CN102043961B - Vein feature extraction method and method for carrying out identity authentication by utilizing double finger veins and finger-shape features - Google Patents
Vein feature extraction method and method for carrying out identity authentication by utilizing double finger veins and finger-shape features Download PDFInfo
- Publication number
- CN102043961B CN102043961B CN2010105779061A CN201010577906A CN102043961B CN 102043961 B CN102043961 B CN 102043961B CN 2010105779061 A CN2010105779061 A CN 2010105779061A CN 201010577906 A CN201010577906 A CN 201010577906A CN 102043961 B CN102043961 B CN 102043961B
- Authority
- CN
- China
- Prior art keywords
- finger
- vein
- image
- point
- type feature
- 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
Images
Abstract
The invention relates to a method for carrying out identity authentication by utilizing double finger veins and finger-shape features, comprising the following steps of: firstly registering and storing vein features and information of finger-shape features of double fingers of a user; collecting, extracting and authenticating the vine features and the information of the finger-shape features of double fingers of a customer; respectively comparing the extracted double finger vein features and finger-shape features with the stored double finger vein features and information of the finger-shape features, and carrying out decision-level fusion; collecting images of hand shapes and finger veins, establishing a rectangular coordinate system by taking an intersection point between the double fingers as an original point and determining a region of interest (ROI); and finally respectively extracting the vein features and the finger-shape features, comparing with the stored double finger veins and information of the finger-shape features, and carrying out decision-level fusion. In the invention, two kinds of information of veins and finger shapes can be respectively extracted from one image and the two kinds of information are fused; and by introducing the coordinate system with the intersection point of the two fingers as the original point, the method provided by the invention not only improves the positioning precision, but also is convenient for users to use and effectively improves the system performance through double feature fusion.
Description
Technical field
The present invention relates to a kind of vena characteristic extracting method and utilize biological characteristic to carry out authentication method, particularly a kind of vena characteristic extracting method and utilize finger vena and method that hand-shaped characteristic carries out authentication.
Background technology
Present social people are more and more higher to the security requirement of information, and how a people's of precise Identification identity, protection information safety, become a crucial social concern that must solve.Traditional authentication, owing to very easily forging and losing, more and more is difficult to meet social demand, and the most convenient and safe solution is exactly biometrics identification technology undoubtedly at present.Biological characteristic is the fake certification scheme that the physical traits such as figure and features according to the people, sound are carried out the science of authentication.Living things feature recognition is exactly, according to unique biological characteristic between each individuality, it is identified to the authentication with identity.
Biometrics identification technology more and more is applied to authentication and identification, as adopt biological identification and the recognition system of the features such as face picture, fingerprint, iris, but no matter be fingerprint or iris face picture, these features all are exposed to surface, be easy to change, such as with advancing age, all change in various degree can be arranged.Finger vena is as a kind of biological characteristic that is positioned at body interior, many-sided superiority is arranged: 1) there is very strong universality and uniqueness, the variation of essence can not occur, different people's vein image differences in most people's vein blood vessel with advancing age; 2) vein is positioned at body interior, is difficult to forge or change by operation; 3) epidermis characteristic can not affect vein pattern; 4) contactless collection, minimizing people's resentment.
Although finger vein features has its superiority, but certain defect is also arranged simultaneously, vein pattern be by vein blood vessel from tripleplane to two-dimentional feature, certain variation occurs in the two dimensional character image that the rotation of very little angle all can make, and affects its First Pass Yield.The finger vein features lines are less, are difficult for determining identical region of interest ROI (Region of Interest).These have brought difficulty all to single finger vena identification and identification.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of method of extracting vein pattern from the vein gray level image, and the method can not be subject to the impact of vein tendency to extract the vein network.Contrast some extracting method in the past, do not need from different directions extract minutiae respectively, then do stack or merge.This method can disposablely be extracted the good vein network of effect.
Another problem that the present invention solves is to provide a kind of two methods that refer to that vein and finger-type feature are carried out authentication of utilizing, the method had both added the advantage of vein pattern, reduced again the reject rate of azygos vein feature, solved in azygos vein feature verification process simultaneously, finger is fixing bad, the problem that need to repeatedly authenticate.
In order to address the above problem, the present invention by the following technical solutions:
A kind of method of extracting vein pattern from the vein gray level image, on the vein gray level image gathered, corresponding region of interest ROI is done N * N matrix diagram, the central point of a little take in outer ring, as symmetric points divide into groups in twos, is divided into as 2N-2 group point, every group of point and the gray-scale value of central point are compared; In 2N-2 group point, if there is the gray-scale value of M group point all to be greater than the gray-scale value of central point, window center point is the vein pattern point, and all unique points that obtain form the vein network image; Vein network image to gained is done skeletonizing, obtains single pixel network characterization, then does two-dimensional dct transform, extracts the feature of the low frequency part of two-dimensional dct transform as the vein network.
A kind of methods of utilizing two finger veins and finger-type feature to carry out authentication, at first two vein patterns that refer to that registration storage user V-arrangement is placed and the information of finger-type feature, then gather two vein patterns that refer to of authentication client V-arrangement placement and the information of finger-type feature, respectively the information of the vein pattern of the vein pattern of collection and finger-type feature and storage and finger-type feature is compared, the level of making decision merges.Further, when storage, extraction, utilize the near infrared collector to gather two vein image and finger-type images that refer to that V-arrangement is placed, extract point of crossing between two fingers, determine respectively vein image region of interest ROI and finger-type interesting image regions ROI, then extract vein network characterization and finger-type feature, in Feature Conversion with after merging, again and the feature of having stored contrasted, obtain final result of determination.
Further, described collect two are referred to vein image and finger-type image carry out pre-service, described collect two are referred to image adopts medium filtering and histogram transformation to carry out pre-service.
The image pre-service comprises two steps, medium filtering and histogram transformation.Due to the singularity of vein paddy shape feature, salt-pepper noise is larger on its impact, tends to extract some isolated noise spots, therefore at first do medium filtering twice, reduce salt-pepper noise, but this process also has corresponding impact to vein pattern.Next image is done respectively to histogram transformation, respectively outstanding profile and two kinds of features of vein.
Medium filtering is based on a kind of nonlinear signal processing technology that can effectively suppress noise of sequencing statistical theory, and two dimension median filter is output as:
g(x,y)=med{f(x-k,y-l),(k,l∈W)},
Wherein, f (x, y), g (x, y) is respectively original image and processes rear image.The two dimension pattern plate that W is 15 * 15.Here do medium filtering twice to gathering the image come, effectively removed the noise of image.
If image is not added to processing, directly extract contour feature and can bring a lot of interference, the high light corrosion of finger edge and some textures in background all can be taken as profile and extract, and real outline line so just is not easily distinguishable out.
Concrete steps are as follows:
(1) Probability p (x) that each gray level of computed image occurs.
(2)p(x)=p(x)×(x/125);
(3) 0~255 calculate the probability sum lower than each gray-scale value appearance of this gray-scale value successively.This sequence is q (x), x ∈ [0,255]
(4) histogram transformation.
k(m,n)=q(x)×255,
K (m, n) is that on image, horizontal ordinate is m, and ordinate is the gray-scale value that n is ordered, and x is the front gray-scale value of pixel value conversion for this reason.
Further, extract respectively vein pattern and two kinds of information of finger-type feature in the described two finger vein images that collect and finger-type image:
To the image after pre-service, extract hand edge contour information with the Canny algorithm, select two fingers to refer to that the extreme point at nest place is that true origin is set up rectangular coordinate system as two ,Yi point of crossing, finger point of crossing, determine vein image region of interest ROI and contour images region of interest ROI;
After obtaining the contour images region of interest ROI, in contour images, point left hand edge with fitting a straight line, the fitting a straight line that obtains of take makes fitting a straight line rotate to the vertical direction parallel with the coordinate Y-axis as benchmark rotation translation image, and with the position of region of interest ROI left hand edge fixed range; Vein image follows that contour images is done same translational movement and the anglec of rotation is adjusted;
Further, in the image after above-mentioned adjustment, extract the contour feature of the finger-type of finger-type transverse width sequence from profile interesting image regions ROI:
For the finger edge information in the contour images region of interest ROI, carry out Feature Conversion, become the characteristic information that is easy to comparison and storage.The present invention extracts the lateral order width information of finger.
Extract the vein pattern of DCT low frequency from the vein image region of interest ROI:
The gray level image of finger vena has following characteristics: the xsect intensity profile of finger vena is paddy shape, more approaches the lowest point, and gray-scale value is lower.
Make like this window of a N * N on gray level image, started by the upper left corner, a pixel of a pixel slides into the lower right corner line by line.When vein is arranged in window, some points of window surrounding outmost turns, its gray-scale value will be greater than the gray-scale value of central point.
On the vein gray level image gathered, corresponding region of interest ROI is made the window of a N * N, by outer ring institute a little, take central point as symmetric points divide into groups in twos, can be divided into altogether 2N-2 group point, by the gray-scale value of every group of point and central point relatively.In this 2N-2 group point, if there is the gray-scale value of M group point all to be greater than the gray-scale value of central point, be the vein Edge Feature Points, all unique points that obtain finally form the vein network image, wherein, experiment draws: the vein that is three of window width N/extraction in a period of time as M is more complete.
What more should emphasize is that the direction tendency of vein is on not impact of the present invention.
The invention concrete steps
Step?1:
One of the image outermost obtained is punctuated and is set to zero point, eliminate the pseudo-special card point of a circle because border is former thereby produce.
Step?2:
Create the equal big or small storage matrix Z of image array that is set to 0 with the previous step outer ring, and each element value is initially to 0.
Step?3:
On the vein gray level image gathered, correspondence is made the window of a N * N, by outer ring institute a little, take central point as symmetric points divide into groups in twos, can be divided into altogether 2N-2 group point, by the gray-scale value of every group of point and central point relatively.
Step?4:
In this 2N-2 group point, if there is the gray-scale value of M group point all to be greater than the gray-scale value of central point, central point is the vein Edge Feature Points, and all unique points that so obtain finally form the vein network image.
When the width of rectangular window is suitable with the blood vessel width, effect is better, and noise is introduced in less meeting.
The vein network image extracted is done to skeletonizing, obtain single pixel network characterization; Finally the single pixel network characterization obtained is done to two-dimensional dct transform, the feature of the low frequency part of extracting two-dimensional dct transform after as last conversion.
Further, the authentication contrast stage, respectively the finger-type feature of two finger vein patterns of the DCT low frequency component of the finger-type feature of two finger vein patterns of the DCT low frequency component of extraction and finger-type transverse width sequence and storage and finger-type transverse width sequence is carried out to Euclidean distance and compare, the level of making decision merges.Only, when both all meet the demands, authentication is passed through.
Beneficial effect of the present invention:
1) in order to be convenient to use finger vein features, overcome its defect simultaneously, the present invention has introduced and the have living space finger contours feature of relation of finger vena, collector once can gather the feature of forefinger and middle finger simultaneously, two fingers are placed with V-type, adopt two fingers to gather image, not only can accurately determine comparison domain, can reduce selection and the distortion of finger, can bring great convenience for the user simultaneously; The authentication method that while is less with respect to the azygos vein contrast characteristic, the fusion of finger vena and finger-type can increase characteristic information again, the elevator system performance.
2) region of interest ROI can be determined in the point of crossing of forefinger and middle finger, take two finger point of crossing as initial point introducing coordinate system, by finger contours, carrys out accurate adjustment, has not only improved positioning precision, is user-friendly to simultaneously, and bicharacteristic merges and effectively improved system performance.
3) the present invention can extract respectively vein and two kinds of information of finger-type from piece image, and it is convenient to extract.
The accompanying drawing explanation
Fig. 1 is the image that the present invention gathers.
Fig. 2 is the schematic diagram that the present invention determines comparison domain.
Fig. 3 is the vein network characteristic pattern that the present invention tentatively extracts.
The profile (a) that Fig. 4 is the last intercepting of the present invention, the comparison domain of vein (b).
Fig. 5 is N * N matrix diagram, and the intermediate point pixel value is X (j, k).
Embodiment
Below in conjunction with accompanying drawing, the present invention is further described.
Embodiment:
With reference to Fig. 1, Fig. 2, the present invention relates to a kind of two finger finger-types and combine and carry out the method for biological identification with vein, contain following steps:
In the image that the above this method is processed, forefinger 1 and middle finger 2 must V-type be put, as shown in Figure 1.Only in this way can extract the middle point of crossing of two fingers, and take this point of crossing 3 as initial point introducing coordinate system, determine area-of-interest R0I 4.
The image that the above this method can gather from single width, extract respectively finger-type profile and vein network.As shown in Figure 2, the vein network as shown in Figure 3 for the finger-type profile extracted.
Wanted advanced column hisgram conversion process before extracting the finger-type profile, concrete steps are as follows:
Concrete steps are as follows:
(1) Probability p (x) that each gray level of computed image occurs.
(2)p(x)=p(x)×(x/125);
(3) 0~255 calculate the probability sum lower than each gray-scale value appearance of this gray-scale value successively.This sequence is q (x), x ∈ [0,255]
(4) histogram transformation.
k(m,n)=q(x)×255,
K (m, n) is that on image, horizontal ordinate is m, and ordinate is the gray-scale value that n is ordered, and x is the front gray-scale value of pixel value conversion for this reason.
Before extracting the vein network, make 16 * 16 medium filtering on the basis of Fig. 1.Then extract the vein network with vein network extraction algorithm, concrete steps are as follows:
Step?1:
One of the image outermost as shown in Figure 1 obtained is punctuated and is set to zero point, eliminate the pseudo-special card point of a circle because border is former thereby produce.
Step?2:
Create the equal big or small storage matrix Z of image array that is set to 0 with the previous step outer ring, and each element value is initially to 0.
Step?3:
Fig. 5 is the window that pointwise is slided on image, size is 17 * 17 (now N=17), at first set count value counter=0, the point that two sequence numbers in comparison diagram 5 rectangular box are O, if P (k)>P (X)+threshold and P (k)>P (X)+threshold, counter is from adding 1.The point that relatively sequence number is 2 again, by that analogy, until 31.
Wherein:
Threshold is threshold value.
Gray-scale value corresponding to point that P (k) is O for sequence number in Fig. 5,0≤k≤31.
Gray-scale value corresponding to central point that P (X) is X for sequence number in Fig. 5.
Step?4:
Judge that whether counter>M is true (now M=5), if true, be set to 255 by X, otherwise be 0.
When the width of rectangular window is suitable with the blood vessel width, effect is better, and noise is introduced in less meeting, and adding compare threshold threshold is also for the Background suppression noise.The vein network characterization tentatively extracted as shown in Figure 3.
The above this method is after the finger contours and vein network image that obtain as shown in Figure 2 and Figure 3, and take in Fig. 2 the two finger point of crossing 3 that obtain is initial point, sets up coordinate system, intercepting region of interest ROI 4.The region of interest ROI size of intercepting is 170 * 256, and unit is pixel.
The above this method is done skeletonizing to the vein network image extracted and is processed, and obtains single pixel network characterization, as Fig. 4 (a) image.Finally the single pixel network characterization obtained is done to two-dimensional dct transform, the vein pattern of the low frequency part of extracting two-dimensional dct transform after as last conversion.Extract the width sequence of Fig. 4 (b) finger-type region of interest ROI as the finger-type feature, the finger-type width sequence obtained is one-dimension array.
The above this method, in last fusion authentication, is taked the fusion certification policy of decision level; Compare respectively the Euclidean distance of vein pattern, the Euclidean distance of finger-type width sequence.When the Euclidean distance of vein pattern is less than 100, and the Euclidean distance of finger-type width sequence is less than at 600 o'clock, and authentication is passed through, otherwise refusal.
Obviously, the above embodiment of the present invention is only for example of the present invention clearly is described, and is not the restriction to embodiments of the present invention.For those of ordinary skill in the field, can also make other changes in different forms on the basis of the above description.Here can't give all embodiments exhaustive.Every still row in protection scope of the present invention of apparent variation that technical scheme of the present invention extends out or change that belong to.
Claims (8)
1. one kind is utilized two methods that refer to that vein and finger-type feature are carried out authentication, it is characterized in that, at first registration storage user couple refers to the vein pattern of interesting image regions ROI and the information of finger-type feature, then gather the two vein pattern of interesting image regions ROI and information of finger-type feature of referring to of authentication client, the finger-type feature of the finger-type feature of the vein pattern of the vein pattern of collection and storage and collection and storage is compared respectively, and the level of making decision merges;
When storage, collection, adopt following method to extract vein pattern: on the vein gray level image gathered, corresponding region of interest ROI is done N * N matrix diagram, the central point of a little take in outer ring, as symmetric points divide into groups in twos, is divided into as 2N-2 group point, every group of point and the gray-scale value of central point are compared; In 2N-2 group point, if there is the gray-scale value of M group point all to be greater than the gray-scale value of central point, window center point is the vein pattern point, and all unique points that obtain form the vein network image; When M extracts vein three of window width N/a period of time; Vein network image to gained is done skeletonizing, obtains single pixel network characterization, then does two-dimensional dct transform, extracts the feature of the low frequency part of two-dimensional dct transform as the vein network.
2. a kind of two methods that refer to that vein and finger-type feature are carried out authentication of utilizing according to claim 1, is characterized in that, when storage, collection, two images that refer to that the described V-arrangement collected is placed carry out pre-service.
3. a kind of two methods that refer to that vein and finger-type feature are carried out authentication of utilizing according to claim 2, is characterized in that, described collect two are referred to image adopts medium filtering and histogram transformation to carry out pre-service.
4. according to a kind of described two methods that refer to that vein and finger-type feature are carried out authentication of utilizing of claim 1 or 3, it is characterized in that, in described two finger theed collect, in image, extract respectively vein pattern and two kinds of information of finger-type feature, comprise the steps:
1) extract finger contours with Boundary extracting algorithm;
2) take extreme point between two fingers sets up rectangular coordinate system as two ,Yi Gai point of crossing, finger point of crossing as true origin, determines the region of interest ROI of vein image and contour images;
3) in contour images, finger left hand edge fitting a straight line, the fitting a straight line that obtains of take is rotated the translation image to fixed position as benchmark; Vein image follows that contour images is done same translational movement and the anglec of rotation is adjusted;
4) in the image after above-mentioned adjustment, extract the vein pattern of DCT low frequency from the vein image region of interest ROI, extract the finger-type feature of finger-type transverse width sequence from profile interesting image regions ROI.
5. a kind of two methods that refer to that vein and finger-type feature are carried out authentication of utilizing according to claim 4, is characterized in that step 1) middle with Canny algorithm extraction edge.
6. a kind of two methods that refer to that veins and finger-type feature are carried out authentication of utilizing according to claim 4, it is characterized in that, the translation of vein step 3), contour images and rotation are adjusted: the left profile of two outline lines of finger edge of take is benchmark, with straight line, carries out matching; By the image rotation certain angle, the fitting a straight line of gained is rotated to the vertical direction parallel with the coordinate Y-axis again, and move to the position with the left hand edge fixed range of region of interest ROI.
7. a kind of two methods that refer to that vein and finger-type feature are carried out authentication of utilizing according to claim 4, is characterized in that, described vein pattern is the DCT low frequency component; Described finger-type is characterized as finger-type transverse width sequence.
8. a kind of two methods that refer to that vein and finger-type feature are carried out authentication of utilizing according to claim 7, is characterized in that, it contains following steps:
1) two information that refer to vein and finger-type feature of registration storage user,
(1) the hand shape that adjacent two fingers that gather with the near infrared collector are separated and the image of finger vena, do the pre-service of medium filtering and histogram transformation;
(2), to the image after pre-service, with the Canny algorithm, extract edge; Then the extreme point between two fingers of take is set up rectangular coordinate system as two ,Yi point of crossing, finger point of crossing coordinates as initial point, determines region of interest ROI;
(3) after the region of interest ROI that obtains vein image and contour images, in contour images, point left hand edge with fitting a straight line, the fitting a straight line that obtains of take makes fitting a straight line rotate to the vertical direction parallel with the coordinate Y-axis as benchmark rotation translation image, and moves to the position with region of interest ROI left hand edge fixed range; Vein image is done same translation and rotation with contour images;
(4), in the vein image region of interest ROI and contour images region of interest ROI from adjusting, extract respectively the vein pattern of low frequency component of two-dimensional dct transform and the finger-type feature of finger-type transverse width sequence component;
(5) by above-mentioned two refer to vein pattern and finger-type feature registration storages;
2) according to 1) in the step of (1)-(4), gather, extract two information that refer to veins and finger-type feature of authentication client, extract respectively low frequency component and the finger-type transverse width sequence of two-dimensional dct transform;
3) by 2) low frequency component and 1 of the two-dimensional dct transform that extracts) Euclidean distance of the DCT low frequency component stored compares, and does decision level fusion; By 2) the finger-type transverse width sequence and 1 extracted) Euclidean distance of the finger-type transverse width sequence of storing does decision level fusion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010105779061A CN102043961B (en) | 2010-12-02 | 2010-12-02 | Vein feature extraction method and method for carrying out identity authentication by utilizing double finger veins and finger-shape features |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010105779061A CN102043961B (en) | 2010-12-02 | 2010-12-02 | Vein feature extraction method and method for carrying out identity authentication by utilizing double finger veins and finger-shape features |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102043961A CN102043961A (en) | 2011-05-04 |
CN102043961B true CN102043961B (en) | 2013-12-11 |
Family
ID=43910087
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2010105779061A Expired - Fee Related CN102043961B (en) | 2010-12-02 | 2010-12-02 | Vein feature extraction method and method for carrying out identity authentication by utilizing double finger veins and finger-shape features |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102043961B (en) |
Families Citing this family (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102194111B (en) * | 2011-06-21 | 2012-12-19 | 哈尔滨工程大学 | Non-contact self-adapting extraction method for ROI (Region of Interest) of fist-clenching-type hand back vein sample |
CN102663393B (en) * | 2012-03-02 | 2014-10-22 | 哈尔滨工程大学 | Method for extracting region of interest of finger vein image based on correction of rotation |
CN106169062B (en) * | 2016-06-08 | 2019-05-21 | 杭州电子科技大学 | A kind of implementation method referring to vein fusion fuzzy vault |
CN109376708B (en) * | 2016-06-08 | 2021-11-26 | 大连民族大学 | Method for extracting ROI |
CN108319887A (en) * | 2017-01-18 | 2018-07-24 | 中国移动通信有限公司研究院 | A kind of identity authentication method and system |
CN107368792A (en) * | 2017-06-29 | 2017-11-21 | 广州智慧城市发展研究院 | A kind of finger vein identification method and system based on wave filter and Bone Edge |
CN108921851B (en) * | 2018-06-06 | 2021-07-09 | 深圳市未来媒体技术研究院 | Medical CT image segmentation method based on 3D countermeasure network |
CN108960120B (en) * | 2018-06-28 | 2023-08-08 | 维沃移动通信有限公司 | Fingerprint identification processing method and electronic equipment |
CN109145792A (en) * | 2018-08-09 | 2019-01-04 | 钧安科技(深圳)有限公司 | Two fingers setting refers to vein identification device and method |
CN109190517B (en) * | 2018-08-14 | 2022-05-10 | 北京凌云天润智能科技有限公司 | Finger vein image edge extraction and direction correction method |
CN109409249A (en) * | 2018-09-30 | 2019-03-01 | 联想(北京)有限公司 | Information processing method and electronic equipment |
CN109598235B (en) * | 2018-12-04 | 2022-05-17 | 云天瀚科技发展有限公司 | Finger vein image authentication method and device |
CN109697347B (en) * | 2018-12-04 | 2020-07-14 | 云天瀚科技发展有限公司 | User characteristic authentication method and device based on finger veins and finger-shaped characteristics |
CN111209851B (en) * | 2020-01-04 | 2023-06-20 | 圣点世纪科技股份有限公司 | Finger vein recognition method based on deep fusion of finger abdominal vein and finger dorsal vein |
CN113780122B (en) * | 2021-08-30 | 2023-12-05 | 沈阳大学 | Palm vein feature encryption-based recognition template generation method and device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101004789A (en) * | 2006-01-17 | 2007-07-25 | 郭岳衡 | Device for recognizing image of finger vein |
CN101477622A (en) * | 2009-01-15 | 2009-07-08 | 山东大学 | Finger vein recognition system |
CN101582115A (en) * | 2008-05-13 | 2009-11-18 | 索尼株式会社 | Authentication apparatus, authentication method, registration apparatus and registration method |
-
2010
- 2010-12-02 CN CN2010105779061A patent/CN102043961B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101004789A (en) * | 2006-01-17 | 2007-07-25 | 郭岳衡 | Device for recognizing image of finger vein |
CN101582115A (en) * | 2008-05-13 | 2009-11-18 | 索尼株式会社 | Authentication apparatus, authentication method, registration apparatus and registration method |
CN101477622A (en) * | 2009-01-15 | 2009-07-08 | 山东大学 | Finger vein recognition system |
Also Published As
Publication number | Publication date |
---|---|
CN102043961A (en) | 2011-05-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102043961B (en) | Vein feature extraction method and method for carrying out identity authentication by utilizing double finger veins and finger-shape features | |
CN104239769B (en) | Personal identification method and system based on finger vein features | |
US10762366B2 (en) | Finger vein identification method and device | |
CN103870808B (en) | Finger vein identification method | |
Chen et al. | Iris recognition based on bidimensional empirical mode decomposition and fractal dimension | |
KR20090087895A (en) | Method and apparatus for extraction and matching of biometric detail | |
CN104091145B (en) | Human body slaps arteries and veins characteristic image acquisition method | |
CN105760841B (en) | Identity recognition method and system | |
CN107169479A (en) | Intelligent mobile equipment sensitive data means of defence based on fingerprint authentication | |
Yang et al. | Finger-vein segmentation based on multi-channel even-symmetric Gabor filters | |
CN107729820A (en) | A kind of finger vein identification method based on multiple dimensioned HOG | |
CN106096569A (en) | A kind of finger vein identification method | |
CN104123547B (en) | Based on improved trend pass filtering and flexible matched recognition methods | |
Kang et al. | The biometric recognition on contactless multi-spectrum finger images | |
Li | An Iris recognition algorithm based on coarse and fine location | |
AU2020432845A1 (en) | An integrated method for finger vein recognition and anti-spoofing, and device, storage medium and equipment therefor | |
CN115223211B (en) | Identification method for converting vein image into fingerprint image | |
Biswas et al. | A new approach of iris detection and recognition | |
CN110147769B (en) | Finger vein image matching method | |
Sun et al. | A DSP-based finger vein authentication system | |
CN106295622A (en) | A kind of recognition methods based on hand back vein | |
Zhichao et al. | Two modality-based bi-finger vein verification system | |
Benziane et al. | Dorsal hand vein identification based on binary particle swarm optimization | |
CN105373781A (en) | Binary image processing method for identity authentication | |
CN105488460A (en) | Physiological feature based image processing method |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131211 Termination date: 20141202 |
|
EXPY | Termination of patent right or utility model |