CN114898413B - Vein identification method based on image contour direction field under complex background - Google Patents
Vein identification method based on image contour direction field under complex background Download PDFInfo
- Publication number
- CN114898413B CN114898413B CN202210828710.8A CN202210828710A CN114898413B CN 114898413 B CN114898413 B CN 114898413B CN 202210828710 A CN202210828710 A CN 202210828710A CN 114898413 B CN114898413 B CN 114898413B
- Authority
- CN
- China
- Prior art keywords
- image
- pixel point
- vein
- continuous pixel
- recognized
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a vein identification method based on an image contour direction field under a complex background, which belongs to the technical field of fingerprint image identification and comprises the following steps: judging whether the image to be identified belongs to empty image or invalid image processing in the environment without fingers or not based on the average gray value; obtaining an edge detection image through edge detection, extracting contour direction field information of the edge detection image, obtaining a plurality of groups of continuous pixel point sets of the edge detection image, and judging whether fingers exist in the collected vein image or not based on the contour direction field information and the plurality of groups of continuous pixel point sets; and carrying out pixel correction on the multiple groups of continuous pixel point sets, and judging whether fingers exist in the collected vein image again based on the corrected multiple groups of continuous pixel point sets. The method judges whether the finger exists in the collected vein image or not in a multi-dimensional and multi-stage fusion mode, can acquire vein data information in an all-around mode under a complex background, judges the vein information of the image, and improves the robustness of identification.
Description
Technical Field
The invention relates to the technical field of fingerprint image identification, in particular to a vein identification method based on an image contour direction field under a complex background.
Background
The vein recognition technology collects the characteristics of the inside of a body, is a distribution image of blood vessel flow in a living finger, is a natural in-vivo characteristic living body recognition technology, is extremely difficult to forge and tamper, and has the characteristics of high anti-counterfeiting performance, high accuracy, stable characteristics and convenience in use compared with other biological recognition technologies, so that the vein recognition technology is widely applied to judicial law, education, social security, finance, medical treatment, sanitation, safety production, fitness venues, home door locks, safe cases and the like. With the continuous expansion of the application scenes, the requirements on the volume, the use convenience and the like of the vein equipment are higher and higher. In order to meet the requirement, a manufacturer of vein equipment develops a point-contact vein recognition device, such as the vein equipment disclosed in the chinese utility model patent with the authorized publication number of CN212749859U and the name of point-contact vein equipment and the chinese invention patent application with the application publication number of CN113505754A and the name of a pure flat ultrathin vein recognition device and a vein recognition method, and the point-contact vein recognition device has no finger placing groove, and can work only by lightly touching the surface of the equipment with the finger tip, so that the volume is small, and the applicable scenes of the point-contact vein recognition device can be greatly increased; and because of not having the finger standing groove, can not restrict user's finger locating position, it is more convenient to use.
Compared with the traditional side-lighting or upper-lighting vein equipment, the finger placement position is not limited by the point-contact vein equipment, and the finger is located in an open space, so that the camera shoots the vein, and simultaneously the complex environment background of the space where the finger is located can be shot, or an empty picture without vein information can be shot by mistakenly touching the equipment by a user. When a user uses the vein recognition equipment at every time, the problems of inconsistent background illumination, complex and changeable background images, large gesture difference of collected fingers and the like on the periphery can influence the vein recognition rate, and how to improve the recognition success rate of the point-contact equipment under different complex backgrounds needs to be solved urgently.
Disclosure of Invention
The invention provides a vein recognition method based on an image contour direction field under a complex background, which aims to solve the problem of low vein recognition rate caused by complex background images, inconsistent illumination, large finger posture difference and the like of vein images acquired by traditional point-contact vein equipment.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
the invention relates to a vein identification method based on an image contour direction field under a complex background, which comprises the following steps:
(1) collecting vein images to form images to be identified;
(2) calculating the average gray value of the image to be recognized, judging whether the image to be recognized belongs to a black image or a bright image or not based on the average gray value, if the image to be recognized is judged not to be the black image or the bright image, entering the step (3), and if the image to be recognized is judged to be the black image or the bright image, ending vein recognition;
(3) adjusting the gray scale of the image to be identified based on the average gray scale value of the image to be identified;
(4) carrying out edge detection on the image to be identified after the gray level adjustment to obtain an edge detection image;
(5) extracting contour direction field information of the edge detection image, acquiring a plurality of groups of continuous pixel point sets of the edge detection image, judging whether fingers exist in the collected vein image or not based on the contour direction field information and the plurality of groups of continuous pixel point sets, entering the step (6) if the judgment result indicates that the fingers exist, and ending vein identification if the judgment result indicates that the fingers do not exist;
(6) performing pixel correction on the multiple groups of continuous pixel point sets, judging whether fingers exist in the collected vein image again based on the corrected multiple groups of continuous pixel point sets, if so, entering the step (7), and if not, ending the vein identification;
(7) and (4) when the judgment result in the step (6) is that the finger exists, carrying out vein recognition on the vein image.
Preferably, in the step (2), a minimum grayscale threshold and a maximum grayscale threshold are set, and when the average grayscale value of the image to be recognized is calculated to be smaller than the minimum grayscale threshold, the image to be recognized is determined to be a black image; and when the average gray value of the image to be recognized is calculated to be larger than the maximum gray threshold value, judging that the image to be recognized is a bright image.
Preferably, the condition of adjusting the gray scale of the image to be recognized in step (3) is as follows:
when the average gray value of the image to be recognized calculated in the step (2) is 30-60, adjusting the gray value of each pixel point of the image to be recognized in proportion to enable the average gray value of the image to be recognized to be 90;
when the average gray value of the image to be recognized calculated in the step (2) is between 60 and 130, adjusting the gray value of each pixel point of the image to be recognized in proportion to enable the average gray value of the image to be recognized to be 80;
and (3) when the average gray value of the image to be recognized calculated in the step (2) is 130-230, proportionally adjusting the gray value of each pixel point of the image to be recognized to enable the average gray value of the image to be recognized to be 120.
Preferably, the step (4) of performing edge detection by using a Sobel algorithm includes: respectively performing horizontal edge convolution and vertical edge convolution on the image to be recognized after the gray level adjustment by utilizing a Sobel horizontal edge template and a Sobel vertical edge template;
the values of the Sobel horizontal edge template are as follows:
the values of the Sobel vertical edge template are as follows:
preferably, the formula for extracting the contour direction field information of the edge detection image in step (5) is as follows:
in the formula (I), the compound is shown in the specification,
represents the magnitude of the directional field of the spot, (ii) andx，y) The coordinates of the point are represented by,
the derivative of the point in the x-direction is indicated,
representing the derivative of the point in the y-direction.
Preferably, the step (5) of obtaining a plurality of groups of continuous pixel point sets of the edge detection image includes: and traversing pixel points of the edge detection image, and acquiring a plurality of groups of continuous pixel point sets in the edge detection image by using an eight-neighborhood edge tracking detection method.
Preferably, before said step (5) determining whether there is a finger in the collected vein image based on the contour direction field information and the sets of multiple groups of continuous pixels, the method further includes: and respectively calculating the lengths of a plurality of groups of continuous pixel point sets, wherein the lengths are the number of the pixel points in the continuous pixel point sets, and if the lengths of the continuous pixel point sets are less than one half of the width of the edge detection image, screening the continuous pixel point sets.
Preferably, the step (5) judges whether the collected vein image has a finger or not based on the contour direction field information and the sets of the multiple groups of continuous pixel points, and the judgment is based on the following basis:
(a) the length of the continuous pixel point set is more than or equal to 80% of the width of the edge detection image;
(b) respectively calculating absolute values of direction field differences of any two pixels in a multi-group continuous pixel point set, wherein the absolute values are less than or equal to 45 degrees;
(c) calculating the total number of pixels in a plurality of groups of continuous pixel point sets, wherein the size of the edge detection image is m x n, and the proportion of the total number of the pixels in the plurality of groups of continuous pixel point sets to the total number of the pixels in the edge detection image is more than or equal to (1/n)% and less than or equal to 70%;
(d) calculating the number of pixels with the direction field between 0 and 45 degrees and between 135 and 180 degrees in the multi-group continuous pixel point set, wherein the proportion of the number in the total number of the pixels in the multi-group continuous pixel point set is more than or equal to 50 percent;
if the number of the conditions satisfied in the judgment criteria (a) to (d) is less than 2, the images are judged to be vein-free images, and if the number of the conditions satisfied is greater than or equal to 2, the images are judged to be vein images of the finger.
Preferably, the step (6) further includes performing pixel correction on the multiple groups of continuous pixel point sets, and further determining whether a finger exists in the vein image based on the corrected multiple groups of continuous pixel point sets.
Preferably, the method for performing pixel correction on the multiple groups of continuous pixel point sets in step (6) includes: classifying the pixels with the direction field range of 60-120 degrees of the pixels in the multi-group continuous pixel point set as invalid points to obtain a corrected multi-group continuous pixel point set; and judging whether the collected vein image has a finger or not again based on the corrected multi-group continuous pixel point set, wherein the judgment basis is as follows:
(e) calculating the ratio of the number of pixels with the direction field range of 0-45 degrees and 135-180 degrees in the corrected multi-group continuous pixel point set to the total number of pixels in the corrected multi-group continuous pixel point set, wherein the ratio is more than 0.5;
(f) calculating the total number of pixels in the corrected multi-group continuous pixel point set, wherein the proportion of the total number of the pixels in the corrected multi-group continuous pixel point set to the total number of the pixels in the edge detection image is more than or equal to 5%;
and if the two judgment bases of (e) and (f) are simultaneously satisfied, determining that the finger exists in the vein image, otherwise, determining that the finger does not exist in the vein image.
Compared with the prior art, the technical scheme provided by the invention has the following beneficial effects:
the vein identification method based on the image contour direction field under the complex background comprises the steps of eliminating invalid images through the average gray value of the images, extracting the contours of the vein images, obtaining the direction field information of the contours of the vein images, judging whether fingers exist in the collected images to be identified or not based on the image contour direction field information, realizing region identification and registration of veins, mapping the image gray space to the direction information space, avoiding the influence of the complex background on vein identification, obtaining vein data information in an all-around mode under the complex background, rapidly judging the image vein information according to the parallel and vertical relations of the vein contour direction field, and improving the robustness of algorithm identification.
Drawings
FIG. 1 is a flow chart of a vein identification method based on an image contour direction field under a complex background;
FIG. 2 is a flowchart of determining whether a finger exists in the collected vein image based on the contour direction field information and the sets of multiple groups of continuous pixels in step (5);
fig. 3 is a flowchart of determining again whether a finger exists in the acquired vein image based on the corrected sets of multiple groups of continuous pixel points in step (6).
Detailed Description
For further understanding of the present invention, the present invention will be described in detail with reference to examples, which are provided for illustration of the present invention but are not intended to limit the scope of the present invention.
Referring to fig. 1, the present invention relates to a vein recognition method based on image contour direction field under complex background, which comprises the following steps:
(1) acquiring a vein image, performing interlaced sampling on the vein image, for example, if the vein image comprises 4 rows and 8 columns, intercepting parts of the 1 st row and the 3 rd row, and 2, 4, 6 and 8 columns to form an image to be identified; through sampling, the image size is reduced and the operation speed is improved under the condition of ensuring that the image information is not incomplete as much as possible.
(2) Setting a minimum gray threshold and a maximum gray threshold, wherein the minimum gray threshold is 30, the maximum gray threshold is 230, calculating an average gray value of the image to be recognized, judging whether the image to be recognized belongs to a black image or a bright image based on the average gray value, namely judging that the image to be recognized is the black image when the average gray value of the image to be recognized is calculated to be less than the minimum gray threshold 30; when the average gray value of the image to be recognized is calculated to be larger than the maximum gray threshold value 230, judging that the image to be recognized is a bright image, and if the image to be recognized is judged to be a black image or a bright image, ending vein recognition; and (4) if the image to be identified is not the black image or the bright image, entering the step (3) and finishing the screening of the first stage of the vein image. Because of the vein collection process surrounding environment is complicated various, user's use habit scheduling problem not inconsistent, overexposure or mistake touch etc. probably appear to lead to the vein image of gathering to appear luminance too high or low problem, this embodiment screens out the screening of bright picture or black picture through average grey value, avoids this type of image to carry out vein identification, can effectively improve vein identification efficiency.
(3) Adjusting the gray scale of the image to be identified based on the average gray scale value of the image to be identified, wherein the condition for adjusting the gray scale of the image to be identified is as follows: when the average gray value of the image to be recognized calculated in the step (2) is 30-60, adjusting the gray value of each pixel point of the image to be recognized in proportion to enable the average gray value of the image to be recognized to be 90; when the average gray value of the image to be recognized calculated in the step (2) is between 60 and 130, adjusting the gray value of each pixel point of the image to be recognized in proportion to enable the average gray value of the image to be recognized to be 80; and (3) when the average gray value of the image to be recognized calculated in the step (2) is 130-230, proportionally adjusting the gray value of each pixel point of the image to be recognized to enable the average gray value of the image to be recognized to be 120. Therefore, the image with lower gray scale can be subjected to gray scale stretching, the image with higher gray scale is subjected to gray scale compression, the gray scale balance of the image can be realized, the image display effect is improved, the image enhancement is realized, and the vein recognition at the later stage is facilitated.
(4) Performing edge detection on the image to be identified after the gray level adjustment by adopting any one algorithm of a Canny algorithm, a Sobel algorithm and a Laplace algorithm to obtain an edge detection image;
the method adopts a Sobel algorithm to carry out edge detection, and comprises the steps of respectively carrying out horizontal edge convolution and vertical edge convolution on an image to be recognized after gray adjustment by utilizing a Sobel horizontal edge template and a Sobel vertical edge template;
the values of the Sobel horizontal edge template are as follows:
the values of the Sobel vertical edge template are as follows:
(5) extracting contour direction field information of the edge detection image, wherein the calculation formula is as follows:
in the formula (I), the compound is shown in the specification,
represents the magnitude of the directional field of the spot, (ii) andx，y) The coordinates of the point are represented by,
the derivative of this point in the x-direction is indicated,
represents the derivative of the point in the y-direction;
acquiring a plurality of groups of continuous pixel point sets of an edge detection image, comprising: traversing pixel points of the edge detection image, acquiring a plurality of groups of continuous pixel point sets in the edge detection image by using an eight-neighborhood edge tracking detection method, respectively calculating the lengths of the plurality of groups of continuous pixel point sets, wherein the lengths are the number of the pixel points in the continuous pixel point sets, and screening out the continuous pixel point sets if the lengths of the continuous pixel point sets are less than one half of the width of the edge detection image;
if in other embodiments, the vein image collected is that the included angle between the connecting line of the fingertip and the finger root and the horizontal axis is smaller than the included angle between the connecting line of the fingertip and the finger root and the vertical axis, and the vein image is rotated by 90 degrees. Theoretically, the collection of multiple groups of continuous pixel points of the obtained edge detection image should be 4 groups, that is: an upper finger contour edge, a lower finger contour edge, a left finger contour edge and a right finger contour edge, and upper finger contour edge = lower finger contour edge > left finger contour edge = right finger contour edge, however, because there are other complex backgrounds in the vein collection process, the number of sets of continuous pixels in the obtained edge detection image is much more than 4, in this embodiment, a part of non-contour pixel sets are screened out by the length of the continuous pixel sets, and in this embodiment, it is considered that, if the length of the pixel sets is less than or equal to the width of the edge detection image, namely, the width of the left finger outline edge and the width of the right finger outline edge are half of the width of the left finger outline edge and the right finger outline edge, the pixel point set is considered to be a non-finger outline pixel point set, and the pixel point set is eliminated, so that the subsequent operation amount is reduced, the operation speed is improved, and the memory loss of the vein equipment is reduced.
Whether the collected vein image has a finger or not is judged based on the contour direction field information and the multiple groups of continuous pixel point sets, if the judgment result is that the finger exists, the step (6) is carried out, if the judgment result is that the finger does not exist, the vein recognition is finished, and the judgment basis is that:
(a) respectively calculating the lengths of a plurality of groups of continuous pixel point sets, wherein the lengths are the number of the pixel points in the continuous pixel point sets, and the lengths of the continuous pixel point sets are more than or equal to 80% of the width of the edge detection image;
(b) respectively calculating absolute values of direction field differences of any two pixels in a multi-group continuous pixel point set, wherein the absolute values are less than or equal to 45 degrees;
(c) calculating the total number of pixels in a plurality of groups of continuous pixel point sets, wherein the size of the edge detection image is m x n, and the proportion of the total number of the pixels in the plurality of groups of continuous pixel point sets to the total number of the pixels in the edge detection image is more than or equal to (1/n)% and less than or equal to 70%;
(d) calculating the number of pixels with the direction field between 0 and 45 degrees and between 135 and 180 degrees in the multi-group continuous pixel point set, wherein the proportion of the number in the total number of the pixels in the multi-group continuous pixel point set is more than or equal to 50 percent;
and if the number of the satisfied conditions in the judgment bases a-d is less than 2, judging the image without veins, and if the number of the satisfied conditions is more than or equal to 2, judging the image with veins.
As described above, theoretically, only four finger contour edges, i.e., upper, lower, left, and right, are present, and the left and right finger contour edges are approximately equal to the width of the edge detection image, in this embodiment, it is considered that at least one complete finger contour edge is also detected; in the vein collection process, the fingers are generally horizontally placed, and even if the fingers are inclined, the inclination angle does not exceed 45 degrees; therefore, the judgment bases of (a) - (d) are set, so that the vein images under a simple environment can be effectively screened out, the non-vein images are removed, the vein recognition of the images is avoided, and the vein recognition efficiency is effectively improved.
(6) Carrying out pixel correction on a plurality of groups of continuous pixel point sets, comprising the following steps: classifying the pixels with the direction field range of 60-120 degrees of the pixels in the multi-group continuous pixel point set as invalid points to obtain a corrected multi-group continuous pixel point set; judging whether the collected vein image has a finger or not again based on the corrected multi-group continuous pixel point set, namely finishing the screening of the third stage of the vein image, if the judgment result is that the finger exists, entering the step (7), if the judgment result is that the finger does not exist, finishing the vein identification, and judging whether the finger exists in the vein image according to the following steps:
(e) calculating the ratio of the number of pixels with the direction field range of pixel between 0 and 45 degrees and between 135 and 180 degrees in the corrected multi-group continuous pixel point set to the total number of pixels in the corrected multi-group continuous pixel point set, wherein the ratio is greater than 0.5;
(f) calculating the total number of pixels in the corrected multi-group continuous pixel point set, wherein the proportion of the total number of the pixels in the corrected multi-group continuous pixel point set to the total number of the pixels in the edge detection image is more than or equal to 5%;
and if the two judgment bases of (e) and (f) are simultaneously satisfied, the image is regarded as a vein image, otherwise, the image is a non-vein image.
(6) And (5) when the judgment result in the step (5) is that the finger exists, carrying out vein recognition on the vein image.
As described above, in the vein collection process, the finger is generally placed horizontally, and even if the finger is tilted, the tilt angle does not exceed 45 degrees, so that if the direction field of the pixel point is between 60 ° and 120 °, the pixel point can be considered as a contour point of a complex background, and therefore the pixel point is considered as an invalid point; through the scheme, the vein images under the complex background can be effectively screened, the non-vein images are removed, the vein recognition of the images is avoided, and the vein recognition efficiency is effectively improved.
The present invention has been described in detail with reference to the embodiments, but the description is only for the preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. All equivalent changes and modifications made within the scope of the present invention shall fall within the scope of the present invention.
Claims (10)
1. A vein identification method based on an image contour direction field under a complex background is characterized in that: which comprises the following steps:
(1) collecting vein images to form images to be identified;
(2) calculating the average gray value of the image to be recognized, judging whether the image to be recognized belongs to a black image or a bright image or not based on the average gray value, if the image to be recognized is judged not to be the black image or the bright image, entering the step (3), and if the image to be recognized is judged to be the black image or the bright image, ending vein recognition;
(3) adjusting the gray scale of the image to be identified based on the average gray scale value of the image to be identified;
(4) carrying out edge detection on the image to be identified after the gray level adjustment to obtain an edge detection image;
(5) extracting contour direction field information of the edge detection image, acquiring a plurality of groups of continuous pixel point sets of the edge detection image, judging whether fingers exist in the collected vein image or not based on the contour direction field information and the plurality of groups of continuous pixel point sets, if so, entering a step (6), and if not, ending vein identification;
(6) performing pixel correction on the multiple groups of continuous pixel point sets, judging whether fingers exist in the collected vein image again based on the corrected multiple groups of continuous pixel point sets, if so, entering the step (7), and if not, ending the vein identification;
(7) and (4) when the judgment result in the step (6) is that the finger exists, carrying out vein recognition on the vein image.
2. The vein identification method based on the image contour direction field under the complex background according to claim 1, characterized in that: in the step (2), a minimum gray threshold and a maximum gray threshold are set, and when the average gray value of the image to be recognized is calculated to be smaller than the minimum gray threshold, the image to be recognized is judged to be a black image; and when the average gray value of the image to be recognized is calculated to be larger than the maximum gray threshold value, judging that the image to be recognized is a bright image.
3. The vein identification method based on the image contour direction field under the complex background according to claim 2, characterized in that: the condition for adjusting the gray scale of the image to be identified in the step (3) is as follows:
when the average gray value of the image to be recognized calculated in the step (2) is 30-60, adjusting the gray value of each pixel point of the image to be recognized in proportion to enable the average gray value of the image to be recognized to be 90;
when the average gray value of the image to be recognized calculated in the step (2) is between 60 and 130, adjusting the gray value of each pixel point of the image to be recognized in proportion to enable the average gray value of the image to be recognized to be 80;
and (3) when the average gray value of the image to be recognized calculated in the step (2) is 130-230, proportionally adjusting the gray value of each pixel point of the image to be recognized to enable the average gray value of the image to be recognized to be 120.
4. The vein identification method based on the image contour direction field under the complex background according to claim 1, characterized in that: the step (4) of adopting Sobel algorithm to carry out edge detection comprises the following steps: respectively performing horizontal edge convolution and vertical edge convolution on the image to be recognized after the gray level adjustment by utilizing a Sobel horizontal edge template and a Sobel vertical edge template;
the values of the Sobel horizontal edge template are as follows:
the values of the Sobel vertical edge template are as follows:
5. the vein identification method based on the image contour direction field under the complex background according to claim 1, characterized in that: the formula for extracting the contour direction field information of the edge detection image in the step (5) is as follows:
in the formula (I), the compound is shown in the specification,
represents the magnitude of the directional field of the spot, (ii) andx，y) The coordinates of the point are represented by a coordinate,
the derivative of the point in the x-direction is indicated,
representing the derivative of the point in the y direction.
6. The vein identification method based on the image contour direction field under the complex background according to claim 5, wherein: the step (5) of obtaining a plurality of groups of continuous pixel point sets of the edge detection image comprises the following steps: and traversing pixel points of the edge detection image, and acquiring a plurality of groups of continuous pixel point sets in the edge detection image by using an eight-neighborhood edge tracking detection method.
7. The vein identification method based on the image contour direction field under the complex background according to claim 6, wherein: before the step (5) judges whether the finger exists in the collected vein image based on the contour direction field information and the multiple groups of continuous pixel point sets, the method further comprises the following steps: and respectively calculating the lengths of a plurality of groups of continuous pixel point sets, wherein the lengths are the number of the pixel points in the continuous pixel point sets, and screening the continuous pixel point sets if the lengths of the continuous pixel point sets are less than one half of the width of the edge detection image.
8. The vein identification method based on the image contour direction field under the complex background according to claim 7, wherein: and (5) judging whether the collected vein image has a finger or not based on the contour direction field information and the multiple groups of continuous pixel point sets, wherein the judgment basis is as follows:
(a) the length of the continuous pixel point set is more than or equal to 80% of the width of the edge detection image;
(b) respectively calculating absolute values of direction field differences of any two pixels in a multi-group continuous pixel point set, wherein the absolute values are less than or equal to 45 degrees;
(c) calculating the total number of pixels of a plurality of groups of continuous pixel point sets, wherein the size of the edge detection image is m × n, and the proportion of the total number of pixels of the plurality of groups of continuous pixel point sets to the total number of pixels of the edge detection image is more than or equal to (1/n)% and less than or equal to 70%;
(d) calculating the number of pixels with a direction field between 0 and 45 degrees and between 135 and 180 degrees in the multi-group continuous pixel point set, wherein the number accounts for more than or equal to 50 percent of the total number of the pixels in the multi-group continuous pixel point set;
if the number of the conditions satisfied in the judgment criteria (a) to (d) is less than 2, the images are judged to be vein-free images, and if the number of the conditions satisfied is greater than or equal to 2, the images are judged to be vein images of the finger.
9. The method for recognizing veins based on image contour direction field in complex background as claimed in claim 8, wherein: the specific way of performing pixel correction on the multi-group continuous pixel point set in the step (6) is as follows: and classifying the pixels with the direction field range of 60-120 degrees of the pixels in the multi-group continuous pixel point set as invalid points to obtain a corrected multi-group continuous pixel point set.
10. The vein identification method based on the image contour direction field under the complex background according to claim 9, wherein: and (4) judging whether the collected vein image has a finger again based on the corrected multiple groups of continuous pixel point sets in the step (6), wherein the judgment basis is as follows:
(e) calculating the ratio of the number of pixels with the direction field range of 0-45 degrees and 135-180 degrees in the corrected multi-group continuous pixel point set to the total number of pixels in the corrected multi-group continuous pixel point set, wherein the ratio is more than 0.5;
(f) calculating the total number of pixels in the corrected multi-group continuous pixel point set, wherein the proportion of the total number of the pixels in the corrected multi-group continuous pixel point set to the total number of the pixels in the edge detection image is more than or equal to 5%;
and if the two judgment bases of (e) and (f) are simultaneously satisfied, determining that the finger exists in the vein image, otherwise, determining that the finger does not exist in the vein image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210828710.8A CN114898413B (en) | 2022-07-15 | 2022-07-15 | Vein identification method based on image contour direction field under complex background |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210828710.8A CN114898413B (en) | 2022-07-15 | 2022-07-15 | Vein identification method based on image contour direction field under complex background |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114898413A CN114898413A (en) | 2022-08-12 |
| CN114898413B true CN114898413B (en) | 2022-09-13 |
Family
ID=82730040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210828710.8A Active CN114898413B (en) | 2022-07-15 | 2022-07-15 | Vein identification method based on image contour direction field under complex background |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114898413B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115311696B (en) * | 2022-10-11 | 2023-02-28 | 山东圣点世纪科技有限公司 | Finger area detection method based on vein texture features |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103310196A (en) * | 2013-06-13 | 2013-09-18 | 黑龙江大学 | Finger vein recognition method by interested areas and directional elements |
| JP5951817B1 (en) * | 2015-02-09 | 2016-07-13 | コリセン カンパニー リミテッドＫｏｒｅｃｅｎ Ｃｏ．，Ｌｔｄ． | Finger vein authentication system |
| CN108830158A (en) * | 2018-05-16 | 2018-11-16 | 天津大学 | The vein area-of-interest exacting method that finger contours and gradient distribution blend |
| CN109190517A (en) * | 2018-08-14 | 2019-01-11 | 北京凌云天润智能科技有限公司 | A kind of finger venous image edge extracting and direction antidote |
-
2022
- 2022-07-15 CN CN202210828710.8A patent/CN114898413B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103310196A (en) * | 2013-06-13 | 2013-09-18 | 黑龙江大学 | Finger vein recognition method by interested areas and directional elements |
| JP5951817B1 (en) * | 2015-02-09 | 2016-07-13 | コリセン カンパニー リミテッドＫｏｒｅｃｅｎ Ｃｏ．，Ｌｔｄ． | Finger vein authentication system |
| CN108830158A (en) * | 2018-05-16 | 2018-11-16 | 天津大学 | The vein area-of-interest exacting method that finger contours and gradient distribution blend |
| CN109190517A (en) * | 2018-08-14 | 2019-01-11 | 北京凌云天润智能科技有限公司 | A kind of finger venous image edge extracting and direction antidote |
Non-Patent Citations (1)
| Title |
|---|
| 手指静脉图像感兴趣区域快速提取方法研究;李杰 等;《电子测量与仪器学报》;20200430;第34卷(第4期);第42-49页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114898413A (en) | 2022-08-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107341473B (en) | Palm characteristic recognition method, palm characteristic identificating equipment and storage medium | |
| CN109815850B (en) | Iris image segmentation and positioning method, system and device based on deep learning | |
| CN109409366B (en) | Distorted image correction method and device based on angular point detection | |
| WO2021027364A1 (en) | Finger vein recognition-based identity authentication method and apparatus | |
| CN105740780B (en) | Method and device for detecting living human face | |
| WO2017059591A1 (en) | Finger vein identification method and device | |
| CN107316031A (en) | The image characteristic extracting method recognized again for pedestrian | |
| CN105335725A (en) | Gait identification identity authentication method based on feature fusion | |
| Yang et al. | Superpixel based finger vein ROI extraction with sensor interoperability | |
| CN110032932B (en) | Human body posture identification method based on video processing and decision tree set threshold | |
| CN110555382A (en) | Finger vein identification method based on deep learning and Wasserstein distance measurement | |
| CN114898413B (en) | Vein identification method based on image contour direction field under complex background | |
| CN106023187A (en) | Image registration method based on SIFT feature and angle relative distance | |
| CN111091075A (en) | Face recognition method and device, electronic equipment and storage medium | |
| US11475707B2 (en) | Method for extracting image of face detection and device thereof | |
| CN106778813B (en) | The self-adaption cluster partitioning algorithm of depth image | |
| CN110084238A (en) | Finger vena image segmentation method, device and storage medium based on LadderNet network | |
| CN111046825A (en) | Human body posture recognition method, device and system and computer readable storage medium | |
| CN112883824A (en) | Finger vein feature recognition device for intelligent blood sampling and recognition method thereof | |
| CN107480678A (en) | A kind of chessboard recognition methods and identifying system | |
| CN111242971B (en) | Target tracking method based on improved double-center particle swarm optimization algorithm | |
| CN110334631B (en) | Sitting posture detection method based on face detection and binary operation | |
| CN114202795A (en) | Method for quickly positioning pupils of old people | |
| CN110348344A (en) | A method of the special facial expression recognition based on two and three dimensions fusion | |
| CN115359249B (en) | Palm image ROI region extraction method and system |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230410 Address after: Room 2309, 23rd Floor, Qidi Building, No. 99, South Tiancheng Road, High Speed Rail New Town, Xiangcheng District, Suzhou City, Jiangsu Province, 215000 - Work Station A029 (cluster registration) Patentee after: Jiangsu Shengdian Century Technology Co.,Ltd. Address before: Room 1602, Qianshan building, building D2, Qingdao International Innovation Park, No. 169, Songling Road, Laoshan District, Qingdao, Shandong 266100 Patentee before: Shandong ShengDian Century Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right |