model, affection by pressure intensity of fingers, dry fingers, wet fingers and the like does not exist and the obtained corresponding two-dimension finger print image is closer to a real finger print image. Thus, precision of the obtained finger print image is high. Compared with a novel method according to which a three-dimension finger print is stretched into a two-dimension finger print image, the three-dimension finger print image ellipsoid fitting processing method does not need enormous data calculating and processing so that a system can rapidly identify and the method is convenient to popularize and use.
Description
Technical field
The invention belongs to the fingerprint identification technology field, relate to image process of fitting treatment method, specifically is a kind of three-dimensional fingerprint image ellipsoid process of fitting treatment method.
Background technology
Along with development of times, society more and more needs efficiently, identification system reliably.Traditional personal identification discriminating means such as key, password, password, identity document, even recognition method such as IC-card, because they have the weakness that can palm off, can forge, can usurp, can decode, the needs that modern society's economic activity and social safety are taken precautions against can not have been satisfied fully; The continuous maturation of recognition technology and fast development of computer technology make various identification systems based on human body physiological characteristics as: recognition technologies such as fingerprint, palm, sound, retina, pupil and face line are come out of from the laboratory one after another.
Fingerprint has uniqueness, unchangeable property, carries and makes things convenient for characteristics such as acquisition mode easily, make fingerprint recognition become first of the bio-identification personal identification, simultaneously also be that present technology is the most ripe, the simplest authentication way of realization, be widely used in many fields such as public security, finance, security, be the checking means that have authority of law, have versatility in the world, its development receives much concern always.
Traditional fingerprint identification technology is to obtain two-dimentional fingerprint image and it is handled and identifies, this two-dimentional fingerprint image is generally pressed by the user or the finger that rolls obtains, such acquisition method makes the fingerprint skin deformation easily, causes different and different along with user's the dynamics of pressing of the fingerprint image quality of gathering and direction; In addition, because the influence of doing environment such as wet degree of fingerprint skin also can cause the uncontinuity, non-renewable of fingerprint, therefore identical fingerprint can increase the complicacy of fingerprint matching in each the collection, and the performance of system has been played negative influence.In order to solve the corresponding problem in the two-dimentional fingerprint recognition process, three-dimensional fingerprint image is suggested gradually.In the scope of applicant retrieval, as follows based on the pertinent literature information of the disposal route of three-dimensional fingerprint image:
Yongchang Wang, Daniel L.Lau, Laurence G.Hassebrook has proposed to obtain in " the match ball of three-dimensional fingerprint image launches and performance evaluation " (Fit-sphere unwrapping and performance analysis of3D fingerprints) literary composition three-dimensional fingerprint image is fitted to the disposal route of spheroid.This method mainly is that finger is approximately spheroid, and the three-dimensional fingerprint that obtains fits to spheroid, projects to two dimensional surface then, obtains corresponding two-dimentional fingerprint image, and corresponding two-dimentional fingerprint image is handled.It is limited that but the match of this method spheroid and fingerprint is counted, and it is less to compare, and fitting effect is not good, makes that resulting respective two-dimensional fingerprint image error is bigger.
Yongchang Wang, Laurence G.Hassebrook, Daniel L.Lau has proposed the structure light scan method in " data of three-dimensional fingerprint are obtained and handled " (Data Acquisition and Processing of3-D Fingerprints) literary composition.This method is incorporated into the Structured Illumination method in the system equipment, obtains the depth information that the higher pictorial information of albedo obtains the fingerprint lines simultaneously.Then finger is approximately right cylinder, the three-dimensional fingerprint image that obtains is fitted to right cylinder, and project on the two dimensional surface, obtain corresponding fingerprint image.But this method is not preserved the relative distance of fingerprint table millet cake, thereby has introduced the horizontal distortion that launches fingerprint, and the respective two-dimensional fingerprint image error that obtains is bigger.
Sara Shafaei, Tamer Inanc, Laurence G.Hassebrook have proposed a kind ofly to launch the method for three-dimensional fingerprint based on three-dimensional surface curvature analysis method in " a kind of three-dimensional fingerprint expands into the new method that two-dimensional phase is answered fingerprint " (A New Approach to Unwrap a3-D Fingerprint to a2-DRolled Equivalent Fingerprint) literary composition.This method uses " spring algorithm " that level and smooth three-dimensional fingerprint surface is changed into two-dimentional fingerprint, and uses the texture of curvature analysis calculated fingerprint.But this method will will be calculated its texture by several equations to each point of fingerprint image, and calculated amount is very big, and processing speed is slow, is not easy to promote the use of.
By above document as can be seen: existing three-dimensional fingerprint identification technology all is contactless, has avoided distortion that traditional two-dimentional fingerprint recognition system causes and distortion etc.But the disposal route of existing three-dimensional fingerprint image at first makes calculated amount very big, and the processing time is longer; Secondly, the model match of selecting in the approximating method is counted limited, makes resulting two-dimensional phase answer the error of fingerprint image bigger, thereby has a strong impact on the treatment effect of three-dimensional fingerprint image.Thereby when avoiding two-dimentional fingerprint conventional identification techniques defective, handling the three-dimensional fingerprint image that obtains how fast, efficiently and accurately is a difficult problem of being badly in need of solution.
Summary of the invention
The objective of the invention is to overcome the deficiency that existing three-dimensional fingerprint image treatment technology calculated amount is big, the result images error is bigger etc., propose the disposal route of a kind of three-dimensional fingerprint image ellipsoid match.In the problem that this method is brought owing to pressure inequality, finger dry and wet state etc. in avoiding the conventional fingerprint identifying, the three-dimensional fingerprint image that obtains is realized handling fast, accurately, its precision height, speed is fast, and is convenient to promote.
In order to achieve the above object, technical scheme of the present invention is:
A kind of three-dimensional fingerprint image ellipsoid process of fitting treatment method may further comprise the steps:
1) the three-dimensional fingerprint image that will obtain is fitted on the ellipsoidal model, determines size and the position of ellipsoidal model;
2) use the linear expansion projecting method that ellipsoidal model is projected to two dimensional surface, obtain undressed two-dimentional fingerprint image;
3) distortion correction step 2) in the two-dimentional fingerprint image is also extracted detailed ridge information, finally obtains the clearly demarcated two-dimentional fingerprint image of detailed ridge paddy.
Ellipsoid match described in the step 1) is specially: at first set an ellipsoidal model, be expressed as:
The lip-deep point of fingerprint (x then
k, y
k, z
k) and the lip-deep point of ellipsoidal model (x
c, y
c, z
c) between distance table be shown:
d=[(x
k-x
c)
2+(y
k-y
c)
2+(z
k-z
c)]
1/2
Secondly, at least three adjacent points on the print image, the range equation between above-mentioned 2 are tried to achieve a bit (x that satisfies apart from the d ellipsoidal model hour by least square method
c, y
c, z
c), accordingly, at least six points on the print image are tried to achieve other 2 points on the ellipsoid, according to three point coordinate on the ellipsoidal model of trying to achieve, determine parameter a, the b in the ellipsoidal model, the value of c, finally determine size and the position of ellipsoidal model; Simultaneously with rectangular coordinate system (x
k, y
k, z
k) be converted to cartesian coordinate system (θ
k, φ
k, ρ
k), wherein, θ
kAnd φ
kBe Rad, ρ
kBe k distance of putting the ellipsoid central point on the fingerprint.
Step 2) determine the size of perspective view before the linear expansion projecting method described in, projection earlier, length and the width of perspective view are respectively L
1And L
2, wherein
In the following formula, max (θ
k) and max (φ
k) be θ
kAnd φ
kMaximal value, min (θ
k) and min (φ
k) be θ
kAnd φ
kMinimum value, t
θBe the step value of θ direction, t
φStep value for the φ direction;
The linear expansion process comprises the projection of both direction, is respectively the projection of θ horizontal direction and φ vertical direction, and wherein, the projection of θ and φ direction is expressed as respectively:
Wherein, l
1Be the pixel value of fingerprint θ direction, and l
1=1,2 ..., L
1, l
2Be the pixel value of fingerprint φ direction, and l
2=1,2 ..., L
2, θ
MinBe the minimum value in the θ direction radian value, φ
MinIt is the minimum value in the φ direction radian value;
In the linear expansion process, fingerprint adopts up-sampling, comes preservation information with this; To each the point (l on two perspective views
1, l
2) corresponding grid value be (
); After the value at each some place of perspective view is obtained, can obtain the respective two-dimensional fingerprint image after the projection.
The described correction distortion of step 3) and the method for extracting ridge information are specially: at first use low-pass filter to step 2) the two-dimentional fingerprint image that obtains carries out filtering, and according to linear interpolation method, along continuous straight runs is adjusted the projection of θ direction, makes the size of perspective view by L
1* L
2Be adjusted into J
1* L
2, and
J
1Be expressed as the length of adjusting the back perspective view; Equally, vertically adjust the projection of φ direction, the size of perspective view is by J
1* L
2Be adjusted into J
1* J
2, and
J
2Be expressed as the width of adjusting the back perspective view, make linear projection become non-linear projection, and fingerprint adopts Downsapling method in this process, to reduce noise, reduce distortion, preserve ridge information; At last by gauss low frequency filter and band-pass filter, and filtered image is carried out histogram equalization obtain the clearly demarcated two-dimentional fingerprint image of detailed ridge paddy.
Compared with prior art, the present invention has following beneficial effect:
The present invention is owing to select ellipsoidal model as the match reference model of three-dimensional fingerprint image, compare spherical and cylindrical reference model, the zone of match is bigger, and be not subjected to the influence of finger presses dynamics, dry and wet state etc., the respective two-dimensional fingerprint image and the actual fingerprint image that obtain are more approaching, so the precision height of gained fingerprint image; Secondly, the present invention expands into the method for two-dimentional fingerprint image than three-dimensional fingerprint, does not need sizable data computation and processing, and the system that makes can identify fast, is convenient to promote the use of.
Description of drawings
The finger contours synoptic diagram that Fig. 1 extracts for the present invention;
Fig. 2 carries out synoptic diagram after the ellipsoid match for the present invention;
Fig. 3 is the comparison diagram of the present invention and traditional spheroidal fitting effect.
Embodiment
The invention will be further described below in conjunction with accompanying drawing:
Referring to Fig. 1 and Fig. 2, at first carry out the Ellipsoidal Surface match, the three-dimensional fingerprint image that obtains is fitted on the ellipsoidal model, determine size and the position of ellipsoidal model; Secondly, the linear expansion projection namely uses the linear expansion projecting method that ellipsoidal model is projected to two dimensional surface, obtains undressed two-dimentional fingerprint image; At last, correct step 2) in the two-dimentional fingerprint image distortion and extract detailed ridge information, finally obtain the clearly demarcated two-dimentional fingerprint image of detailed ridge paddy.So both can be complementary with traditional two-dimentional fingerprint database, can use traditional disposal route to two-dimentional fingerprint image again.As shown in Figure 3, compare spherical approximating method, the match point of ellipsoid fitting process is more, and fitted area is big, and the respective two-dimensional fingerprint image that the match of use ellipsoid obtains is more near real fingerprint image, and precision is higher.
The concrete following steps that the present invention includes:
1) ellipsoid match
At first from the former figure of finger-image, extract the profile of finger, as shown in Figure 1, namely extract three-dimensional fingerprint image; Then, as shown in Figure 2, the three-dimensional fingerprint image that obtains is fitted on the ellipsoidal model that configures in advance, determines size and the position of ellipsoidal model.Concrete grammar is as follows: at first set an ellipsoidal model, be expressed as:
The lip-deep point of fingerprint (x then
k, y
k, z
k) and the lip-deep point of ellipsoidal model (x
c, y
c, z
c) between distance table be shown:
d=[(x
k-x
c)
2+(y
k-y
c)
2+(z
k-z
c)]
1/2
Secondly, at least three adjacent points on the print image, the range equation between above-mentioned 2 are tried to achieve a bit (x that satisfies apart from the d ellipsoidal model hour by least square method
c, y
c, z
c), accordingly, at least six points on the print image are tried to achieve other 2 points on the ellipsoid, according to three point coordinate on the ellipsoidal model of trying to achieve, determine parameter a, the b in the ellipsoidal model, the value of c, finally determine size and the position of ellipsoidal model; Simultaneously with rectangular coordinate system (x
k, y
k, z
k) be converted to cartesian coordinate system (θ
k, φ
k, ρ
k), wherein, θ
kAnd φ
kBe Rad, ρ
kBe k distance of putting the ellipsoid central point on the fingerprint.
2) linear expansion projection
Use the linear expansion projecting method that ellipsoidal model is projected to two dimensional surface, obtain undressed two-dimentional fingerprint image; Determine the size of perspective view before the projection earlier, length and the width of perspective view are respectively L
1And L
2, wherein
In the following formula, max (θ
k) and max (φ
k) be θ
kAnd φ
kMaximal value, min (θ
k) and min (φ
k) be θ
kAnd φ
kMinimum value, t
θBe the step value of θ direction, t
φStep value for the φ direction;
The linear expansion process comprises the projection of both direction, is respectively the projection of θ horizontal direction and φ vertical direction, and wherein, the projection of θ and φ direction is expressed as respectively:
Wherein, l
1Be the pixel value of fingerprint θ direction, and l
1=1,2 ..., L
1, l
2Be the pixel value of fingerprint φ direction, and l
2=1,2 ..., L
2, θ
MinBe the minimum value in the θ direction radian value, φ min is the minimum value in the φ direction radian value;
In the linear expansion process, fingerprint adopts up-sampling, comes preservation information with this; To each the point (l on two perspective views
1, l
2) corresponding grid value be (
); After the value at each some place of perspective view is obtained, can obtain the respective two-dimensional fingerprint image after the projection.
3) correct distortion and extraction ridge information
Correction step 2) distortion in the two-dimentional fingerprint image is also extracted detailed ridge information, finally obtains the clearly demarcated two-dimentional fingerprint image of detailed ridge paddy.In order to reduce noise, at first using low-pass filter to step 2) the two-dimentional fingerprint image that obtains carries out filtering, and according to linear interpolation method, along continuous straight runs is adjusted the projection of θ direction, makes the size of perspective view by L
1* L
2Be adjusted into J
1* L
2, and
J
1Be expressed as the length of adjusting the back perspective view; Equally, vertically adjust the projection of φ direction, the size of perspective view is by J
1* L
2Be adjusted into J
1* J
2, and
J
2Be expressed as the width of adjusting the back perspective view, make linear projection become non-linear projection, and fingerprint adopts Downsapling method in this process, to reduce noise, reduce distortion, preserve ridge information; At last by gauss low frequency filter and band-pass filter, and filtered image is carried out histogram equalization obtain the clearly demarcated two-dimentional fingerprint image of detailed ridge paddy.
Above content is to further describing that the present invention does in conjunction with concrete preferred implementation; can not assert that the specific embodiment of the present invention only limits to this; for the general technical staff of the technical field of the invention; without departing from the inventive concept of the premise; can also make some simple deduction or replace, all should be considered as belonging to the present invention and determine scope of patent protection by claims of submitting to.
Claims (4)
1. a three-dimensional fingerprint image ellipsoid process of fitting treatment method is characterized in that, may further comprise the steps:
1) the three-dimensional fingerprint image that will obtain is fitted on the ellipsoidal model, determines size and the position of ellipsoidal model;
2) use the linear expansion projecting method that ellipsoidal model is projected to two dimensional surface, obtain undressed two-dimentional fingerprint image;
3) distortion correction step 2) in the two-dimentional fingerprint image is also extracted detailed ridge information, finally obtains the clearly demarcated two-dimentional fingerprint image of detailed ridge paddy.
2. three-dimensional fingerprint image ellipsoid process of fitting treatment method according to claim 1 is characterized in that the ellipsoid match described in the step 1) is specially: at first set an ellipsoidal model, be expressed as:
The lip-deep point of fingerprint (x then
k, y
k, z
k) and the lip-deep point of ellipsoidal model (x
c, y
c, z
c) between distance table be shown:
d=[(x
k-x
c)
2+(y
k-y
c)
2+(z
k-a
c)]
1/2
Secondly, at least three adjacent points on the print image, the range equation between above-mentioned 2 are tried to achieve a bit (x that satisfies apart from the d ellipsoidal model hour by least square method
c, y
c, z
c), accordingly, at least six points on the print image are tried to achieve other 2 points on the ellipsoid, according to three point coordinate on the ellipsoidal model of trying to achieve, determine parameter a, the b in the ellipsoidal model, the value of c, finally determine size and the position of ellipsoidal model; Simultaneously with rectangular coordinate system (x
k, y
k, z
k) be converted to cartesian coordinate system (θ
k, φ
k, ρ
k), wherein, θ
kAnd φ
kBe Rad, ρ
kBe k distance of putting the ellipsoid central point on the fingerprint.
3. three-dimensional fingerprint image ellipsoid process of fitting treatment method according to claim 2 is characterized in that step 2) described in the linear expansion projecting method, determine the size of perspective view before the projection earlier, length and the width of perspective view are respectively L
1And L
2, wherein
In the following formula, max (θ
k) and max (φ
k) be θ
kAnd φ
kMaximal value, min (θ
k) and min (φ
k) be θ
kAnd φ
kMinimum value, t
θBe the step value of θ direction, t
φStep value for the φ direction;
The linear expansion process comprises the projection of both direction, is respectively the projection of θ horizontal direction and φ vertical direction, and wherein, the projection of θ and φ direction is expressed as respectively:
Wherein, l
1Be the pixel value of fingerprint θ direction, and l
1=1,2 ..., L
1, l
2Be the pixel value of fingerprint φ direction, and l
2=1,2 ..., L
2, θ
MinBe the minimum value in the θ direction radian value, φ
MinIt is the minimum value in the φ direction radian value;
In the linear expansion process, fingerprint adopts up-sampling, comes preservation information with this; To each the point (l on two perspective views
1, l
2) corresponding grid value be (
); After the value at each some place of perspective view is obtained, can obtain the respective two-dimensional fingerprint image after the projection.
4. three-dimensional fingerprint image ellipsoid process of fitting treatment method according to claim 3, it is characterized in that, the described correction distortion of step 3) and the method for extracting ridge information are specially: at first use low-pass filter to step 2) the two-dimentional fingerprint image that obtains carries out filtering, according to linear interpolation method, along continuous straight runs is adjusted the projection of θ direction, makes the size of perspective view by L
1* L
2Be adjusted into J
1* L
2, and
J
1Be expressed as the length of adjusting the back perspective view; Equally, vertically adjust the projection of φ direction, the size of perspective view is by J
1* L
2Be adjusted into J
1* J
2, and
J
2Be expressed as the width of adjusting the back perspective view, make linear projection become non-linear projection, and fingerprint adopts Downsapling method in this process, to reduce noise, reduce distortion, preserve ridge information; At last by gauss low frequency filter and band-pass filter, and filtered image is carried out histogram equalization obtain the clearly demarcated two-dimentional fingerprint image of detailed ridge paddy.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013101429522A CN103268473A (en) | 2013-04-23 | 2013-04-23 | Three-dimension finger print image ellipsoid fitting processing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013101429522A CN103268473A (en) | 2013-04-23 | 2013-04-23 | Three-dimension finger print image ellipsoid fitting processing method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103268473A true CN103268473A (en) | 2013-08-28 |
Family
ID=49012100
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2013101429522A Pending CN103268473A (en) | 2013-04-23 | 2013-04-23 | Three-dimension finger print image ellipsoid fitting processing method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103268473A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104484679A (en) * | 2014-09-17 | 2015-04-01 | 北京邮电大学 | Non-standard gun shooting bullet trace image automatic identification method |
CN106934777A (en) * | 2017-03-10 | 2017-07-07 | 北京小米移动软件有限公司 | Scan image acquisition methods and device |
CN109902569A (en) * | 2019-01-23 | 2019-06-18 | 上海思立微电子科技有限公司 | Conversion method, device and the fingerprint identification method of fingerprint image |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040008875A1 (en) * | 2002-07-09 | 2004-01-15 | Miguel Linares | 3-D fingerprint identification system |
CN102509094A (en) * | 2011-11-25 | 2012-06-20 | 哈尔滨工业大学深圳研究生院 | Structured-light-based embedded 3D (three dimensional) fingerprint collecting method and system |
CN102880860A (en) * | 2012-08-31 | 2013-01-16 | 成都国腾实业集团有限公司 | Laser-scanning-based three-dimensional fingerprint acquisition device |
-
2013
- 2013-04-23 CN CN2013101429522A patent/CN103268473A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040008875A1 (en) * | 2002-07-09 | 2004-01-15 | Miguel Linares | 3-D fingerprint identification system |
CN102509094A (en) * | 2011-11-25 | 2012-06-20 | 哈尔滨工业大学深圳研究生院 | Structured-light-based embedded 3D (three dimensional) fingerprint collecting method and system |
CN102880860A (en) * | 2012-08-31 | 2013-01-16 | 成都国腾实业集团有限公司 | Laser-scanning-based three-dimensional fingerprint acquisition device |
Non-Patent Citations (1)
Title |
---|
YONGCHANG WANG ET AL: "fit-sphere unwrapping and performance analysis of 3D fingerprints", 《APPLIED OPTICS》, vol. 49, no. 4, 1 February 2010 (2010-02-01), pages 592 - 597, XP001551412, DOI: doi:10.1364/AO.49.000592 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104484679A (en) * | 2014-09-17 | 2015-04-01 | 北京邮电大学 | Non-standard gun shooting bullet trace image automatic identification method |
CN104484679B (en) * | 2014-09-17 | 2017-12-22 | 北京邮电大学 | Non- standard rifle shooting warhead mark image automatic identifying method |
CN106934777A (en) * | 2017-03-10 | 2017-07-07 | 北京小米移动软件有限公司 | Scan image acquisition methods and device |
CN106934777B (en) * | 2017-03-10 | 2020-07-14 | 北京小米移动软件有限公司 | Scanning image acquisition method and device |
CN109902569A (en) * | 2019-01-23 | 2019-06-18 | 上海思立微电子科技有限公司 | Conversion method, device and the fingerprint identification method of fingerprint image |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yuan et al. | Fingerprint liveness detection using an improved CNN with image scale equalization | |
Kanhangad et al. | A unified framework for contactless hand verification | |
Chen et al. | 3D free-form object recognition in range images using local surface patches | |
Sheela et al. | Iris recognition methods-survey | |
US10762366B2 (en) | Finger vein identification method and device | |
Pflug et al. | Ear biometrics: a survey of detection, feature extraction and recognition methods | |
CN103268497B (en) | A kind of human face posture detection method and the application in recognition of face | |
CN105335722A (en) | Detection system and detection method based on depth image information | |
CN110287918B (en) | Living body identification method and related product | |
CN104573634A (en) | Three-dimensional face recognition method | |
CN104036254A (en) | Face recognition method | |
Frejlichowski et al. | The west pomeranian university of technology ear database–a tool for testing biometric algorithms | |
CN105760841A (en) | Identify recognition method and identify recognition system | |
CN104102903A (en) | Secondary face recognition method based on SRC | |
TW201211913A (en) | A method for recognizing the identity of user by palm vein biometric | |
CN104123547B (en) | Based on improved trend pass filtering and flexible matched recognition methods | |
CN109977764A (en) | Vivo identification method, device, terminal and storage medium based on plane monitoring-network | |
CN103268473A (en) | Three-dimension finger print image ellipsoid fitting processing method | |
CN103324921A (en) | Mobile identification method based on inner finger creases and mobile identification equipment thereof | |
CN109598235B (en) | Finger vein image authentication method and device | |
CN104573628A (en) | Three-dimensional face recognition method | |
Tewari et al. | Fingerprint recognition and feature extraction using transform domain techniques | |
Sapkale et al. | A finger vein recognition system | |
Benziane et al. | Dorsal hand vein identification based on binary particle swarm optimization | |
Zhang et al. | Finger vein recognition based on Gabor filter |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130828 |