CN104820828A - Identity authentication method based on multi-directional mixed features of dual inner phalangeal prints - Google Patents
Identity authentication method based on multi-directional mixed features of dual inner phalangeal prints Download PDFInfo
- Publication number
- CN104820828A CN104820828A CN201510219816.8A CN201510219816A CN104820828A CN 104820828 A CN104820828 A CN 104820828A CN 201510219816 A CN201510219816 A CN 201510219816A CN 104820828 A CN104820828 A CN 104820828A
- Authority
- CN
- China
- Prior art keywords
- feature
- finger
- phalangeal
- formula
- identity authentication
- 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.)
- Granted
Links
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The invention relates to an identity authentication method based on multi-directional mixed features of dual inner phalangeal prints and is mainly involved in Gabor conversion, a horizontal local binary pattern, and a score level fusion strategy. An authentication flow mainly includes five parts: finger image collection, inner phalangeal print feature positioning, inner phalangeal print feature extraction, score level feature fusion, and decision matching. According to the invention, two print features of a single finger are used for designing an identity authentication method, thereby overcoming a defect of a single print feature of a single inner knuckle. With fully utilization of the biological features of the single finger and the Gabor and horizontal local binary pattern, multi-directional mixed features are extracted, so that a defect that is caused by only utilization of the unidirectional Gabor feature according to the traditional method can be overcome and the inner phalangeal prints can be used for identity authentication information loss. Therefore, safe and reliable identity authentication can be realized.
Description
Technical field
The invention belongs to Digital Image Processing, mode identification technology, relate to the new method of living things feature recognition authentication techniques, particularly a kind of identity identifying method based on the multi-direction composite character of two interior phalangeal configurations.
Background technology
Identity verify and the authentication method of the mankind roughly experienced by three phases.Early stage mainly by the method for article (Possession, What you have), such as I.D., passport, magnetic card, key etc. are as the proof of identification under different occasion.The method of article is easily lost, damage, stolen and imitated, baffled faitour is had an opportunity to take advantage of.Along with the development of electronic information technology, there is Knowledge based engineering method (Knowledge, What you know), as password (Token), password (Password) etc., these class methods are easily forgotten, or easily victim cracks.
Under this background, the novel unique advantage had because of it based on the identity identifying method of biological characteristic, overcome front two generation identification authentication mode defect, be day by day subject to the favor of people.Compared with traditional mode, the mankind can not lose, forget oneself biological characteristic, and therefore more biometric identity Verification System has more security, reliability and validity, all have broad application prospects and huge economic benefit in many security fields.
The articulations digitorum manus junction of finger inner side, interior phalangeal configurations position, not easy to wear.For individual in other phalangeal configurations covered by other objects, peel, have the skin of scar, some physical labourers that callus is more on hand etc. and cannot identification injure skin identification, can not be interfered on the whole, there is not the low problem of discrimination; Insensitive to temperature and humidity, less demanding to finger cleanliness; Less demanding to acquisition resolution.Current existing research also proves that interior phalangeal configurations can be used as a kind of biological characteristic for authentication.It is convenient that interior phalangeal configurations gathers, and is particularly useful for user's handheld mobile device, and is applicable to the multiple biological characteristic joint qualification of hand, have vast potential for future development.
Interior phalangeal configurations has attracted the sight of numerous researcher with its outstanding advantage, but the inferior position of interior phalangeal configurations also exposes gradually.Interior phalangeal configurations structure is single, and texture is simple, and along with the raising of safety requirements, the feature of single phalangeal configurations is difficult to meet.Interior phalangeal configurations multiple features fusion is a kind of thinking solving this problem.But the interior phalangeal configurations Limited Number of individual, the use of multiple interior phalangeal configurations will consume personal biology characteristics resource greatly.At present, the existing research for interior phalangeal configurations all fails to overcome the above problems preferably.
Summary of the invention
The present invention mainly solves the technical matters existing for prior art; Provide one and utilize single finger first phalanges and the second phalanges, the identity identifying method based on the multi-direction composite character of two interior phalangeal configurations between the inner side texture of the second phalanges and the 3rd phalanges junction and two textures.
Above-mentioned technical matters of the present invention is mainly solved by following technical proposals:
Based on an identity identifying method for the multi-direction composite character of two interior phalangeal configurations, it is characterized in that, comprise the following steps:
The step that an original image gathers: specifically hand images collection, collected by camera comprises the hand images of complete finger and palm, this process entails finger closes up parallel, palm is perpendicular to lens direction, palm is parallel with minute surface, under collection environment is placed in even smooth black background, avoid light position on the impact of recognition effect;
One is carried out pretreated step for above-mentioned original image: pre-service is intended to locate the area-of-interest for feature extraction, specifically comprises:
S1, locates single finger areas: select any one in middle finger, the third finger and forefinger, according to the prior imformation of finger position, intercepts key point locating area image f from original palmprint image f (x, y)
1(x, y), this edge, left and right, region is in the second knuckle of selected finger, and lower edges is on adjacent two fingers of selected finger; Use large rule algorithm picks appropriate threshold binaryzation f
1(x, y) obtains f
2(x, y); To f
2(x, y) projects along webs direction radon, by two border key points of the selected finger in histogram extreme value location; Use the prior imformation of two borders and the 3rd phalangeal region to be partitioned into and comprise complete middle finger first knuckle line and second knuckle line region f
3(x, y);
S2, phalangeal configurations area-of-interest in location is two: to image f
3(x, y) vertically carries out radon projection, and the extreme point according to two regions of projection can judge first knuckle line and second knuckle line position, thus is partitioned into the area-of-interest R (x, y) comprising first knuckle line and second knuckle line;
One is carried out the step of feature extraction for above-mentioned pretreated image: the hybrid feature extraction specifically on phalangeal configurations 3 directions, specifically comprises:
S1, global characteristics extracts: to area-of-interest R (x, y) respectively from 5 °, 0 ° ,-5 ° are carried out Gabor transformation, obtain RG
1(x, y), RG
2(x, y), RG
3(x, y);
S2, local shape factor: to RG
1(x, y), RG
2(x, y), RG
3the texture maps in (x, y) three directions carries out horizontal LBP computing, obtains Local textural feature;
S3, blocked histogram: be divided into mutually disjoint sub-block to Local textural feature, the histogram feature of each sub-block of phalangeal configurations in calculating respectively, the histogram feature of contiguous block obtains composite character HRG
1(x, y), HRG
2(x, y), HRG
3(x, y);
S4, feature coding: use DP-SMAWK to choose N number of optimal threshold, carries out coding to connection histogram feature and obtains code figure P
1(x, y), P
2(x, y), P
3(x, y);
A feature for said extracted carries out the step of characteristic matching fusion, specifically comprises:
S1, characteristic matching: mated respectively at the template in database by 3 feature templates, matching way adopts normalized Hamming distance;
H (a, b)=min (| a-b|, N-|a-b|) formula two
P wherein
i(x, y) is phalangeal configurations feature in test, Q
i(x, y) is phalangeal configurations feature in database, i=1,2,3;
S2, score normalization: because the coupling score distribution of all directions is different, need to be normalized; The parameter of normalization process needs to train on the database on certain basis to obtain, and normalization mode is as follows:
D represents D
ithe true coupling of category feature and false coupling distribution, D* represents D
ithe true coupling distribution of category feature, α represents training parameter, i=1,2,3;
S3, mark merges: fractional layer amalgamation mode uses weighted registration, and formula is as follows
Dis=W
1d
1'+W
2d
2'+W
3d
3' formula four
Described weights W
ichoose according to error rate (EER) e such as the training of each adaptation
idetermine, i=1,2,3, computing formula is as follows:
The step that a matching result generates: be exactly specifically match decision, according to the distance according to training
The optimal threshold chosen judges the correctness of coupling, thus judges whether user passes through certification.
Therefore, tool of the present invention has the following advantages: fully use the singlehanded biological characteristic referred to, and use Gabor and horizontal local binary patterns to extract multidirectional composite character and overcome the loss that interior phalangeal configurations that tradition only uses one direction Gabor characteristic to cause can be used for authentication information, realize safe and reliable authentication.
Accompanying drawing explanation
Fig. 1 is method flow diagram of the present invention.
Fig. 2 is algorithm structure process flow diagram of the present invention.
Embodiment
Below by embodiment, and by reference to the accompanying drawings, technical scheme of the present invention is described in further detail.
Embodiment:
As shown in drawings, the present invention is based on phalangeal configurations feature extraction algorithm in Gabor and horizontal LBP, and it is divided into five major parts: finger-image collection, interior phalangeal configurations feature location, interior phalangeal configurations feature extraction, fractional layer Fusion Features, and decision-making is mated.
For middle finger phalangeal configurations, specific implementation process is as follows:
T1: hand images collection, collected by camera comprises the hand images of complete finger and palm, and this process entails finger closes up parallel, palm is perpendicular to lens direction, palm is parallel with minute surface, under collection environment is placed in even smooth black background, avoids light position on the impact of recognition effect.
T2: pre-service is intended to locate the area-of-interest for feature extraction, and concrete steps comprise:
S1: locate single finger areas.
According to the prior imformation of finger position, from original palmprint image f (x, y), intercept key point locating area image f
1(x, y), this edge, left and right, region is in middle finger second knuckle, and lower edges is on forefinger and the third finger.Use large rule algorithm picks appropriate threshold binaryzation f
1(x, y) obtains f
2(x, y).To f
2(x, y) projects along webs direction radon, by two border key points of histogram extreme value location middle finger.Use the prior imformation of two borders and the 3rd phalangeal region to be partitioned into and comprise complete middle finger first knuckle line and second knuckle line region f
3(x, y).
S2: phalangeal configurations area-of-interest in location is two
To image f
3(x, y) vertically carries out radon projection, and the extreme point according to two regions of projection can judge first knuckle line and second knuckle line position, thus is partitioned into the area-of-interest R (x, y) comprising first knuckle line and second knuckle line.
T3: feature extraction, the hybrid feature extraction on interior phalangeal configurations 3 directions.
S1: global characteristics extracts.
To area-of-interest R (x, y) respectively from 5 °, 0 ° ,-5 ° are carried out Gabor transformation, obtain RG
1(x, y), RG
2(x, y), RG
3(x, y).
S2: local shape factor.
To RG
1(x, y), RG
2(x, y), RG
3the texture maps in (x, y) three directions carries out horizontal LBP computing, obtains Local textural feature.
S3: blocked histogram.
Be divided into mutually disjoint sub-block to Local textural feature, the histogram feature of each sub-block of phalangeal configurations in calculating respectively, the histogram feature of contiguous block obtains composite character HRG
1(x, y), HRG
2(x, y), HRG
3(x, y).
S4: feature coding.
Use DP-SMAWK to choose N number of optimal threshold, coding is carried out to connection histogram feature and obtains code figure P
1(x, y), P
2(x, y), P
3(x, y).
T4: characteristic matching merges.
S1: characteristic matching.
Mated respectively at the template in database by 3 feature templates, matching way adopts normalized Hamming distance.
h(a,b)＝min(|a-b|,N-|a-b|) (2)
P wherein
i(x, y) is phalangeal configurations feature in test, Q
i(x, y) is phalangeal configurations feature in database, i=1,2,3.
S2: score normalization.
Because the coupling score distribution of all directions is different, need to be normalized.The parameter of normalization process needs to train on the database on certain basis to obtain, and normalization mode is as follows:
D represents D
ithe true coupling of category feature and false coupling distribution, D* represents D
ithe true coupling distribution of category feature, α represents training parameter, i=1,2,3.
S3: mark merges.
Fractional layer amalgamation mode uses weighted registration, and formula is as follows
dis＝W
1·D
1'+W
2·D
2'+W
3·D
3' (4)
Weights W
ichoose according to error rate (EER) e such as the training of each adaptation
idetermine, i=1,2,3, computing formula is as follows:
T5: match decision.
Judge the correctness of coupling according to the optimal threshold chosen according to the distance of training, thus judge whether user passes through certification.
Specific embodiment described herein is only to the explanation for example of the present invention's spirit.Those skilled in the art can make various amendment or supplement or adopt similar mode to substitute to described specific embodiment, but can't depart from spirit of the present invention or surmount the scope that appended claims defines.
Claims (1)
1., based on an identity identifying method for the multi-direction composite character of two interior phalangeal configurations, it is characterized in that, comprise the following steps:
The step that an original image gathers: specifically hand images collection, collected by camera comprises the hand images of complete finger and palm, this process entails finger closes up parallel, palm is perpendicular to lens direction, palm is parallel with minute surface, under collection environment is placed in even smooth black background, avoid light position on the impact of recognition effect;
One is carried out pretreated step for above-mentioned original image: pre-service is intended to locate the area-of-interest for feature extraction, specifically comprises:
S1, locates single finger areas: select any one in middle finger, the third finger and forefinger, according to the prior imformation of finger position, intercepts key point locating area image f from original palmprint image f (x, y)
1(x, y), this edge, left and right, region is in the second knuckle of selected finger, and lower edges is on adjacent two fingers of selected finger; Use large rule algorithm picks appropriate threshold binaryzation f
1(x, y) obtains f
2(x, y); To f
2(x, y) projects along webs direction radon, by two border key points of the selected finger in histogram extreme value location; Use the prior imformation of two borders and the 3rd phalangeal region to be partitioned into and comprise complete middle finger first knuckle line and second knuckle line region f
3(x, y);
S2, phalangeal configurations area-of-interest in location is two: to image f
3(x, y) vertically carries out radon projection, and the extreme point according to two regions of projection can judge first knuckle line and second knuckle line position, thus is partitioned into the area-of-interest R (x, y) comprising first knuckle line and second knuckle line;
One is carried out the step of feature extraction for above-mentioned pretreated image: the hybrid feature extraction specifically on phalangeal configurations 3 directions, specifically comprises:
S1, global characteristics extracts: to area-of-interest R (x, y) respectively from 5 °, 0 ° ,-5 ° are carried out Gabor transformation, obtain RG
1(x, y), RG
2(x, y), RG
3(x, y);
S2, local shape factor: to RG
1(x, y), RG
2(x, y), RG
3the texture maps in (x, y) three directions carries out horizontal LBP computing, obtains Local textural feature;
S3, blocked histogram: be divided into mutually disjoint sub-block to Local textural feature, the histogram feature of each sub-block of phalangeal configurations in calculating respectively, the histogram feature of contiguous block obtains composite character HRG
1(x, y), HRG
2(x, y), HRG
3(x, y);
S4, feature coding: use DP-SMAWK to choose N number of optimal threshold, carries out coding to connection histogram feature and obtains code figure P
1(x, y), P
2(x, y), P
3(x, y);
A feature for said extracted carries out the step of characteristic matching fusion, specifically comprises:
S1, characteristic matching: mated respectively at the template in database by 3 feature templates, matching way adopts normalized Hamming distance;
H (a, b)=min (| a-b|, N-|a-b|) formula two
P wherein
i(x, y) is phalangeal configurations feature in test, Q
i(x, y) is phalangeal configurations feature in database, i=1,2,3;
S2, score normalization: because the coupling score distribution of all directions is different, need to be normalized; The parameter of normalization process needs to train on the database on certain basis to obtain, and normalization mode is as follows:
D represents D
ithe true coupling of category feature and false coupling distribution, D* represents D
ithe true coupling distribution of category feature, α represents training parameter, i=1,2,3;
S3, mark merges: fractional layer amalgamation mode uses weighted registration, and formula is as follows
Dis=W
1d
1'+W
2d
2'+W
3d
3' formula four
Described weights W
ichoose according to error rate (EER) e such as the training of each adaptation
idetermine, i=1,2,3, computing formula is as follows:
The step that a matching result generates: be exactly specifically match decision, judges the correctness of coupling, thus judges whether user passes through certification according to the optimal threshold chosen according to the distance of training.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510219816.8A CN104820828B (en) | 2015-04-30 | 2015-04-30 | A kind of identity identifying method based on double interior multi-direction composite characters of phalangeal configurations |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510219816.8A CN104820828B (en) | 2015-04-30 | 2015-04-30 | A kind of identity identifying method based on double interior multi-direction composite characters of phalangeal configurations |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104820828A true CN104820828A (en) | 2015-08-05 |
| CN104820828B CN104820828B (en) | 2018-05-11 |
Family
ID=53731118
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510219816.8A Active CN104820828B (en) | 2015-04-30 | 2015-04-30 | A kind of identity identifying method based on double interior multi-direction composite characters of phalangeal configurations |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104820828B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107392160A (en) * | 2017-07-27 | 2017-11-24 | 北京小米移动软件有限公司 | Optical finger print recognition methods and device, computer-readable recording medium |
| CN107516088A (en) * | 2017-09-02 | 2017-12-26 | 宜宾学院 | A kind of more finger segments lines recognition methods |
| CN107578009A (en) * | 2017-09-02 | 2018-01-12 | 宜宾学院 | The recognition methods of more finger tip interphalangeal joint lines |
| CN108764127A (en) * | 2018-05-25 | 2018-11-06 | 京东方科技集团股份有限公司 | Texture Recognition and its device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100177937A1 (en) * | 2009-01-15 | 2010-07-15 | Lei Zhang | Method and system for identifying a person using their finger-joint print |
| CN103324921A (en) * | 2013-06-28 | 2013-09-25 | 华南理工大学 | Mobile identification method based on inner finger creases and mobile identification equipment thereof |
| CN104112125A (en) * | 2014-07-24 | 2014-10-22 | 大连大学 | Method for identity recognition based on palm print and finger crease feature fusion |
-
2015
- 2015-04-30 CN CN201510219816.8A patent/CN104820828B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100177937A1 (en) * | 2009-01-15 | 2010-07-15 | Lei Zhang | Method and system for identifying a person using their finger-joint print |
| CN103324921A (en) * | 2013-06-28 | 2013-09-25 | 华南理工大学 | Mobile identification method based on inner finger creases and mobile identification equipment thereof |
| CN104112125A (en) * | 2014-07-24 | 2014-10-22 | 大连大学 | Method for identity recognition based on palm print and finger crease feature fusion |
Non-Patent Citations (2)
| Title |
|---|
| GUANGWEI GAO ET AL.: "Intergration of multiple orientation and texture information for finger-knuckle-print verification", 《NEUROCOMPUTING》 * |
| LORIS NANNI ET AL.: "A multi-matcher system based on knuckle-based features", 《NEURAL COMPUTING & APPLICATIONS》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107392160A (en) * | 2017-07-27 | 2017-11-24 | 北京小米移动软件有限公司 | Optical finger print recognition methods and device, computer-readable recording medium |
| US10885298B2 (en) | 2017-07-27 | 2021-01-05 | Beijing Xiaomi Mobile Software Co., Ltd. | Method and device for optical fingerprint recognition, and computer-readable storage medium |
| CN107516088A (en) * | 2017-09-02 | 2017-12-26 | 宜宾学院 | A kind of more finger segments lines recognition methods |
| CN107578009A (en) * | 2017-09-02 | 2018-01-12 | 宜宾学院 | The recognition methods of more finger tip interphalangeal joint lines |
| CN107578009B (en) * | 2017-09-02 | 2020-04-10 | 宜宾学院 | Method for identifying lines of multi-finger distal interphalangeal joints |
| CN107516088B (en) * | 2017-09-02 | 2020-05-22 | 宜宾学院 | Multi-knuckle grain identification method |
| CN108764127A (en) * | 2018-05-25 | 2018-11-06 | 京东方科技集团股份有限公司 | Texture Recognition and its device |
| US11170515B2 (en) | 2018-05-25 | 2021-11-09 | Boe Technology Group Co., Ltd. | Texture recognition method and apparatus, and computer-readable storage medium thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104820828B (en) | 2018-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fei et al. | Local discriminant direction binary pattern for palmprint representation and recognition | |
| Valdes-Ramirez et al. | A Review of Fingerprint Feature Representations and Their Applications for Latent Fingerprint Identification: Trends and Evaluation. | |
| CN108960039B (en) | Irreversible fingerprint template encryption method based on symmetric hash | |
| CN102254165B (en) | Hand back vein identification method based on fusion of structural coding features and texture coding features | |
| CN101246543B (en) | Examiner identity identification method based on bionic and biological characteristic recognition | |
| Su et al. | Fingerprint indexing with pose constraint | |
| CN107169479A (en) | Intelligent mobile equipment sensitive data means of defence based on fingerprint authentication | |
| Alheeti | Biometric iris recognition based on hybrid technique | |
| CN102629316B (en) | Image enhancement method in automatic fingerprint identification technology | |
| CN104112115A (en) | Three-dimensional face detection and identification technology | |
| CN103886283A (en) | Method for fusing multi-biometric image information for mobile user and application thereof | |
| CN103268497A (en) | Gesture detecting method for human face and application of gesture detecting method in human face identification | |
| CN104820828A (en) | Identity authentication method based on multi-directional mixed features of dual inner phalangeal prints | |
| CN101794374A (en) | Method and system for identifying a person using their finger-joint print | |
| CN102629320A (en) | Ordinal measurement statistical description face recognition method based on feature level | |
| Krishneswari et al. | A review on palm print verification system | |
| CN103679136A (en) | Hand back vein identity recognition method based on combination of local macroscopic features and microscopic features | |
| CN102938055A (en) | Hand bone identification system | |
| CN112686191B (en) | Living body anti-counterfeiting method, system, terminal and medium based on three-dimensional information of human face | |
| CN107958208A (en) | A kind of fingerprint crossing storehouse matching method based on propagation algorithm | |
| CN103886303A (en) | Palmprint recognition method and device | |
| Ramakrishnan et al. | An efficient automatic attendance system using fingerprint reconstruction technique | |
| Xia et al. | Building instance mapping from ALS point clouds aided by polygonal maps | |
| CN102819754A (en) | Fingerprint score fusion system and method based on Sigmoid expansion | |
| Akulwar et al. | Secured multi modal biometric system: a review |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |