CN104268533A - Non-contact revocable palm print bimodal authentication method - Google Patents
Non-contact revocable palm print bimodal authentication method Download PDFInfo
- Publication number
- CN104268533A CN104268533A CN201410529257.6A CN201410529257A CN104268533A CN 104268533 A CN104268533 A CN 104268533A CN 201410529257 A CN201410529257 A CN 201410529257A CN 104268533 A CN104268533 A CN 104268533A
- Authority
- CN
- China
- Prior art keywords
- template
- revocable
- palm
- veins
- palmmprint
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/13—Sensors therefor
- G06V40/1312—Sensors therefor direct reading, e.g. contactless acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a non-contact revocable palm print bimodal authentication method. A non-contact device is used for collection a palm print. Point set operation is adopted for preprocessing. A Gabor filter is adopted for extracting texture features. Random projection and quantization are conducted on the texture features and a random matrix with the participation of external variable parameters. A revocable template of a 2DPalmHash code is generated, and the generation direction of the revocable template is optimized according to a low correlation policy. During registration, the template of the bimodal 2DPalmHash code is stored in a database; during authentication, the template is extracted from the database and contrasted with authentication templates one by one, fractional order integration is conducted, and revocable palm print authentication is realized. According to the non-contact revocable palm print bimodal authentication method, the accuracy of the template is improved as the point set operation is adopted; the revocable template is given updating capacity; the feature template is optimized according to a proper strategy, and authentication precision is improved without increasing time and space complexity.
Description
Technical field
The present invention relates to the authentication method of biological characteristic, particularly the contactless revocable palm ridge bimodal authentication method of one.
Background technology
When contact palmmprint and palm arteries and veins harvester Long-Time Service, easily cause equipment surface to pollute, cause the problems such as precision reduction, health concerns.Although contactless collection overcomes above-mentioned defect, the preprocessing process of palmmprint and palm arteries and veins is easily subject to more multifactor interference, and such as, under natural scene, the situation such as background, illumination is complicated; Unfixed palm position makes shooting distance, angle and rotation amount disunity; Attitude (as finger closes up, the opens) disunity of palm.Therefore, the more high-precision Preprocessing Algorithm of design is needed to improve Position location accuracy.
Existing palmmprint and palm arteries and veins Verification System use primitive character template, there is many safety and privacy concern: when feature is stolen, cannot cancel and upgrade; Along with the popularization of range of application, multiple database may store same characteristic features template, if the feature templates in lower security level database is stolen, the feature templates of other databases will all lose efficacy; Primitive character may reveal the privacy information (as gene defect, disease etc.) of user.Therefore, need design and use shielded revocable biological characteristic.
The bimodal certification of palmmprint and palm arteries and veins can improve the antifalsification of authentication precision and biological characteristic, but in order to not increase characteristic amount and computing cost, must the suitable template of preference policy to two kinds of mode of design select.
Summary of the invention
For solving the problem, the object of the present invention is to provide a kind of contactless revocable palm ridge bimodal authentication method, using the method can realize high-precision contactless palmmprint and the pre-service of palm arteries and veins; The method adopts revocable biological characteristic, generates shielded feature templates, solves security and privacy problem; By suitable tactful preferred feature template, under the condition not increasing Time & Space Complexity, improve authentication precision.
For achieving the above object, the present invention realizes by the following technical solutions:
A kind of contactless revocable palm ridge bimodal authentication method, is characterized in that, mainly comprise the following steps:
One, the non-contact capture of palmmprint palm arteries and veins: use contactless palmmprint and palm arteries and veins harvester to gather palmmprint palm arteries and veins;
Two, the revocable template construct of palmmprint palm arteries and veins:
(1) point set operation, is adopted to carry out pre-service to palmmprint palm arteries and veins, translation, intersecting and merging computing are carried out to segmentation palm area point set, obtain palm and refer to root region, namely root point set is referred to, then according to constraint condition, finger root point set element is screened, obtain the accurate location that four refer to root point, select two reference points to set up corresponding coordinate system, Kinematic Positioning area-of-interest;
(2) Gabor filter, is adopted to carry out Gabor filtering, texture feature extraction to palmmprint palm arteries and veins;
(3), control pseudorandom number generator by outside variable element and generate stochastic matrix, by accidental projection and quantification structure irreversible transformation, change the textural characteristics of original palmmprint and palm arteries and veins into revocable form, i.e. two-dimentional palm Hash codes, also known as 2DPalmHash code, if revocable template is on the hazard, reset the outside variable element controlling stochastic matrix, generate new variable formwork, ensure that protected template possesses and replace and issue capability again;
(4), carry out preferably to the fusion template generation direction of palmmprint and palm arteries and veins: the covariance of adding up four direction coupling mark in often kind of mode respectively, according to little covariance selection strategy, from palmmprint and palm arteries and veins, the both direction of low correlation is selected to generate 2DPalmHash code as revocable template to be fused respectively, the coupling mark of multiple template is merged by average strategy, effective reduction coupling mark variance, realizes the certification of high precision revocable palm ridge bimodal;
Three, template registration: bimodal 2DPalmHash code mask is stored in database;
Four, template certification: extract registered template from database, with template to be certified comparison carry out the fusion of mark level one by one, realizes the certification of revocable palm ridge.
Further, described non-contact capture device is made up of acquisition port, camera, background light source, and described acquisition port is positioned at device outside, and described camera is positioned at device inside and is positioned at the below of acquisition port, described camera is two, and a camera directly gathers palmprint image; Another camera assists infrared fileter collection to slap arteries and veins image, background light source below described camera, described background light source is made up of three warm yellow light LED lamps and four infrared light compensating lamps, and warm yellow light LED lamp forms combination array with triarray around the mode of infrared light supply.
Further, described outside variable element is user key or smart card ID.
The present invention, owing to taking above technical scheme, has the following advantages:
(1) non-contact capture device avoids collecting device surface contamination, causes the shortcoming that authentication precision reduces; Overcome the hygienic issues that contact process causes, decrease the risk of communicate illness; Part traditional national culture forbids the same article surface of opposite sex contact, thus improves the applicability of Verification System in these ethnic mimority area.
(2) utilize volar direction information, directly region is operated, improve anti-interference and the accuracy of contactless palmmprint and the framing of palm arteries and veins.
(3) protected template possesses the contactless palm ridge authentication method of defeasibility.
Accompanying drawing explanation
Accompanying drawing 1, contactless revocable palm ridge bimodal Verification System framework of the present invention;
Accompanying drawing 2, contactless harvester design diagram;
Accompanying drawing 3, for left hand the palm arteries and veins " point set operation " location algorithm process;
Accompanying drawing 4,2DPalmHash code generative process;
The preferred revocable palm ridge template procedure of accompanying drawing 5, low correlation strategy;
Accompanying drawing 6, revocable palm ridge mark level merge certification.
Embodiment
Describe the present invention below in conjunction with drawings and Examples.
Fig. 1 is contactless revocable palm ridge bimodal Verification System framework of the present invention, adopts contactless palmmprint and palm arteries and veins harvester, introduces " point set operation " and carry out pre-service, adopt Gabor filter texture feature extraction.Control pseudorandom number generator by outside variable element and generate stochastic matrix.Textural characteristics and the stochastic matrix of palmmprint and palm arteries and veins carry out accidental projection and quantification, and generate revocable template, called after " two-dimentional palm Hash codes ", is also called " 2DPalmHash code ", preferred to revocable template generation direction by low correlation strategy.During registration, bimodal 2DPalmHash code mask is stored in database; During certification, from database, extract template, with the comparison carry out the fusion of mark level one by one of certification template, realize the certification of revocable palm ridge.
Fig. 2 is harvester design diagram, and as shown in the figure, this harvester acquisition port 1 is positioned at device outside, and acquisition camera 4 is arranged on device inside, and acquisition camera is two, and a camera directly gathers palmprint image; Another camera assists infrared fileter collection to slap arteries and veins image, the array structure of light source is had bottom the inner described acquisition camera of device, described array of source is made up of three warm yellow light LED lamps 2 and four infrared light compensating lamps 3, warm yellow light LED lamp 2 forms combination array with triarray around the mode of infrared light supply 3, and line outlet is positioned at device side.
Fig. 3 is for left hand palm arteries and veins image, and " point set operation " position fixing process of noncontact palmmprint and palm arteries and veins is described.During collection, palm the five fingers direction upwards.Three directions (to the right, upwards, left) translation is carried out to the palm area of segmentation, and calculates with original area respectively and occur simultaneously, obtain residual error.Three common factors (point set) calculate common factor again, obtain rough interdigital area territory.Obtain candidate root points region by regional choice, and calculate each regional barycenter as finger root point.Refer to that left and right palm is determined in the position of root points by four, and with the root point between the root point between forefinger and middle finger and nameless and little finger of toe for reference point, set up coordinate system.The length of side of cutting square area-of-interest is set to the distance between two reference points
m.Between two reference points line vertical line under 0.2
mplace, intercepting the length of side is
msquare as area-of-interest, and its size to be normalized.
Figure 4 shows that 2DPalmHash code generative process, Gabor filtering is carried out, texture feature extraction to palmmprint and palm arteries and veins.Controlled by outside variable element to generate the random number meeting Gaussian distribution, composition stochastic matrix.Gabor textural characteristics matrix and the stochastic matrix of palmmprint and palm arteries and veins carry out random projection transforms, generate Random Maps matrix, then carry out quantification generation 2DPalmHash code.Due to the compression of accidental projection, the element number of 2DPalmHash code is the half of textural characteristics matrix element number.Authentication performance is tested in following three kinds of situations.
(1) optimal cases: when whole user uses different outside variable elements and respective palmmprint (or palm arteries and veins) feature, 2DPalmHash code can reach desirable authentication precision.
(2) worst condition: when whole user uses identical outside variable element and respective palmmprint (or palm arteries and veins) feature, the authentication performance of 2DPalmHash code directly quantizes the PalmCode generated a little less than original Gabor textural characteristics.
(3) template renewal: when certain user uses different outside variable elements and the palmmprint of oneself (or palm arteries and veins) feature, the diversified 2DPalmHash code of satisfied enough diversity factoies can be generated, ensure that the success of original 2DPalmHash code is cancelled and upgrades the successful issue of 2DPalmHash code.
Fig. 5 is the process of the preferred revocable palm ridge template of low correlation strategy.When using multiple directions Gabor filter texture feature extraction, if interval, Gabor filter direction is excessive, then the discriminating that can not give full play to all directions is complementary; If interval too small, not only computation complexity is excessive, and the coupling fractional dependence of adjacent angular strengthens, and reduces syncretizing effect.Usually 4 or 6 direction Gabor filter texture feature extractions can be chosen.Balance complexity and accuracy requirement, the present invention to palmmprint and palm arteries and veins extract respectively 4 directions (0,
π/ 4,
π/ 2,3
π/ 4) Gabor textural characteristics.The coupling fractional statistics correlativity of the palmmprint code (PalmCode) after quantizing palmmprint 4 direction Gabor textural characteristics, the both direction selecting correlativity the most weak generates 2DPalmHash code.Also similar statistics and selection course are carried out to palm arteries and veins.4 the 2DPalmHash codes finally optimized constitute bimodal revocable palm ridge template.The coupling mark of multiple low correlation template carries out organically blending and effectively can reduce variance, improves authentication performance.
Fig. 6 is that revocable palm ridge mark level merges authentication framework.During registration, bimodal revocable palm ridge template is stored in database; During certification, from database, extract corresponding templates, with 4 preferred 2DPalmHash code comparisons respectively of certification template, calculate Hamming distances and measure diversity factor as coupling mark.To 4 coupling fractional statistics arithmetic mean numbers, realize mark level and merge bimodal certification.Its authentication performance is better than the syncretizing effect of lower 4 the direction 2DPalmHash code coupling marks of palmmprint (or palm arteries and veins) single mode, thus under the condition not increasing Time & Space Complexity, achieve the high precision bimodal certification that revocable palm ridge mark level merges.
Claims (3)
1. a contactless revocable palm ridge bimodal authentication method, is characterized in that, mainly comprise the following steps:
One, the non-contact capture of palmmprint palm arteries and veins: use contactless palmmprint and palm arteries and veins harvester to gather palmmprint palm arteries and veins;
Two, the revocable template construct of palmmprint palm arteries and veins:
(1) point set operation, is adopted to carry out pre-service to palmmprint palm arteries and veins, translation, intersecting and merging computing are carried out to segmentation palm area point set, obtain palm and refer to root region, namely root point set is referred to, then according to constraint condition, finger root point set element is screened, obtain the accurate location that four refer to root point, select two reference points to set up corresponding coordinate system, Kinematic Positioning area-of-interest;
(2) Gabor filter, is adopted to carry out Gabor filtering, texture feature extraction to palmmprint palm arteries and veins;
(3), control pseudorandom number generator by outside variable element and generate stochastic matrix, by accidental projection and quantification structure irreversible transformation, change the textural characteristics of original palmmprint and palm arteries and veins into revocable form, i.e. 2DPalmHash code, if revocable template is on the hazard, reset and control the outside variable element of stochastic matrix, generate new variable formwork, ensure that protected template possesses and replace and issue capability again;
(4), carry out preferably to the fusion template generation direction of palmmprint and palm arteries and veins: the covariance of adding up four direction coupling mark in often kind of mode respectively, according to little covariance selection strategy, from palmmprint and palm arteries and veins, the both direction of low correlation is selected to generate 2DPalmHash code as revocable template to be fused respectively, the coupling mark of multiple template is merged by average strategy, effective reduction coupling mark variance, realizes the certification of high precision revocable palm ridge bimodal;
Three, template registration: bimodal 2DPalmHash code mask is stored in database;
Four, template certification: extract registered template from database, with template to be certified comparison carry out the fusion of mark level one by one, realizes the certification of revocable palm ridge.
2. the contactless revocable palm ridge bimodal authentication method of one according to claim 1, it is characterized in that, described non-contact capture device is made up of acquisition port, camera, background light source, described acquisition port is positioned at device outside, described camera is positioned at device inside and is positioned at the below of acquisition port, described camera is two, and a camera directly gathers palmprint image; Another camera assists infrared fileter collection to slap arteries and veins image, background light source below described camera, described background light source is made up of three warm yellow light LED lamps and four infrared light compensating lamps, and warm yellow light LED lamp forms combination array with triarray around the mode of infrared light supply.
3. the contactless revocable palm ridge bimodal authentication method of one according to claim 1, it is characterized in that, the outside variable element described in step 2 is user key or smart card ID.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410529257.6A CN104268533A (en) | 2014-10-10 | 2014-10-10 | Non-contact revocable palm print bimodal authentication method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410529257.6A CN104268533A (en) | 2014-10-10 | 2014-10-10 | Non-contact revocable palm print bimodal authentication method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104268533A true CN104268533A (en) | 2015-01-07 |
Family
ID=52160053
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410529257.6A Pending CN104268533A (en) | 2014-10-10 | 2014-10-10 | Non-contact revocable palm print bimodal authentication method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104268533A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105141424A (en) * | 2015-08-14 | 2015-12-09 | 南昌航空大学 | Key extraction method of palm print-palm vein double-source fuzzy vault |
| CN107958210A (en) * | 2017-11-20 | 2018-04-24 | 济南大学 | A kind of palm grain identification method based on decimal system map locating table |
| CN108960003A (en) * | 2017-05-18 | 2018-12-07 | 济南大学 | Based on Gabor and the palm print characteristics of chaotic maps generate and authentication method |
| CN110012034A (en) * | 2019-05-10 | 2019-07-12 | 济南大学 | Biometric authentication method under cloud environment with secret protection |
| CN110516594A (en) * | 2019-08-27 | 2019-11-29 | 安徽大学 | A kind of guard method and its protective device for referring to vein and can cancelling feature templates |
| CN110580783A (en) * | 2019-09-19 | 2019-12-17 | 京麦智能科技(苏州)有限公司 | Intelligent locker control system based on palm vein and palm print dual-mode identity authentication |
| CN112651304A (en) * | 2020-12-11 | 2021-04-13 | 西安电子科技大学 | Revocable palm print template generation method, device, equipment and storage medium based on feature fusion |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1664848A (en) * | 2005-03-24 | 2005-09-07 | 上海邦震科技发展有限公司 | Non-contact type self-feedback iris image acquisition device |
-
2014
- 2014-10-10 CN CN201410529257.6A patent/CN104268533A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1664848A (en) * | 2005-03-24 | 2005-09-07 | 上海邦震科技发展有限公司 | Non-contact type self-feedback iris image acquisition device |
Non-Patent Citations (6)
| Title |
|---|
| GOH KAH ONG MICHAEL ETAL.: "A contactless biometric system using palm print and palm vein features", 《PROC OF ADVANCED BIOMETRIC TECHNOLOGIES》 * |
| LU LENG ETAL.: "Logical Conjunction of Triple-perpendicular-directional Translation Residual for Contactless Palmprint Preprocessing", 《2014 11TH INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGY: NEW GENERATION》 * |
| YAGIZ SUTCU ETAL.: "A secure biometric authentication scheme based on robust hashing", 《IN ACM CONFERENCE ON THE WORKSHORP ON MULTIMEDIA AND SECURITY》 * |
| 冷璐: "可撤除掌纹识别与认证技术研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
| 冷璐等: "可撤销掌纹脉二维PalmHash码的共轭认证", 《光电子激光》 * |
| 杨汉辉等: "一种可撤销指纹身份认证方案", 《福建电脑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105141424A (en) * | 2015-08-14 | 2015-12-09 | 南昌航空大学 | Key extraction method of palm print-palm vein double-source fuzzy vault |
| CN105141424B (en) * | 2015-08-14 | 2018-08-07 | 南昌航空大学 | A kind of palm print and palm vein double source obscures the key extraction method of safety box |
| CN108960003A (en) * | 2017-05-18 | 2018-12-07 | 济南大学 | Based on Gabor and the palm print characteristics of chaotic maps generate and authentication method |
| CN107958210A (en) * | 2017-11-20 | 2018-04-24 | 济南大学 | A kind of palm grain identification method based on decimal system map locating table |
| CN110012034A (en) * | 2019-05-10 | 2019-07-12 | 济南大学 | Biometric authentication method under cloud environment with secret protection |
| CN110012034B (en) * | 2019-05-10 | 2021-06-01 | 济南大学 | Biological characteristic authentication method with privacy protection function in cloud environment |
| CN110516594A (en) * | 2019-08-27 | 2019-11-29 | 安徽大学 | A kind of guard method and its protective device for referring to vein and can cancelling feature templates |
| CN110516594B (en) * | 2019-08-27 | 2022-03-18 | 安徽大学 | Protection method and protection device for finger vein feature template capable of being cancelled |
| CN110580783A (en) * | 2019-09-19 | 2019-12-17 | 京麦智能科技(苏州)有限公司 | Intelligent locker control system based on palm vein and palm print dual-mode identity authentication |
| CN112651304A (en) * | 2020-12-11 | 2021-04-13 | 西安电子科技大学 | Revocable palm print template generation method, device, equipment and storage medium based on feature fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104268533A (en) | Non-contact revocable palm print bimodal authentication method | |
| CN102542258B (en) | Imaging device based on finger biometric information and multimoding identity recognition method | |
| US9095285B2 (en) | Portable biometric identification device using a dorsal hand vein pattern | |
| CN103106401B (en) | Mobile terminal iris recognition device with human-computer interaction mechanism | |
| CN108573203B (en) | Identity authentication method and device and storage medium | |
| CN107004114A (en) | Blood-vessel image filming apparatus and individual authentication system | |
| TWI407377B (en) | Biometric information processing device | |
| US9289160B2 (en) | Portable biometric identification device using a dorsal hand vein pattern | |
| CN103246883B (en) | A kind of underground coal mine thermal infrared images face identification method | |
| WO2017059591A1 (en) | Finger vein identification method and device | |
| US8577094B2 (en) | Image template masking | |
| CN103886283A (en) | Method for fusing multi-biometric image information for mobile user and application thereof | |
| CN103559489B (en) | Palm feature extracting method under a kind of noncontact imaging mode | |
| CN106127181A (en) | One is virtual tries manicure method, system on | |
| JP2010244365A (en) | Device and method for providing service, server and method for authenticating biometric information, program, and service providing system | |
| CN106326327A (en) | Methods and apparatuses for updating user authentication data | |
| KR102558741B1 (en) | Device and method to register user | |
| CN104063679B (en) | Blood-vessel image filming apparatus | |
| US20150154474A1 (en) | Biometric information registration device and method | |
| CN102592115A (en) | Hand positioning method and system | |
| JP6617570B2 (en) | Biometric authentication device, biometric authentication method, and biometric authentication program | |
| WO2019024350A1 (en) | Biometric recognition method and apparatus | |
| CN111104859A (en) | Authentication method and system based on multispectral identification | |
| US20140375786A1 (en) | Obtaining a biometric image and handling damaged biometrics | |
| WO2010084963A1 (en) | Image processing device, biometric authentication device, image processing method, and recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150107 |
|
| RJ01 | Rejection of invention patent application after publication |