CN113609943A - Finger vein recognition method, electronic device and storage medium - Google Patents
Finger vein recognition method, electronic device and storage medium Download PDFInfo
- Publication number
- CN113609943A CN113609943A CN202110850771.XA CN202110850771A CN113609943A CN 113609943 A CN113609943 A CN 113609943A CN 202110850771 A CN202110850771 A CN 202110850771A CN 113609943 A CN113609943 A CN 113609943A
- Authority
- CN
- China
- Prior art keywords
- finger vein
- registered finger
- registered
- image
- images
- 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
- 210000003462 vein Anatomy 0.000 title claims abstract description 442
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000001514 detection method Methods 0.000 claims abstract description 21
- 238000012795 verification Methods 0.000 claims abstract description 4
- 230000004044 response Effects 0.000 claims description 131
- 230000004927 fusion Effects 0.000 claims description 81
- 239000000126 substance Substances 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 19
- 238000009826 distribution Methods 0.000 description 17
- 238000012545 processing Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 102000001554 Hemoglobins Human genes 0.000 description 1
- 108010054147 Hemoglobins Proteins 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000010420 art technique Methods 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a finger vein identification method, an electronic device and a storage medium. The finger vein recognition method comprises the following steps: responding to a finger vein registration request, and acquiring a binary registered finger vein image of a plurality of acquisition angles; fusing a plurality of registered finger vein images to obtain a registered finger vein template; responding to a finger vein detection request, and acquiring a binary finger vein image to be compared; judging the consistency of the finger vein image to be compared and the registered finger vein template; and performing identity verification on the finger vein image to be compared based on the consistency of the finger vein image to be compared and the registered finger vein template. The invention provides a novel finger vein finger image template generation method and a finger vein recognition method, and the comparison efficiency is improved because only one template needs to be compared. Meanwhile, the template covers all the acquisition angles, so that the comparison accuracy, efficiency and stability are improved.
Description
Technical Field
The invention relates to the technical field related to biological identification, in particular to a finger vein identification method, electronic equipment and a storage medium.
Background
The finger vein is a second generation biological identification technology, and utilizes the absorption effect of hemoglobin in flowing blood in the finger on near infrared light with the wavelength of 700-1000nm (muscles and bones are easy to penetrate), obtains a vein blood vessel distribution image in the finger through the cooperation of an infrared light group and a camera, converts the image into digital characteristics through an algorithm, and compares the vein characteristic values, thereby carrying out identity identification.
Currently, when a user registers to acquire a vein image with the same finger, N templates (e.g., 3 templates) need to be acquired for subsequent comparison.
As shown in fig. 1, the existing acquisition method includes:
step S101, collecting finger vein registration images;
step S102, judging the image quality;
step S103, preprocessing an image;
step S104, generating a standard binary image of the finger vein;
step S105, generating a user template I, a user template II and a user template III;
step S106, carrying out finger vein identification operation on the user;
step S107, image acquisition and digital image processing are carried out;
step S108, comparing the collected image with a template I to a template III respectively;
and step S109, if the identity judgment condition is met, the identity is legal, if the identity is not legal, the identity is judged to be illegal, the authentication is carried out again, and the step S106 is executed.
Specifically, the existing finger vein recognition technology is as follows:
(1) when a user registers, 3 images are collected for fingers as comparison templates (template generation process: quality judgment-preprocessing-digital processing-template images);
(2) the 3 templates are unrelated and exist independently;
(3) when the user identities are compared, the collected user images are respectively compared with 3 templates, and the similarity is judged according to the digital characteristics and a certain rule, so that the validity of the identities is confirmed.
(4) If the collection angle (for example > 15 ℃) of the finger relative to the light source and the camera is larger than the allowable comparison requirement when the user identities are compared, the difference between the collected image and the template is large, and therefore misjudgment is made.
Therefore, in the existing finger vein recognition technology, because the finger placing angle of the user cannot be consistent with the registered template, the difference between the collected vein image and the template is easy to be large, and thus an incorrect identification conclusion is made.
Disclosure of Invention
Accordingly, it is desirable to provide a finger vein recognition method, an electronic device, and a storage medium, which solve the technical problem that the finger vein recognition in the related art is prone to misjudgment.
The invention provides a finger vein identification method, which comprises the following steps:
responding to a finger vein registration request, and acquiring a binary registered finger vein image of a plurality of acquisition angles;
fusing a plurality of registered finger vein images to obtain a registered finger vein template;
responding to a finger vein detection request, and acquiring a binary finger vein image to be compared;
judging the consistency of the finger vein image to be compared and the registered finger vein template;
and performing identity verification on the finger vein image to be compared based on the consistency of the finger vein image to be compared and the registered finger vein template.
Further, the obtaining of the registered finger vein template after fusing the plurality of registered finger vein images specifically includes:
carrying out corner detection on each registered finger vein image to obtain corners in the registered finger vein images, wherein each corner has a corner position and a corner response value;
taking the corner points of which the corner point response values meet the corner point threshold condition as effective corner points, wherein the corner point response values of the effective corner points are effective corner point response values;
taking a region, in which the relation between effective corner response values in the two registered finger vein images meets a preset fusion condition, as a fusion region of the two registered finger vein images;
and fusing corresponding fusion areas in the plurality of registered finger vein images to obtain a registered finger vein template.
Further, the preset fusion condition is as follows: the difference value between the sum of the effective corner response values of each column of one registered finger vein image and the sum of the effective corner response values of the corresponding column of the other registered finger vein image in the area is smaller than a column sum difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each column of one registered finger vein image and the mean square error of the effective corner response value of the corresponding column of the other registered finger vein image in the area is smaller than a column mean square error difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each line of one registered finger vein image and the mean square error of the effective corner response value of the corresponding line of the other registered finger vein image in the region is smaller than a line mean square error difference threshold value.
Still further, the taking a region in which a relationship between effective corner response values in two registered finger vein images meets a preset fusion condition as a fusion region of the two registered finger vein images specifically includes:
calculating the sum of effective corner response values of each column for each registered finger vein image;
if an area exists, the difference value of the sum of the effective corner point response values of each column of the two registered finger vein images and the sum of the effective corner point response values of the corresponding column of the other registered finger vein image in the area is smaller than a column sum difference threshold value, and the two registered finger vein images are judged to have a suspected overlapping area;
respectively calculating the mean square deviation value of the effective corner response value of each column in the suspected overlapping area in the two registered finger vein images of the suspected overlapping area;
if the difference value of the mean square error of the effective corner point response value of each column of one registered finger vein image in the suspected overlapping area and the mean square error of the effective corner point response value of the corresponding column of the other registered finger vein image is smaller than a column mean square error difference value threshold, then respectively calculating the mean square error value of the effective corner point response value of each line in the suspected overlapping area in the two registered finger vein images in the suspected overlapping area;
if the difference value of the mean square error of the effective corner point response value of each line of one registered finger vein image in the suspected overlapping area and the mean square error of the effective corner point response value of the corresponding line of the other registered finger vein image is less than the line mean square error difference threshold value, the suspected overlapping area is judged to be the high suspected overlapping area of the two registered finger vein images;
and determining a fusion area of the two registered finger vein images based on the high suspected overlapping area.
Furthermore, the taking a region in which a relationship between effective corner response values in two registered finger vein images satisfies a preset fusion condition as a fusion region of the two registered finger vein images specifically includes:
taking a region, in which the relation between effective corner response values in the two registered finger vein images meets a preset fusion condition, as a highly suspected overlapping region of the two registered finger vein images;
calculating the cross-correlation coefficient of the highly suspected overlapping area of the two registered finger vein images;
and if the cross-correlation coefficient is larger than a preset cross-correlation coefficient threshold value, judging that the high suspected overlapping area is a fusion area of the two registered finger vein images.
Still further, the determining that the suspected overlapping area is a fusion area of the two registered finger vein images specifically further includes:
and if the two registered finger vein images comprise a plurality of high suspected overlapping areas with cross correlation coefficients larger than a preset cross correlation coefficient threshold value, selecting the high suspected overlapping area with the maximum cross correlation coefficient as a fusion area of the two registered finger vein images.
Still further, the acquiring registered finger vein images at a plurality of acquisition angles specifically includes:
acquiring registration finger vein images of a plurality of different acquisition angles, and performing binarization processing to obtain binarized registration finger vein images of the plurality of acquisition angles;
and rotationally adjusting the registered finger vein images of a plurality of different acquisition angles to the same angle.
Still further, the registered finger vein images of the plurality of different acquisition angles at least comprise a registered finger vein image with the maximum deflection of the hand pointing to a first direction and a registered finger vein image with the maximum deflection of the hand pointing to a second direction, and the first direction is opposite to the second direction.
Still further, the acquiring registered finger vein images at a plurality of acquisition angles specifically includes:
and detecting the acquisition angle of the registered finger vein image, and if the registered finger vein image with the maximum deflection of the hand pointing to the first direction is lacked or the registered finger vein image with the maximum deflection of the hand pointing to the second direction is lacked, executing a reminding operation.
Still further, the determining the consistency between the finger vein image to be compared and the registered finger vein template specifically includes:
and if the registered finger vein template comprises the finger vein image to be compared, judging that the finger vein image to be compared is consistent with the registered finger vein template.
The present invention provides an electronic device, including:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to at least one of the processors; wherein the content of the first and second substances,
the memory stores instructions executable by at least one of the processors to enable the at least one of the processors to perform the finger vein recognition method as previously described.
The present invention provides a storage medium storing computer instructions for performing all the steps of the finger vein recognition method as described above when the computer executes the computer instructions.
The invention provides a novel finger vein finger image template generation method and a finger vein recognition method, and the comparison efficiency is improved because only one template needs to be compared. Meanwhile, the template covers all the acquisition angles, so that the acquisition angles of the fingers of the user are within the coverage range of the final template no matter what acquisition angles the fingers of the user are in, and the comparison accuracy, efficiency and stability are improved.
Drawings
FIG. 1 is a flowchart illustrating a conventional finger vein recognition method;
FIG. 2 is a flowchart illustrating a method for identifying finger veins according to the present invention;
FIG. 3 is a schematic view of a finger vein image capture device;
FIG. 4 is a flowchart illustrating a method for finger vein recognition according to an embodiment of the present invention;
FIG. 5a is a diagram illustrating binarization of an image of a finger vein deflected to the left according to an example of the present invention;
FIG. 5b is a diagram illustrating an exemplary registration finger vein image binarization in an intermediate state according to the present invention;
FIG. 5c is a diagram illustrating a binarized representation of a rightward deflected registered finger vein image according to an example of the present invention;
FIG. 6a is a distribution diagram of the effective corner points of FIG. 5 a;
FIG. 6b is a diagram of the effective corner of FIG. 5 b;
FIG. 6c is a diagram of the effective corner points of FIG. 5 c;
fig. 7a is a schematic diagram of a fusion region of the distribution diagrams of effective corners of the first registered finger vein image and the second registered finger vein image;
fig. 7b is a schematic diagram of a fusion region of the distribution diagrams of the effective corner points of the second registered finger vein image and the third registered finger vein image;
FIG. 8 is a fused diagram of the distribution map of the effective corners of three registered finger vein images;
FIG. 9 is a user template after three registered finger vein images are fused;
FIG. 10 is a flowchart illustrating a method for identifying finger veins according to a preferred embodiment of the present invention;
FIG. 11 is a flowchart illustrating a method for determining an image fusion region according to a preferred embodiment of the present invention;
fig. 12 is a schematic diagram of a hardware structure of an electronic device according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples.
Example one
Fig. 2 is a flowchart illustrating a method for identifying finger veins according to the present invention, which includes:
step S201, responding to a finger vein registration request, and acquiring a binaryzation registration finger vein image of a plurality of acquisition angles;
step S202, fusing a plurality of registered finger vein images to obtain a registered finger vein template;
step S203, responding to the finger vein detection request, and acquiring a binary finger vein image to be compared;
step S204, judging the consistency of the finger vein image to be compared and the registered finger vein template;
step S205, based on the consistency between the finger vein image to be compared and the registered finger vein template, performing identity authentication on the finger vein image to be compared.
Specifically, the present invention can be applied to an Electronic Control Unit (ECU) of a vehicle.
Acquiring finger vein images of a plurality of collection angles as registered finger vein images by executing step S201, and performing binarization processing on the registered finger vein images to obtain binarized registered finger vein images to finish user registration. For example, with a finger vein image capture device as shown in fig. 3, a finger 31 is illuminated by a near infrared light source 32, and then a finger vein image of the finger is captured by an image sensor (not shown in the figure) at the bottom of the finger 31. Then, the existing binarization processing is adopted, namely, the gray level image is converted into a binary image. The pixel gray scale larger than a certain critical gray scale value is set as a gray scale maximum value, and the pixel gray scale smaller than the value is set as a gray scale minimum value, so that binarization is realized. Each registered finger vein image is a binary image with the same width and height, and a row-column coordinate system of each image is established.
Then, step S202 is executed to fuse the plurality of registered finger vein images to obtain a registered finger vein template. The registered finger vein template is obtained by fusing registered finger vein images from a plurality of different acquisition angles. Therefore, the registered finger vein template covers all the acquisition angles, so that the finger of the user is in the coverage range of the final template no matter what acquisition angle the finger of the user is in, and the comparison accuracy, efficiency and stability are improved.
Then, when the user needs to identify, a finger vein detection request is generated, for example, by contacting a sensor of the finger vein image acquisition device, and the finger vein detection request is generated, the step S203 is triggered to be executed, the finger vein image to be compared is acquired, and the binarization processing is performed to obtain the binarized finger vein image to be compared. For example, when a user places a finger on the finger vein image capture device shown in fig. 3, the finger vein image of the user can be acquired, and the finger vein image is binarized to be used as a finger vein image to be compared.
And step S204 is executed, and whether the finger vein image to be compared is consistent with the registered finger vein template is judged. Then, in step S205, the finger vein image to be compared may be authenticated based on the consistency between the finger vein image to be compared and the registered finger vein template.
For example, if only one registered finger vein template is included, when the finger vein image to be compared is consistent with the registered finger vein template, the user authentication is judged to be passed, otherwise, the user authentication is judged to be failed.
For another example, if multiple registered finger vein templates are included, each registered finger vein template may be associated with one user. And if the finger vein image to be compared is consistent with one of the registered finger vein templates, judging that the user corresponding to the finger vein image to be compared is the user corresponding to the registered finger vein template consistent with the finger vein image to be compared. And if the finger vein image to be compared is inconsistent with all the registered finger vein templates, judging that the user authentication fails.
The invention provides a novel finger vein finger image template generation method and a finger vein recognition method, and the comparison efficiency is improved because only one template needs to be compared. Meanwhile, the template covers all the acquisition angles, so that the acquisition angles of the fingers of the user are within the coverage range of the final template no matter what acquisition angles the fingers of the user are in, and the comparison accuracy, efficiency and stability are improved.
Example two
Fig. 4 is a flowchart illustrating a method for identifying finger veins according to an embodiment of the present invention, including:
step S401, responding to a finger vein registration request, and acquiring registered finger vein images of a plurality of different acquisition angles;
in one embodiment, the registered finger vein images of the plurality of different acquisition angles at least comprise a registered finger vein image with the maximum deflection of the hand pointing to a first direction and a registered finger vein image with the maximum deflection of the hand pointing to a second direction, and the first direction is opposite to the second direction.
In one embodiment, the acquiring registered finger vein images at a plurality of acquisition angles specifically further includes:
and detecting the acquisition angle of the registered finger vein image, and if the registered finger vein image with the maximum deflection of the hand pointing to the first direction is lacked or the registered finger vein image with the maximum deflection of the hand pointing to the second direction is lacked, executing a reminding operation.
Step S402, the registered finger vein images of a plurality of different collection angles are adjusted to the same angle in a rotating mode, and each registered finger vein template generates a binary image with the same size;
step S403, performing corner detection on each registered finger vein image to obtain corners in the registered finger vein image, wherein each corner has a corner position and a corner response value;
step S404, taking the corner points of which the corner point response values meet the corner point threshold condition as effective corner points, wherein the corner point response values of the effective corner points are effective corner point response values;
step S405, taking a region, in which the relation between effective corner response values in two registered finger vein images meets a preset fusion condition, as a fusion region of the two registered finger vein images, wherein the preset fusion condition is as follows:
the difference value between the sum of the effective corner response values of each column of one registered finger vein image and the sum of the effective corner response values of the corresponding column of the other registered finger vein image in the area is smaller than a column sum difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each column of one registered finger vein image and the mean square error of the effective corner response value of the corresponding column of the other registered finger vein image in the area is smaller than a column mean square error difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each line of one registered finger vein image and the mean square error of the effective corner response value of the corresponding line of the other registered finger vein image in the region is smaller than a line mean square error difference threshold value.
In one embodiment, the taking a region in which a relationship between effective corner response values in two registered finger vein images meets a preset fusion condition as a fusion region of the two registered finger vein images specifically includes:
calculating the sum of effective corner response values of each column for each registered finger vein image;
if an area exists, the difference value of the sum of the effective corner point response values of each column of the two registered finger vein images and the sum of the effective corner point response values of the corresponding column of the other registered finger vein image in the area is smaller than a column sum difference threshold value, and the two registered finger vein images are judged to have a suspected overlapping area;
respectively calculating the mean square deviation value of the effective corner response value of each column in the suspected overlapping area in the two registered finger vein images of the suspected overlapping area;
if the difference value of the mean square error of the effective corner point response value of each column of one registered finger vein image in the suspected overlapping area and the mean square error of the effective corner point response value of the corresponding column of the other registered finger vein image is less than a column mean square error difference value threshold, calculating the mean square error value of the effective corner point response value of each line in the suspected overlapping area in the two registered finger vein images in the suspected overlapping area respectively;
if the difference value of the mean square error of the effective corner point response value of each line of one registered finger vein image in the suspected overlapping area and the mean square error of the effective corner point response value of the corresponding line of the other registered finger vein image is less than the line mean square error difference threshold value, the suspected overlapping area is judged to be the high suspected overlapping area of the two registered finger vein images;
and determining a fusion area of the two registered finger vein images based on the high suspected overlapping area.
In one embodiment, the high suspected overlapping area is directly used as a fusion area of the two registered finger vein images.
In one embodiment, the taking a region in which a relationship between effective corner response values in two registered finger vein images meets a preset fusion condition as a fusion region of the two registered finger vein images specifically includes:
taking a region, in which the relation between effective corner response values in the two registered finger vein images meets a preset fusion condition, as a highly suspected overlapping region of the two registered finger vein images;
calculating the cross-correlation coefficient of the highly suspected overlapping area of the two registered finger vein images;
and if the cross correlation coefficient is larger than a preset cross correlation coefficient threshold value, judging that the suspected overlapping area is a fusion area of the two registered finger vein images.
In one embodiment, the determining that the suspected overlapping area is a fusion area of the two registered finger vein images specifically further includes:
and if the two registered finger vein images comprise a plurality of high suspected overlapping areas with cross correlation coefficients larger than a preset cross correlation coefficient threshold value, selecting the high suspected overlapping area with the maximum cross correlation coefficient as a fusion area of the two registered finger vein images.
Step S406, fusing corresponding fusion areas in the plurality of registered finger vein images to obtain registered finger vein templates;
step S407, responding to the finger vein detection request, and acquiring a finger vein image to be compared;
step S408, if the registered finger vein template comprises the finger vein image to be compared, judging that the finger vein image to be compared is consistent with the registered finger vein template;
and step S409, performing identity verification on the finger vein image to be compared based on the consistency of the finger vein image to be compared and the registered finger vein template.
Specifically, by executing step S401 to acquire finger vein images of a plurality of collection angles as registered finger vein images, user registration is completed. For example, with a finger vein image capture device as shown in fig. 3, a finger 31 is illuminated by a near infrared light source 32, and then a finger vein image of the finger is captured by an image sensor (not shown in the figure) at the bottom of the finger 31.
During collection, the registered finger vein images of a plurality of different collection angles are collected. The method at least comprises a registered finger vein image with the maximum deflection of the hand towards the first direction and a registered finger vein image with the maximum deflection of the hand towards the second direction, so that all acquisition angles are covered to the maximum extent. The first direction is opposite to the second direction, and a reference direction may be specifically set, and the first direction and the second direction are symmetrical with respect to the reference direction. For example, two near-infrared light sources 32 are set to be parallel, an image sensor is provided at the bottom of the middle region between the two near-infrared light sources 32, and the extending direction of the image sensor is set as a reference direction 33. The first direction deflection is a left deflection in the reference direction 33 and the second direction deflection is a right deflection in the reference direction 33, or the first direction deflection is a right deflection in the reference direction 33 and the second direction deflection is a left deflection in the reference direction 33.
The maximum deflection may be determined by calibration so that most users cannot exceed the maximum deflection when in use.
Meanwhile, if the collected registered finger vein images are detected, and registered finger vein images with the most deflection of the hand pointing to the first direction or registered finger vein images with the most deflection of the hand pointing to the second direction are lacked, a reminding operation is performed, for example, a user is prompted in a sound-light alarm mode. For example, in the detection regions of the two near-infrared light sources 32, a light and shadow effect of the maximum deflection in the first direction and a light and shadow effect of the maximum deflection in the second direction are displayed by the light and shadow effect.
Then step S402 rotationally adjusts the registered finger vein images of a plurality of different acquisition angles to the same angle, for example, all to the reference direction 33 as shown in fig. 3. The rotation adjustment mode can adopt the existing rotation adjustment mode. During adjustment, adjustment can be performed based on the detected included angle between the finger reference line and the reference direction, so that the finger reference lines of all the registered finger vein images coincide with the reference direction, and then the registered finger vein images at different acquisition angles are adjusted to the same angle in a rotating manner. The selection of the finger reference line may be determined in an existing manner.
Then, step S403 performs corner detection on each registered finger vein image to obtain corners in the registered finger vein image, where each corner has a corner position and a corner response value. The corner detection and its response values can be obtained using prior art techniques. For example using Harris corner detection algorithm. And calculating the relation between the fusion area and the position of the plurality of registered finger vein images by comparing the angular point response values.
And the Harris corner response algorithm utilizes a local window to move on the image to judge the position of large change of the gray scale. The gray scale change E (u, v) resulting from the image window translation u, v is calculated.
Wherein:
[ u, v ] is the offset of the window W;
(x, y) is the pixel coordinate position corresponding to window W;
i (x, y) is the image gray scale value at pixel coordinate position (x, y);
i (x + u, y + v) is the image gray scale value at pixel coordinate position (x + u, y + v);
w (x, y) is a window function.
Finally, a corner response value R is calculated as the corner response value.
R＝det(M)-k(trace(M))2
det(M)＝λ1λ2
trace(M)＝λ1+λ2
Wherein λ is1,λ2Is the eigenvalue of the matrix M in both the x and y gray scale change directions, and k is a prescribed coefficient.
Then, step S404 is executed to use the corner points meeting the corner point threshold condition as valid corner points. The corner threshold condition is determined according to the selected corner response algorithm. For example, for Harris corner response algorithm, the corner threshold condition is that the corner response value R is greater than the preset detection value threshold. And (3) removing the characteristic points with the corner response value lower than the N value to generate a standard response value R (x, y) distribution image of each template. The point with the response value lower than the N value is removed, so that the interference of the low-gradient corner point (possibly caused by interference factors such as the outside world) can be effectively removed, and the accuracy and the response speed of the algorithm can be improved. And finally obtaining the effective corner point and the position coordinates of each registered finger vein image.
Specifically, a distribution graph of the corresponding effective corner can be generated for each registered finger vein image, and the distribution graph of the effective corner marks the corner identification and the corresponding effective corner response value at the pixel point corresponding to the effective corner in the registered finger vein image.
As shown in fig. 5a, 5b, and 5c, three registered finger vein images are shown. As shown in fig. 6a, 6b, and 6c, the distributions of the effective corners corresponding to fig. 5a, 5b, and 5c calculated by using the Harris algorithm are shown, and when the Harris algorithm is used, the distribution of the effective corners may be the distribution of the effective corners. The dots in the figure are corner marks, and each corner mark records a corresponding effective corner response value and a position coordinate thereof.
Then, step S405 is executed to use a region in which the relationship between the effective corner response values in the two registered finger vein images satisfies a preset fusion condition as a fusion region of the two registered finger vein images.
In particular, a map of the active corners may be used to determine the blending region. The distribution map of the effective corner points visually shows the positions of the corner points on the registered finger vein image. However, since the detection of the fusion region is based on the effective corner response value for matching, the pixel points corresponding to the effective corners may also be directly recorded in the registered finger vein image, and then the fusion region may be determined.
The fusion conditions of the fusion region were:
the difference value between the sum of the effective corner response values of each column of one registered finger vein image and the sum of the effective corner response values of the corresponding column of the other registered finger vein image in the area is smaller than a column sum difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each column of one registered finger vein image and the mean square error of the effective corner response value of the corresponding column of the other registered finger vein image in the area is smaller than a column mean square error difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each line of one registered finger vein image and the mean square error of the effective corner response value of the corresponding line of the other registered finger vein image in the region is smaller than a line mean square error difference threshold value.
During calculation, the sum of the effective corner response values of each column is calculated by taking the columns as a unit, then the two registered finger vein images are compared column by column, and an area, in which the difference of the sum of the effective corner response values of the columns between the two registered finger vein images is smaller than a column sum difference threshold value, is searched to serve as a suspected overlapping area. Wherein, for two registered finger vein images, when the difference between the sum of the effective corner point response values of the a-th column of the first registered finger vein image and the sum of the effective corner point response values of the b-th column of the second registered finger vein image is found to be less than a column sum difference threshold, whether the difference between the sum of the effective corner point response values of the a + 1-th column of the first registered finger vein image and the sum of the effective corner point response values of the b + 1-th column of the second registered finger vein image is less than a column sum difference threshold is continuously judged until the difference between the sum of the effective corner point response values of the a + c-th column of the first registered finger vein image and the sum of the effective corner point response values of the b + c-th column of the second registered finger vein image is greater than or equal to the column sum difference threshold, the region from the a-th column to the a + c-th column of the first registered finger vein image and the region from the b-th column to the b + c-th column of the second registered finger vein image are respectively obtained from the column sum threshold and the column sum difference threshold, as a pseudo-overlap area present in the first registered finger vein image and the second registered finger vein image.
Then, the areas of the a-th column to the a + c-th column of the first registered finger vein image and the areas of the b-th column to the b + c-th column of the second registered finger vein image are calculated, and the mean square deviation value of the effective corner response value of each column is calculated. And comparing whether the difference value of the mean square error of the effective corner response value of the a-th column of the first registered finger vein image and the mean square error of the effective corner response value of the b-th column of the second registered finger vein image is less than a column mean square error difference threshold value. If the difference value of the mean square error of the effective corner response value of the a-th column of the first registered finger vein image and the mean square error of the effective corner response value of the b-th column of the second registered finger vein image is smaller than the column mean square error difference threshold value, whether the difference value of the mean square error of the effective corner response value of the a + 1-th column of the first registered finger vein image and the mean square error of the effective corner response value of the b + 1-th column of the second registered finger vein image is smaller than the column mean square error difference threshold value or not is continuously judged until the mean square error of the effective corner response value of the a + c-th column of the first registered finger vein image and the mean square error of the effective corner response value of the b + c-th column of the second registered finger vein image are compared. And if the mean square error of the effective corner response value of each column from the a-th column to the a + c-th column of the first registered finger vein image and the mean square error of the effective corner response value of each column from the b-th column to the b + c-th column of the second registered finger vein image are both smaller than a column mean square error difference threshold value, calculating the mean square error of the effective corner response value of each row in the suspected overlapping area in the two registered finger vein images in the suspected overlapping area.
If the difference value between the mean square error of the effective corner point response value of one column of the registered finger vein image and the mean square error of the effective corner point response value of the corresponding column of the other registered finger vein image is larger than or equal to the column mean square error difference threshold value, the suspected overlapping area is recalculated.
For example, if the difference between the mean square error of the effective corner response value of the P-th column of the first registered finger vein image and the mean square error of the effective corner response value of the Q-th column corresponding to the second registered finger vein image in the suspected overlapping area is greater than or equal to the column mean square error difference threshold, the suspected overlapping area is reduced to the P-1 th column of the first registered finger vein image, and the suspected overlapping area is reduced to the Q-1 th column of the second registered finger vein image.
And if the difference value between the mean square error of the effective corner response values of the first registered finger vein image in each line of the suspected overlapping area and the mean square error of the effective corner response values of the second registered finger vein image in each line of the suspected overlapping area is smaller than a line mean square error difference threshold value, judging that the suspected overlapping area is the fusion area of the two registered finger vein images.
For example, in the suspected overlapping areas of the first registered finger vein image and the second registered finger vein image, the mean square error of the valid corner point response value of each line is calculated from the first line to the last line, and if the difference value of the mean square error of the valid corner point response value of each line in the suspected overlapping area of the first registered finger vein image and the mean square error of the valid corner point response value of each line in the suspected overlapping area of the second registered finger vein image is smaller than the line sum difference threshold value, the suspected overlapping area is determined to be the highly suspected overlapping area of the two registered finger vein images.
The highly suspected overlapping region may be directly used as the fusion region.
After the highly suspected overlapping area is calculated, the cross-correlation coefficient may be calculated for the highly suspected overlapping areas of the two registered finger vein images by using a cross-correlation coefficient method, so as to determine whether the highly suspected overlapping areas in the two registered finger vein images have correlation. And if the cross correlation coefficient of the highly suspected overlapping area in the two registered finger vein images is larger than a preset cross correlation coefficient threshold value, judging that the highly suspected overlapping area is a fusion area of the two registered finger vein images.
The cross-correlation coefficient method can be implemented using existing techniques.
For a highly suspected overlapping region of the first registered finger vein image and the second registered finger vein image, one way to calculate the cross-correlation coefficient C is:
wherein m is the total number of rows of the highly suspected overlapping area, and n is the total number of columns of the highly suspected overlapping area. f (x)i,yj) For the xth finger vein image of the first registrationiLine yjGray value of pixel point of column, wherein, xiThe number of lines in the first registered finger vein image, y, which is the ith line of the highly suspected overlapping areajThe number of columns in the first registered finger vein image is the jth row of the highly suspected overlapping area.
Is the x-th suspected overlapping area in the second registered finger vein imagei *Line yj *Gray value of pixel point of column, wherein xi *The number of rows in the second registered finger vein image, y, which is the ith row of the highly suspected overlapping areaj *The number of columns in the second registered finger vein image is the jth row of the highly suspected overlapping area.
After C is calculated, if C is 0, it indicates no correlation, and if C is 1, it indicates complete correlation. In the correlation operation, the position of the maximum correlation coefficient is the position of the target image. And finally, comparing the C values obtained by the correlation operations for several times, and determining the position of the optimal registration point of the two images.
Therefore, if two registered finger vein images include a plurality of highly suspected overlapping areas with cross correlation coefficients larger than a preset cross correlation coefficient threshold value, the highly suspected overlapping area with the largest cross correlation coefficient is selected as a fusion area of the two registered finger vein images.
Fig. 7a shows a fused region 711 determined on the histogram 71 of the first effective corner point and a fused region 721 determined on the histogram 72 of the second effective corner point based on the histogram 71 of the first registered finger vein image (i.e., the schematic diagram of fig. 6 a) and the histogram 72 of the second effective corner point of the second registered finger vein image (i.e., the schematic diagram of fig. 6 b). The distribution map of the effective corner points visually shows the positions of the corner points on the registered finger vein image. And because the detection of the fusion region is based on the effective corner response value for matching, the fusion region can also be determined by adopting the method of the invention directly based on the effective corner response value recorded in the registered finger vein image.
Similarly, as shown in fig. 7b, a fused region 722 determined on the histogram 72 of the second valid corner and a fused region 731 determined on the histogram 73 of the third valid corner are determined based on the histogram 72 of the second registered finger vein image and the histogram 73 of the third valid corner of the third registered finger vein image (i.e., the schematic diagram of fig. 6 c).
The first registered finger vein image is preferably a finger vein image acquired at the maximum deflection to the left, the second registered finger vein image is preferably a finger vein image acquired at the middle direction, and the third registered finger vein image is preferably a finger vein image acquired at the maximum deflection to the right.
And step S406 is executed to fuse corresponding fusion regions in the plurality of registered finger vein images to obtain a registered finger vein template. Specifically, as shown in fig. 8, a distribution diagram 81 of new effective corner points obtained based on fusion of the fusion regions is shown, the distribution diagram 81 of effective corner points corresponds to the image region of the original registered finger vein image by using a corner point identifier 811, and is restored to the binary image, so as to obtain the registered finger vein template 91 shown in fig. 9.
And if the registered finger vein image is directly adopted to calculate the fusion area and carry out fusion, the registered finger vein image can be directly fused after the fusion area is calculated, so as to obtain the registered finger vein template. The fusion mode can be realized by adopting the existing image fusion mode.
In general, three registered finger vein images are used, and a fusion region of the first registered finger vein image and the second registered finger vein image, a fusion region of the first registered finger vein image and the third registered finger vein image, and a fusion region of the second registered finger vein image and the third registered finger vein image are calculated, respectively. In fact, the two registered finger vein images do not necessarily have a fusion region. For example, as shown in fig. 7a and 7b, the first registered finger vein image 71 and the second registered finger vein image 72 have a fusion region and can be fused, and the second registered finger vein image 72 and the third registered finger vein image 73 have a fusion region and can be fused, however, the first registered finger vein image 71 and the third registered finger vein image 73 have no fusion region.
Finally, when the user needs to identify, a finger vein detection request is generated, for example, the finger vein detection request is generated by contacting a sensor of the finger vein image acquisition device, and the steps S407 to S409 are triggered to be executed, so as to obtain a finger vein image to be compared and perform authentication. For example, when the user places a finger on the finger vein image capture device as shown in fig. 3, the finger vein image of the user can be acquired as the finger vein image to be compared. And then, when the registered finger vein template comprises the finger vein image to be compared, judging that the finger vein image to be compared is consistent with the registered finger vein template.
It should be noted that, in the present application, rows and columns are opposite. That is, for an image, if the pixel points extending along the first direction in the image are taken as rows, the pixel points extending along the second direction in the image are taken as columns, wherein the first direction is perpendicular to the second direction. For example, if the pixels extending in the length direction in the image are taken as rows, the pixels extending in the width direction in the image are taken as columns. Similarly, if the pixel points extending along the width direction in the image are taken as rows, the pixel points extending along the length direction in the image are taken as columns.
The embodiment provides a new finger vein finger image template generation method and a finger vein recognition method, and only one template needs to be compared, so that the comparison efficiency is improved. Meanwhile, the registered finger vein image of the embodiment comprises the registered finger vein image with the maximum deflection of the hand pointing to the first direction and the registered finger vein image with the maximum deflection of the hand pointing to the second direction, so that all acquisition angles are covered, all user use conditions are covered, and the accuracy and the stability of subsequent identification are improved.
Meanwhile, the angular point detection of the registered finger vein image can remarkably improve the response speed of the algorithm and improve the fusion accuracy. And the elimination of the point with the response value lower than the N value can effectively eliminate the interference of the low-gradient corner point (possibly caused by interference factors such as the outside world and the like), and can improve the accuracy and the response speed of the algorithm. The fusion mode is verified to be effective, the calculation response speed and accuracy can be effectively improved, and the fusion quality meets the design requirements. And finally, further determining a fusion area through the correlation, and further improving the accuracy.
Fig. 10 is a flowchart illustrating a method for identifying finger veins according to a preferred embodiment of the present invention, including:
step S1001, collecting finger vein registration images, and registering three states of maximum leftward deflection, middle state and maximum rightward deflection of fingers;
step S1002, judging image quality;
step S1003, image preprocessing;
step S1004, indicating a vein standard binary image;
step S1005, generating a user template I, a user template II and a user template III;
step S1006, a Harris angular point detection algorithm is respectively adopted for the first user template, the second user template and the third user template to calculate the lambda of the characteristic point and extract an angular point response value R, and a distribution diagram of the effective angular points of the templates is generated;
step S1008, fusing the three templates to generate a new user template;
step S1009, the user refers to a vein recognition operation;
step S1010, image acquisition and digital image processing are carried out;
step S1011, comparing the collected image with a new template;
in step S1012, if the identity determination condition is satisfied, the identity is legal, and if not, the identity is determined to be illegal, and the authentication is performed again, and step S1009 is performed.
Fig. 11 is a flowchart illustrating a method for determining an image fusion region according to a preferred embodiment of the present invention, which includes:
step S1101, along the line extension direction of the template, quickly calculating the sum SumR (x) of the corner response values R of all effective corners on the same x value, wherein x is a column serial number;
step S1102, whether the three templates have suspected overlapping areas or not is judged, the SumR (x) difference value of each x value of the three templates is smaller than the standard R1 value, if yes, the step S1102 is executed, and if not, the step S1102 is ended;
step S1103, calculating a mean square deviation σ r (x) of the valid corner response values of the suspected overlapping area and each x value;
step S1104, in the suspected overlapping area, if σ r (x) of each x value is smaller than the standard σ 1 value, step S1105 is executed, otherwise, the process is ended;
step S1105, calculating the mean square deviation value sigma R (y) of the effective angular point response value of each y value of the suspected overlapping area, wherein y is the line serial number;
step S1106, in the suspected overlapping area, if σ r (y) of each y value is smaller than the standard σ 2 value, step S1107 is executed, otherwise, the process is ended;
step S1107, preliminarily determine the position and size of the suspected overlapping area;
step S1108, determining the correlation degree of the suspected overlapping area block by using a cross correlation coefficient method, if the cross correlation coefficient is greater than a preset threshold value, judging the suspected overlapping area as a fusion area, confirming an optimal matching area, and executing step S1109, otherwise, ending;
and step S1109, completing the splicing, fusion and reduction among the templates.
Fig. 12 is a schematic diagram of a hardware structure of an electronic device according to the present invention, which includes:
at least one processor 1201; and the number of the first and second groups,
a memory 1202 communicatively coupled to at least one of the processors 1201; wherein the content of the first and second substances,
the memory 1202 stores instructions executable by at least one of the processors to enable the at least one of the processors to perform a finger vein recognition method as previously described.
Fig. 12 illustrates an example of one processor 1201.
The electronic device may further include: an input device 1203 and a display device 1204.
The processor 1201, the memory 1202, the input device 1203, and the display device 1204 may be connected by a bus or other means, and are illustrated as being connected by a bus.
The memory 1202 is a non-volatile computer-readable storage medium, and can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as program instructions/modules corresponding to the finger vein recognition method in the embodiment of the present application, for example, the method flow shown in fig. 1. The processor 1201 executes various functional applications and data processing, that is, implements the finger vein recognition method in the above-described embodiment, by executing nonvolatile software programs, instructions, and modules stored in the memory 1202.
The memory 1202 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the finger vein recognition method, and the like. Further, the memory 1202 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the memory 1202 may optionally include memory remotely located from the processor 1201, and such remote memory may be connected over a network to a device that performs the finger vein recognition method. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 1203 may receive the input user click and generate signal inputs related to user settings and function control of the finger-finger vein recognition method. The display device 1204 may include a display screen or the like.
The one or more modules stored in the memory 1202, when executed by the one or more processors 1201, perform the finger vein recognition method of any of the method embodiments described above.
The invention provides a novel finger vein finger image template generation method and a finger vein recognition method, and the comparison efficiency is improved because only one template needs to be compared. Meanwhile, the template covers all the acquisition angles, so that the acquisition angles of the fingers of the user are within the coverage range of the final template no matter what acquisition angles the fingers of the user are in, and the comparison accuracy, efficiency and stability are improved.
An embodiment of the present invention provides a storage medium storing computer instructions for performing all the steps of the finger vein recognition method as described above when the computer executes the computer instructions.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (12)
1. A finger vein recognition method is characterized by comprising the following steps:
responding to a finger vein registration request, and acquiring a binary registered finger vein image of a plurality of acquisition angles;
fusing a plurality of registered finger vein images to obtain a registered finger vein template;
responding to a finger vein detection request, and acquiring a binary finger vein image to be compared;
judging the consistency of the finger vein image to be compared and the registered finger vein template;
and performing identity verification on the finger vein image to be compared based on the consistency of the finger vein image to be compared and the registered finger vein template.
2. The finger vein recognition method according to claim 1, wherein the obtaining of the registered finger vein template after fusing the plurality of registered finger vein images specifically comprises:
carrying out corner detection on each registered finger vein image to obtain corners in the registered finger vein images, wherein each corner has a corner position and a corner response value;
taking the corner points of which the corner point response values meet the corner point threshold condition as effective corner points, wherein the corner point response values of the effective corner points are effective corner point response values;
taking a region, in which the relation between effective corner response values in the two registered finger vein images meets a preset fusion condition, as a fusion region of the two registered finger vein images;
and fusing corresponding fusion areas in the plurality of registered finger vein images to obtain a registered finger vein template.
3. The finger vein recognition method according to claim 2, wherein the preset fusion condition is:
the difference value between the sum of the effective corner response values of each column of one registered finger vein image and the sum of the effective corner response values of the corresponding column of the other registered finger vein image in the area is smaller than a column sum difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each column of one registered finger vein image and the mean square error of the effective corner response value of the corresponding column of the other registered finger vein image in the area is smaller than a column mean square error difference threshold value; and is
The difference value of the mean square error of the effective corner response value of each line of one registered finger vein image and the mean square error of the effective corner response value of the corresponding line of the other registered finger vein image in the region is smaller than a line mean square error difference threshold value.
4. The finger vein recognition method according to claim 3, wherein the step of taking a region in which a relationship between effective corner response values in two registered finger vein images satisfies a preset fusion condition as a fusion region of the two registered finger vein images specifically comprises:
calculating the sum of effective corner response values of each column for each registered finger vein image;
if an area exists, the difference value of the sum of the effective corner point response values of each column of the two registered finger vein images and the sum of the effective corner point response values of the corresponding column of the other registered finger vein image in the area is smaller than a column sum difference threshold value, and the two registered finger vein images are judged to have a suspected overlapping area;
respectively calculating the mean square deviation value of the effective corner response value of each column in the suspected overlapping area in the two registered finger vein images of the suspected overlapping area;
if the difference value of the mean square error of the effective corner point response value of each column of one registered finger vein image in the suspected overlapping area and the mean square error of the effective corner point response value of the corresponding column of the other registered finger vein image is less than a column mean square error difference value threshold, calculating the mean square error value of the effective corner point response value of each line in the suspected overlapping area in the two registered finger vein images in the suspected overlapping area respectively;
if the difference value of the mean square error of the effective corner point response value of each line of one registered finger vein image in the suspected overlapping area and the mean square error of the effective corner point response value of the corresponding line of the other registered finger vein image is less than the line mean square error difference threshold value, the suspected overlapping area is judged to be the high suspected overlapping area of the two registered finger vein images;
and determining a fusion area of the two registered finger vein images based on the high suspected overlapping area.
5. The finger vein recognition method according to claim 2, wherein the step of taking a region in which a relationship between effective corner response values in two registered finger vein images satisfies a preset fusion condition as a fusion region of the two registered finger vein images specifically comprises:
taking a region, in which the relation between effective corner response values in the two registered finger vein images meets a preset fusion condition, as a highly suspected overlapping region of the two registered finger vein images;
calculating the cross-correlation coefficient of the highly suspected overlapping area of the two registered finger vein images;
and if the cross-correlation coefficient is larger than a preset cross-correlation coefficient threshold value, judging that the high suspected overlapping area is a fusion area of the two registered finger vein images.
6. The method according to claim 5, wherein the determining that the suspected overlapping area is a fusion area of the two registered finger vein images further comprises:
and if the two registered finger vein images comprise a plurality of high suspected overlapping areas with cross correlation coefficients larger than a preset cross correlation coefficient threshold value, selecting the high suspected overlapping area with the maximum cross correlation coefficient as a fusion area of the two registered finger vein images.
7. The finger vein recognition method according to any one of claims 1 to 6, wherein the acquiring of the binarized registered finger vein images of a plurality of acquisition angles specifically comprises:
acquiring registered finger vein images of a plurality of different acquisition angles;
and rotating and adjusting the registered finger vein images of a plurality of different acquisition angles to the same angle, and generating a binary image with the same size by each registered finger vein template.
8. The finger vein recognition method according to claim 7, wherein the registered finger vein images of the plurality of different capturing angles include at least a registered finger vein image with a maximum deflection of a hand in a first direction and a registered finger vein image with a maximum deflection of a hand in a second direction, the first direction being opposite to the second direction.
9. The finger vein recognition method according to claim 8, wherein the acquiring registered finger vein images from a plurality of acquisition angles further comprises:
and detecting the acquisition angle of the registered finger vein image, and if the registered finger vein image with the maximum deflection of the hand pointing to the first direction is lacked or the registered finger vein image with the maximum deflection of the hand pointing to the second direction is lacked, executing a reminding operation.
10. The finger vein recognition method according to any one of claims 1 to 6, wherein the determining the consistency between the finger vein image to be compared and the registered finger vein template specifically comprises:
and if the registered finger vein template comprises the finger vein image to be compared, judging that the finger vein image to be compared is consistent with the registered finger vein template.
11. An electronic device, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to at least one of the processors; wherein the content of the first and second substances,
the memory stores instructions executable by at least one of the processors to enable the at least one of the processors to perform a method of finger vein recognition according to any one of claims 1 to 10.
12. A storage medium storing computer instructions for performing all the steps of the finger vein recognition method according to any one of claims 1 to 10 when the computer instructions are executed by a computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110850771.XA CN113609943B (en) | 2021-07-27 | 2021-07-27 | Finger vein recognition method, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110850771.XA CN113609943B (en) | 2021-07-27 | 2021-07-27 | Finger vein recognition method, electronic device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113609943A true CN113609943A (en) | 2021-11-05 |
| CN113609943B CN113609943B (en) | 2024-05-17 |
Family
ID=78305561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110850771.XA Active CN113609943B (en) | 2021-07-27 | 2021-07-27 | Finger vein recognition method, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113609943B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101110122A (en) * | 2007-08-31 | 2008-01-23 | 北京工业大学 | Large cultural heritage picture pattern split-joint method based on characteristic |
| CN102903085A (en) * | 2012-09-25 | 2013-01-30 | 福州大学 | Rapid image mosaic method based on corner matching |
| JP2014191814A (en) * | 2013-03-28 | 2014-10-06 | Hitachi Solutions Ltd | Finger vein authentication system, and authentication accuracy improvement method of finger vein authentication system |
| CN104376548A (en) * | 2014-11-07 | 2015-02-25 | 中国电子科技集团公司第二十八研究所 | Fast image splicing method based on improved SURF algorithm |
| CN104637041A (en) * | 2014-12-31 | 2015-05-20 | 浙江理工大学 | Wide fabric image acquiring and splicing method based on reference characteristics |
| WO2016072921A1 (en) * | 2014-11-07 | 2016-05-12 | Fingerprint Cards Ab | Fingerprint authentication using stitch and cut |
| CN106096569A (en) * | 2016-06-20 | 2016-11-09 | 广州中国科学院先进技术研究所 | A kind of finger vein identification method |
| CN107766776A (en) * | 2016-08-17 | 2018-03-06 | 广州广电运通金融电子股份有限公司 | A kind of finger vein identification method and device |
| CN107909085A (en) * | 2017-12-01 | 2018-04-13 | 中国科学院长春光学精密机械与物理研究所 | A kind of characteristics of image Angular Point Extracting Method based on Harris operators |
| US9990753B1 (en) * | 2017-01-11 | 2018-06-05 | Macau University Of Science And Technology | Image stitching |
| CN109345458A (en) * | 2018-10-25 | 2019-02-15 | 常州市第二人民医院 | One kind splicing ultrasound image method based on improved harris Corner Detection |
| CN109934114A (en) * | 2019-02-15 | 2019-06-25 | 重庆工商大学 | A kind of finger vena template generation and more new algorithm and system |
| CN111639559A (en) * | 2020-05-15 | 2020-09-08 | 圣点世纪科技股份有限公司 | Finger vein machine learning identification method and device based on topographic concave-convex characteristics |
| CN112101332A (en) * | 2020-11-23 | 2020-12-18 | 北京圣点云信息技术有限公司 | Feature extraction and comparison method and device based on 3D finger veins |
| CN112883356A (en) * | 2021-03-31 | 2021-06-01 | 中国工商银行股份有限公司 | Identity authentication method, device and equipment |
-
2021
- 2021-07-27 CN CN202110850771.XA patent/CN113609943B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101110122A (en) * | 2007-08-31 | 2008-01-23 | 北京工业大学 | Large cultural heritage picture pattern split-joint method based on characteristic |
| CN102903085A (en) * | 2012-09-25 | 2013-01-30 | 福州大学 | Rapid image mosaic method based on corner matching |
| JP2014191814A (en) * | 2013-03-28 | 2014-10-06 | Hitachi Solutions Ltd | Finger vein authentication system, and authentication accuracy improvement method of finger vein authentication system |
| CN104376548A (en) * | 2014-11-07 | 2015-02-25 | 中国电子科技集团公司第二十八研究所 | Fast image splicing method based on improved SURF algorithm |
| WO2016072921A1 (en) * | 2014-11-07 | 2016-05-12 | Fingerprint Cards Ab | Fingerprint authentication using stitch and cut |
| CN104637041A (en) * | 2014-12-31 | 2015-05-20 | 浙江理工大学 | Wide fabric image acquiring and splicing method based on reference characteristics |
| CN106096569A (en) * | 2016-06-20 | 2016-11-09 | 广州中国科学院先进技术研究所 | A kind of finger vein identification method |
| CN107766776A (en) * | 2016-08-17 | 2018-03-06 | 广州广电运通金融电子股份有限公司 | A kind of finger vein identification method and device |
| US9990753B1 (en) * | 2017-01-11 | 2018-06-05 | Macau University Of Science And Technology | Image stitching |
| CN107909085A (en) * | 2017-12-01 | 2018-04-13 | 中国科学院长春光学精密机械与物理研究所 | A kind of characteristics of image Angular Point Extracting Method based on Harris operators |
| CN109345458A (en) * | 2018-10-25 | 2019-02-15 | 常州市第二人民医院 | One kind splicing ultrasound image method based on improved harris Corner Detection |
| CN109934114A (en) * | 2019-02-15 | 2019-06-25 | 重庆工商大学 | A kind of finger vena template generation and more new algorithm and system |
| CN111639559A (en) * | 2020-05-15 | 2020-09-08 | 圣点世纪科技股份有限公司 | Finger vein machine learning identification method and device based on topographic concave-convex characteristics |
| CN112101332A (en) * | 2020-11-23 | 2020-12-18 | 北京圣点云信息技术有限公司 | Feature extraction and comparison method and device based on 3D finger veins |
| CN112883356A (en) * | 2021-03-31 | 2021-06-01 | 中国工商银行股份有限公司 | Identity authentication method, device and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113609943B (en) | 2024-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110674712A (en) | Interactive behavior recognition method and device, computer equipment and storage medium | |
| JP5574515B2 (en) | Biometric device and method | |
| EP1380984A2 (en) | Method and apparatus for supporting a biometric registration performed on an authentication server | |
| CN110807491A (en) | License plate image definition model training method, definition detection method and device | |
| JPH11250261A (en) | Fingerprint picture characteristic correlator | |
| WO2011022212A1 (en) | Automatic identification of fingerprint inpainting target areas | |
| JP5468332B2 (en) | Image feature point extraction method | |
| US10621765B2 (en) | Image reconstruction | |
| US11436774B2 (en) | Pattern mapping | |
| EP1318484A2 (en) | Apparatus and method for authenticating a user by employing feature points of a fingerprint image of the user | |
| CN109492647A (en) | A kind of power grid robot barrier object recognition methods | |
| US20090022405A1 (en) | System and method for video object identification | |
| CN116188024A (en) | Medical safety payment system | |
| CN109313708B (en) | Image matching method and vision system | |
| KR102260556B1 (en) | Deep learning-based parking slot detection method and apparatus integrating global and local information | |
| US10331936B2 (en) | Duplicate pattern reconstructions | |
| CN113609943B (en) | Finger vein recognition method, electronic device and storage medium | |
| CN112525355A (en) | Image processing method, device and equipment | |
| CN109840515B (en) | Face posture adjusting method and device and terminal | |
| CN110084157B (en) | Data processing method and device for image re-recognition | |
| CN112070035A (en) | Target tracking method and device based on video stream and storage medium | |
| CN108364024B (en) | Image matching method and device, computer equipment and storage medium | |
| CN108710849B (en) | Fingerprint block splicing method and system | |
| KR20180072517A (en) | Method for detecting borderline between iris and sclera | |
| JPH01271883A (en) | Detecting system for center of fingerprint |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |