CN111639557A - Intelligent registration feedback method for finger vein image - Google Patents
Intelligent registration feedback method for finger vein image Download PDFInfo
- Publication number
- CN111639557A CN111639557A CN202010414631.3A CN202010414631A CN111639557A CN 111639557 A CN111639557 A CN 111639557A CN 202010414631 A CN202010414631 A CN 202010414631A CN 111639557 A CN111639557 A CN 111639557A
- Authority
- CN
- China
- Prior art keywords
- image
- vein
- finger vein
- score
- noise
- 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 151
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000001514 detection method Methods 0.000 claims abstract description 11
- 238000003708 edge detection Methods 0.000 claims description 23
- 238000013139 quantization Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 101100129590 Schizosaccharomyces pombe (strain 972 / ATCC 24843) mcp5 gene Proteins 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 3
- 230000000877 morphologic effect Effects 0.000 claims description 3
- 238000011002 quantification Methods 0.000 claims 1
- 238000000605 extraction Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 230000001788 irregular Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001727 in vivo Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06V40/1382—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger
- G06V40/1388—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger using image processing
-
- 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
-
- 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
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- 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
-
- 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
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/758—Involving statistics of pixels or of feature values, e.g. histogram matching
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Human Computer Interaction (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to an intelligent finger vein image registration feedback method, which comprises the following steps: 1) collecting a finger vein registration image; 2) detecting the light leakage degree of the finger vein image to obtain a corresponding score; 3) detecting the vein definition of the finger vein image to obtain a corresponding score; 4) detecting the noise degree of the finger vein image to obtain a corresponding score; 5) detecting image similarity between finger vein registration images to obtain corresponding scores of the finger vein registration images; 6) and feeding back the problems existing in the current finger placement according to the four scores, and prompting the user to correct. According to the invention, four detection methods are set according to problems possibly occurring during actual registration, the problems existing in the current registration are effectively fed back, the real finger vein features are more accurately described, the user experience is improved, and a good foundation is laid for subsequent feature extraction and identification.
Description
Technical Field
The invention belongs to the technical field of finger vein identification and information security, and particularly relates to an intelligent finger vein image registration feedback method.
Background
The finger vein recognition technology is a new biological characteristic recognition technology, and utilizes vein grain images obtained after near infrared light penetrates through fingers to perform identity recognition, so that internal information of a human body is obtained in the process, and the finger vein recognition technology has the advantages of high safety and difficulty in counterfeiting, meanwhile, the finger vein recognition technology cannot be lost, is safe and convenient, and is widely applied to authentication systems and equipment in the field of public safety.
The finger vein recognition technology is a process of extracting features of an obtained finger vein image and comparing the extracted features with a feature template of a registered image, so that the quality of the registered image seriously affects the finger vein recognition performance. In the prior art, a feedback method of image information disclosed in patent No. CN110086999A and a terminal device or an image capturing feedback disclosed in patent No. CN105580051B are used in extracting a registered image, and whether image acquisition is good or bad is judged by the similarity or the score, and the image with good quality is retained.
However, when the finger vein device is used for the first time, because the finger vein belongs to in vivo characteristics, the acquired image is influenced by registration behaviors, and many irregular behaviors may occur in the registration process, such as the situations of putting a biased finger, pressing a finger heavily, dirtying a finger and the like, so that the corresponding characteristics of the acquired finger vein image are incomplete or have deviation, the real finger vein characteristics cannot be accurately described, and the comparison fails when the verification image characteristics have great deviation with the registration image characteristics, thereby causing the reduction of the finger vein identification performance.
Disclosure of Invention
The invention aims to solve the problem that the real finger vein information cannot be accurately expressed because the behavior specification of the registered image is not considered in the prior art, and provides an intelligent finger vein image registration feedback method, which effectively feeds back the non-standard behavior of the registered finger and prepares for the subsequent identification process.
In order to achieve the purpose, the technical scheme provided by the invention is as follows:
the invention relates to an intelligent finger vein image registration feedback method, which comprises the following steps:
1) collecting a finger vein registration image;
2) detecting the light leakage degree of the finger vein image, and calculating a light leakage fraction;
3) detecting the vein definition of the finger vein image, and calculating a vein definition score;
4) detecting the noise degree of the finger vein image, and calculating a noise rate score;
5) detecting image similarity between finger vein registration images, and calculating a similarity score;
6) and feeding back the problems existing in the current finger placement according to the four scores, and prompting the user to correct.
Preferably, the step of detecting the light leakage degree of the finger vein image and calculating the light leakage fraction in the step 2) includes:
2.1) carrying out sobel operator edge detection on the finger vein image to obtain the gray value of each pixel point:
wherein G isx、GyG represents the gray value of the image after the detection of the transverse, longitudinal and transverse and longitudinal combined edges respectively, and A represents the original image;
2.2) setting a threshold T1, comparing the gray value G of the image subjected to the horizontal and vertical combination edge detection with the threshold T1, judging whether the gray value G is an edge point, setting the edge point to be 1 and setting the non-edge point to be 0, and obtaining a binary image I subjected to the edge detection1:
Wherein I1(x, y) denotes second after edge detectionA grey value of the value image;
2.3) obtaining a binary image I1Calculating the number of the maximum continuous black pixel points in each row according to the rows, namely counting the number num1 of all the pixel points in the edge line to obtain the light leakage fraction score1 of the image:
score1=num1/(N×M) (4),
n, M indicates the number of rows and columns of the finger vein image.
Preferably, the step of detecting the vein definition of the finger vein image and calculating the vein definition score in step 3) includes:
3.1) carrying out Gaussian filtering smoothing treatment on the finger vein image;
3.2) carrying out maximum curvature filtering on the smoothed image to obtain image section maps of different angles, and calculating the curvature k (z) of the section maps, wherein the calculation formula of the curvature k (z) of the section maps is as follows:
wherein z is the position of each pixel point in the same section, pf(z) is the gray value of the pixel point;
3.3) intercepting points with the curvature k (z) larger than 0, finding out the point with the maximum curvature as a vein central point, and storing the vein central points at different angles to obtain a vein line binary image;
3.4) carrying out morphological closed operation processing on the vein pattern binary image to obtain a final vein pattern binary image I2;
3.5) statistics of vein line binary image I2In a binary image I1And (3) carrying out fractional quantization on the number num2 of all white pixel points in the margin line according to the vein definition of different image libraries, and calculating the vein definition score 2.
Preferably, the step of detecting the noise degree of the finger vein image and calculating the noise rate score in the step 4) includes:
4.1) carrying out sobel operator edge detection on the finger vein image to obtain the gray value of each pixel point:
wherein G isx、GyG represents the gray value of the image after the detection of the transverse, longitudinal and transverse and longitudinal combined edges respectively, and A is an original image;
4.2) setting the threshold T2, T2<T1, comparing the gray value G of the image after the horizontal and vertical combination edge detection with a threshold T2, judging whether the gray value G is an edge point, setting the edge point to be 1 and setting the non-edge point to be 0, and obtaining a binary image I after the edge detection3:
Wherein I3(x, y) is the gray value of the binary image after edge detection;
4.3) from the binary image I3Edge line of (1), statistics of3And (3) performing fractional quantization on the number num3 of all white pixel points in the edge line according to the noise degrees of the image libraries with different noise rates to obtain a noise rate score 3.
Preferably, the step of detecting image similarity between finger vein registration images and calculating a similarity score in step 5) includes:
5.1) calculating the similarity SSIM of adjacent registered images:
wherein x and y respectively represent two adjacent registration images, mux、μyMean values of x, y, respectively, σx、σyRespectively, the standard deviation of x and y, sigmaxyDenotes the covariance of x and y, c1、c2Is a constant;
5.2) obtaining a similarity score4 of the registered image according to the similarity of the adjacent registered images:
wherein N' is the number of registered images, AiIs the ith registered image.
Preferably, the step of feeding back the user information according to the image score in step 6) includes:
6.1) setting a threshold S1 of image light leakage degree, a threshold S2 of image vein definition, a threshold S3 of image noise degree, and thresholds S4, S5 of registration image similarity;
6.2) comparing the noise rate score with the magnitude of S3, and if the noise rate score is less than S3, prompting the user to check whether the finger is dirty or the mirror surface of the equipment is wiped;
6.3) comparing the light leakage fraction with the magnitude of S1, and if the light leakage fraction is smaller than S1, prompting a user to cover the mirror surface of the equipment as much as possible when placing a finger;
6.4) comparing the vein definition score with the size of S2, and if the vein definition score is smaller than S2, prompting the user to put a finger lightly;
6.5) comparing the similarity score with the sizes of S4 and S5, if the similarity score is smaller than S4, prompting the user to reduce the finger variation range, and if the similarity score is larger than S5, prompting the user to vary the finger posture.
Preferably, the score quantization is performed according to the vein definition of different image libraries in the step 3.5), and the specific step of calculating the vein definition score2 is: finding out the images with the least veins and the most veins in the image library, wherein the number num2 of white pixel points of the images with the least veins and the most veins is a and b respectively, and carrying out normalization processing to obtain the vein definition score 2.
Preferably, in the step 3.5), a score corresponding to num2 ═ a is 30 points, a score corresponding to num2 ═ b is 100 points, and the vein definition score2 is calculated according to the following formula:
preferably, in the step 4.3), performing fractional quantization according to noise degrees of the image library with different noise rates to obtain the noise rate score3 specifically includes: and finding out the image with the least noise and the most noise in the image library, wherein the noise of the image with the least noise and the noise of the image with the most noise are c and d respectively, and performing normalization processing to obtain a noise rate score 3.
Preferably, in the step 4.3), the score corresponding to num3 ═ c is set to 100 points, the score corresponding to num3 ═ d is set to 20 points, and the calculation formula of the noise rate score3 is as follows:
compared with the prior art, the technical scheme provided by the invention has the following beneficial effects:
aiming at the problems possibly occurring when fingers are registered, the method sets four detection indexes, namely the light leakage degree of an image, the definition of an image vein, the image noise rate and the similarity between images, respectively corresponds to the bad operation of placing a biased finger, a heavy finger and a dirty finger and too large or too small finger placement change in the registration process, feeds back user operation information according to four scores, enables the finger vein registration to be more intelligent, can more accurately describe the real finger vein characteristics, and compared with the hard operation that the image which does not meet the conditions cannot be registered, feeds back information to a user, enables the user physical examination to be more friendly in a mode of feeding back information to the user, and lays a good foundation for subsequent characteristic extraction and identification.
Drawings
FIG. 1 is a flow chart of an intelligent finger vein image registration feedback method according to the present invention;
FIG. 2 is a collected finger vein registration image;
FIG. 3 is a binary image after edge detection during light leak detection;
FIG. 4 is an image after Gaussian filtering in the vein sharpness detection process;
FIG. 5 is an image after maximum curvature processing during vein sharpness detection;
FIG. 6 is a binary image after edge detection in a noise detection process;
FIG. 7 is a schematic diagram of an experimental image library in the example.
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 intelligent finger vein image registration feedback method according to the present invention includes the following steps:
1) the finger vein registration image is captured, and an image with a size of M × N is obtained, wherein M, N represents the number of rows and columns of the image, respectively, M × N is 400 × 160 in this embodiment, as shown in fig. 2.
2) Detecting the light leakage degree of the finger vein image, and calculating the light leakage fraction, wherein the method specifically comprises the following steps:
2.1) carrying out sobel operator edge detection on the finger vein image to obtain the gray value of each pixel point:
wherein G isx、GyG represents the gray value of the image subjected to the transverse, longitudinal and transverse and longitudinal combined edge detection respectively, and A is an original image;
2.2) setting a threshold T1, in the embodiment, T1 is 150, comparing the gray value G of the image subjected to the horizontal and vertical combination edge detection with the threshold T1, judging whether the gray value G is an edge point, setting the edge point to be 1 and setting the non-edge point to be 0, and obtaining a binary image I subjected to edge detection1As shown in FIG. 3, a binary image I1The calculation formula of (2) is as follows:
wherein I1(x, y) represents the gray value of the binary image after edge detection;
2.3) obtaining a binary image I1Calculating the number of the maximum continuous black pixel points in each row according to the rows, namely counting the number num1 of all the pixel points in the edge line to obtain the light leakage fraction score1 of the image:
score1=num1/(N×M) (4),
n, M respectively indicate the number of rows and columns of the vein image, which in this embodiment are 160 and 400 respectively.
3) Detecting the vein definition of a finger vein image, and calculating a vein definition score, wherein the method specifically comprises the following steps:
3.1) avoiding the influence of noise on the vein definition, and performing Gaussian filtering smoothing treatment on the finger vein image to obtain an image as shown in FIG. 4;
3.2) because the gray scale of the vein part is low, the smoothed image is subjected to maximum curvature filtering, image section maps of different angles are obtained firstly, and 0 is selected in the embodiment0、300、600、1200、1500、1800The curvature k (z) of the cross-sectional view is calculated for the 6 angles respectively, and the calculation formula of the curvature k (z) is as follows:
wherein z is the position of each pixel point in the same section, pfAnd (z) is the gray value of the pixel point.
3.3) intercepting points with the curvature k (z) larger than 0, finding out the point with the maximum curvature as the vein central point, and storing the vein central points at different angles to obtain a vein line binary image;
3.4) carrying out morphological closed operation processing on the vein pattern binary image to obtain a final vein pattern binary image I2As shown in fig. 5;
3.5) obtaining the binary image I according to the step 2.2)1The edge line of the vein is counted to obtain a binary image I of vein lines2In a binary image I1The number num2 of all white pixel points in the margin line is subjected to fractional quantization according to the vein definition of different image libraries to obtain a vein definition fraction score2, namely the score2
Calculating the vein definition of different image libraries to obtain num2 corresponding to each image, finding out the image with the least veins and the most veins in the image library, wherein the white pixel numbers num2 of the image with the least veins and the image with the most veins are 7000 and 22000 respectively, namely a is 7000 and b is 22000, setting the two numbers to correspond to 30 minutes and 100 minutes respectively, and carrying out normalization processing to obtain a vein definition score 2:
4) the method comprises the following steps of detecting the noise degree of the finger vein image and calculating the noise rate score, and comprises the following specific steps:
4.1) carrying out sobel operator edge detection on the finger vein image to obtain the gray value of each pixel point:
wherein G isx、GyG represents the gray value of the image after the detection of the transverse, longitudinal and transverse and longitudinal combined edges respectively, and A is the original image.
4.2) setting the threshold T2, T2<T1, in this embodiment, T2 is 100, the gray value G of the image after edge detection is compared with the threshold T2 to determine whether it is an edge point, and the edge point is set to 1 and the non-edge point is set to 0, so as to obtain the edge-detected binary image I3As shown in fig. 6:
wherein I3(x, y) is the gray value of the binary image after edge detection;
4.3) from the binary image I3Edge line of (1), statistics of3In the margin line, the number num3 of all white pixels is subjected to fractional quantization according to the noise degree of the image library with different noise rates to obtain a noise rate fraction score3, namely, the noise rate fraction score3
Calculating the noise degrees of different noise rate image libraries to obtain num3 corresponding to each image, finding out the image with the least noise and the most noise in the image library, wherein the number of the white pixel points num3 of the image with the least noise and the image with the most noise is 100 and 500 respectively, namely c is 100, d is 500, setting the two numbers to correspond to 100 minutes and 20 minutes respectively, and performing normalization processing to obtain a noise rate score 3:
5) detecting image similarity between finger vein registration images and calculating a similarity score, and the specific steps comprise:
5.1) calculating the similarity SSIM of adjacent registered images:
wherein x and y respectively represent two adjacent registration images, mux、μyMean values of x, y, respectively, σx、σyRespectively, the standard deviation of x and y, sigmaxyDenotes the covariance of x and y, c1、c2Is a constant;
5.2) obtaining a similarity score4 between the registered images according to the similarity of the adjacent registered images:
wherein N' is the number of registered images, AiIs the ith registered image.
6) Feeding back user information according to the image scores to prompt a user to correct, and the specific steps comprise:
6.1) setting a threshold value S1 of the light leakage degree of the image, a threshold value S2 of the vein definition of the image, a threshold value S3 of the noise degree of the image and threshold values S4 and S5 of the similarity of the registered images, wherein S1, S2, S3, S4 and S5 are respectively 80, 70, 80, 70 and 95;
6.2) since the noise can affect the judgment of other conditions, the noise degree is firstly detected, the sizes of score3 and S3 are judged, if score3 is smaller than S3, the user is prompted to check whether the finger is dirty or the mirror surface of the equipment is wiped, and if not, the next step is executed;
6.3) judging the sizes of score1 and S1, if score1 is smaller than S1, prompting the user to cover the mirror surface of the equipment as much as possible when placing a finger;
6.4) judging the sizes of score2 and S2, and if score2 is smaller than S2, prompting the user to put a finger lightly;
6.5) judging the sizes of score4 and S4 and S5, if score4 is smaller than S4, prompting the user to reduce the finger variation amplitude, and if score4 is larger than S5, prompting the user to change the finger posture.
The following are experimental results and analyses of different image libraries using the method of the invention.
In order to verify the intelligent registration feedback method of the finger vein image, 4 groups of finger vein image databases corresponding to different image scores are established, each group of image databases consists of different registration modes, the first group is an image database (figure 7(1)) acquired by normally placing fingers and placing biased fingers, the second group is an image database (figure 7(2)) acquired by normally placing fingers and pressing fingers lightly, the third group is an image database (figure 7(3)) acquired by normally placing fingers and placing dirty fingers, the fourth group is an image database (figure 7(4)) acquired by placing fingers with too small change, the fifth group is an image database (figure 7(5)) acquired by placing fingers with too large change, and each group is placed 500 times to form an experimental database. MATLAB2016a was used as the compiler software, and the operating system of the computer used was 64-bit Window10, memory 8G, and main frequency 2.30 GHz. For each group of image libraries, the corresponding scores obtained by normal placement and special placement were detected, and the score results are shown in table 1:
TABLE 1 corresponding scores for Normal Placement and Special Placement
The above table shows that the scores obtained by the five groups of specially placed fingers are different from the scores obtained by the normal placement, and therefore, the intelligent finger vein image registration feedback method provided by the invention can effectively indicate the problem of irregular operation in the registration process, and provides a good guarantee for finger vein identification.
The above examples are illustrative of the present invention, but are not to be construed as limiting the scope of the 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. An intelligent finger vein image registration feedback method is characterized in that: which comprises the following steps:
1) collecting a finger vein registration image;
2) detecting the light leakage degree of the finger vein image, and calculating a light leakage fraction;
3) detecting the vein definition of the finger vein image, and calculating a vein definition score;
4) detecting the noise degree of the finger vein image, and calculating a noise rate score;
5) detecting image similarity between finger vein registration images, and calculating a similarity score;
6) and feeding back the problems existing in the current finger placement according to the four scores, and prompting the user to correct.
2. The intelligent finger vein image registration feedback method according to claim 1, characterized in that: the step of detecting the light leakage degree of the finger vein image and calculating the light leakage fraction in the step 2) comprises the following steps:
2.1) carrying out sobel operator edge detection on the finger vein image to obtain the gray value of each pixel point:
wherein G isx、GyG represents the gray value of the image after the detection of the transverse, longitudinal and transverse and longitudinal combined edges respectively, and A represents the original image;
2.2) setting a threshold T1, comparing the gray value G of the image subjected to the horizontal and vertical combination edge detection with the threshold T1, judging whether the gray value G is an edge point, setting the edge point to be 1 and setting the non-edge point to be 0, and obtaining a binary image I subjected to the edge detection1:
Wherein I1(x, y) represents the gray value of the binary image after edge detection;
2.3) obtaining a binary image I1Calculating the number of the maximum continuous black pixel points in each row according to the rows, namely counting the number num1 of all the pixel points in the edge line to obtain the light leakage fraction score1 of the image:
score1=num1/(N×M) (4),
n, M indicates the number of rows and columns of the finger vein image.
3. The intelligent finger vein image registration feedback method according to claim 2, characterized in that: the step of detecting the vein definition of the finger vein image and calculating the vein definition score in the step 3) comprises the following steps:
3.1) carrying out Gaussian filtering smoothing treatment on the finger vein image;
3.2) carrying out maximum curvature filtering on the smoothed image to obtain image section maps of different angles, and calculating the curvature k (z) of the section maps, wherein the calculation formula of the curvature k (z) of the section maps is as follows:
wherein z is the position of each pixel point in the same section, pf(z) is the gray value of the pixel point;
3.3) intercepting points with the curvature k (z) larger than 0, finding out the point with the maximum curvature as a vein central point, and storing the vein central points at different angles to obtain a vein line binary image;
3.4) carrying out morphological closed operation processing on the vein pattern binary image to obtain a final vein pattern binary image I2;
3.5) statistics of vein line binary image I2In a binary image I1And (3) carrying out fractional quantization on the number num2 of all white pixel points in the margin line according to the vein definition of different image libraries, and calculating the vein definition score 2.
4. The intelligent finger vein image registration feedback method according to claim 3, characterized in that: the step of detecting the noise degree of the finger vein image and calculating the noise rate score in the step 4) comprises the following steps:
4.1) carrying out sobel operator edge detection on the finger vein image to obtain the gray value of each pixel point:
wherein G isx、GyG represents the gray value of the image after the detection of the transverse, longitudinal and transverse and longitudinal combined edges respectively, and A is an original image;
4.2) setting the threshold T2, T2<T1, comparing the gray value G of the image after detecting the horizontal and vertical combination edge with the threshold T2, and determining whether it isSetting the edge point as 1 and the non-edge point as 0 to obtain the edge detected binary image I3:
Wherein I3(x, y) is the gray value of the binary image after edge detection;
4.3) from the binary image I3Edge line of (1), statistics of3And (3) performing fractional quantization on the number num3 of all white pixel points in the edge line according to the noise degrees of the image libraries with different noise rates to obtain a noise rate score 3.
5. The intelligent finger vein image registration feedback method according to claim 1, characterized in that: the step of detecting image similarity between finger vein registration images and calculating a similarity score in the step 5) comprises the following steps:
5.1) calculating the similarity SSIM of adjacent registered images:
wherein x and y respectively represent two adjacent registration images, mux、μyMean values of x, y, respectively, σx、σyRespectively, the standard deviation of x and y, sigmaxyDenotes the covariance of x and y, c1、c2Is a constant;
5.2) obtaining a similarity score4 of the registered image according to the similarity of the adjacent registered images:
wherein N' is the number of registered images, AiIs the ith registered image.
6. The intelligent finger vein image registration feedback method according to claim 1, characterized in that: the step of feeding back the user information according to the image score in the step 6) comprises:
6.1) setting a threshold S1 of image light leakage degree, a threshold S2 of image vein definition, a threshold S3 of image noise degree, and thresholds S4, S5 of registration image similarity;
6.2) comparing the noise rate score with the magnitude of S3, and if the noise rate score is less than S3, prompting the user to check whether the finger is dirty or the mirror surface of the equipment is wiped;
6.3) comparing the light leakage fraction with the magnitude of S1, and if the light leakage fraction is smaller than S1, prompting a user to cover the mirror surface of the equipment as much as possible when placing a finger;
6.4) comparing the vein definition score with the size of S2, and if the vein definition score is smaller than S2, prompting the user to put a finger lightly;
6.5) comparing the similarity score with the sizes of S4 and S5, if the similarity score is smaller than S4, prompting the user to reduce the finger variation range, and if the similarity score is larger than S5, prompting the user to vary the finger posture.
7. The intelligent finger vein image registration feedback method according to claim 3, characterized in that: the step 3.5) of performing score quantification according to the vein definition of different image libraries comprises the specific steps of calculating a vein definition score 2: finding out the images with the least veins and the most veins in the image library, wherein the number num2 of white pixel points of the images with the least veins and the most veins is a and b respectively, and carrying out normalization processing to obtain the vein definition score 2.
8. The intelligent finger vein image registration feedback method according to claim 7, wherein: in the step 3.5), the score corresponding to num2 ═ a is set to be 30 points, the score corresponding to num2 ═ b is set to be 100 points, and the calculation formula of the vein definition score2 is as follows:
9. the intelligent finger vein image registration feedback method according to claim 4, wherein: in the step 4.3), performing fractional quantization according to the noise degrees of the image library with different noise rates to obtain a noise rate score3 specifically includes: and finding out the image with the least noise and the most noise in the image library, wherein the noise of the image with the least noise and the noise of the image with the most noise are c and d respectively, and performing normalization processing to obtain a noise rate score 3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010414631.3A CN111639557B (en) | 2020-05-15 | 2020-05-15 | Intelligent registration feedback method for finger vein image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010414631.3A CN111639557B (en) | 2020-05-15 | 2020-05-15 | Intelligent registration feedback method for finger vein image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111639557A true CN111639557A (en) | 2020-09-08 |
CN111639557B CN111639557B (en) | 2023-06-20 |
Family
ID=72330847
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010414631.3A Active CN111639557B (en) | 2020-05-15 | 2020-05-15 | Intelligent registration feedback method for finger vein image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111639557B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112200159A (en) * | 2020-12-01 | 2021-01-08 | 四川圣点世纪科技有限公司 | Non-contact palm vein identification method based on improved residual error network |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2006121631A2 (en) * | 2005-04-26 | 2006-11-16 | Novadaq Technologies, Inc. | Real time imaging during solid organ transplant |
CN101441711A (en) * | 2008-11-18 | 2009-05-27 | 华南理工大学 | Identification authentication system based on finger print and hand back vein |
US20110304720A1 (en) * | 2010-06-10 | 2011-12-15 | The Hong Kong Polytechnic University | Method and apparatus for personal identification using finger imaging |
CN105580051A (en) * | 2013-10-30 | 2016-05-11 | 英特尔公司 | Image capture feedback |
CN105760841A (en) * | 2016-02-22 | 2016-07-13 | 桂林航天工业学院 | Identify recognition method and identify recognition system |
CN107862282A (en) * | 2017-11-07 | 2018-03-30 | 深圳市金城保密技术有限公司 | A kind of finger vena identification and safety certifying method and its terminal and system |
CN108960068A (en) * | 2018-06-05 | 2018-12-07 | 天津大学 | For acquiring the light source brightness adjusting device and method of finger venous image |
CN109784174A (en) * | 2018-12-14 | 2019-05-21 | 深圳壹账通智能科技有限公司 | A kind of login method and equipment of user account |
CN109815869A (en) * | 2019-01-16 | 2019-05-28 | 浙江理工大学 | A kind of finger vein identification method based on the full convolutional network of FCN |
CN110086999A (en) * | 2019-05-28 | 2019-08-02 | 维沃移动通信有限公司 | The feedback method and terminal device of image information |
CN110119724A (en) * | 2019-05-16 | 2019-08-13 | 天津科技大学 | A kind of finger vein identification method |
CN110599606A (en) * | 2019-07-30 | 2019-12-20 | 广东工业大学 | Subway ticketing system based on finger vein recognition |
CN110751029A (en) * | 2019-09-12 | 2020-02-04 | 南京邮电大学 | Maximum curvature-based adaptive finger vein line extraction method |
CN110781708A (en) * | 2018-07-31 | 2020-02-11 | 贾志恒 | Finger vein image recognition system and finger vein image recognition method based on acquisition equipment |
-
2020
- 2020-05-15 CN CN202010414631.3A patent/CN111639557B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2006121631A2 (en) * | 2005-04-26 | 2006-11-16 | Novadaq Technologies, Inc. | Real time imaging during solid organ transplant |
CN101441711A (en) * | 2008-11-18 | 2009-05-27 | 华南理工大学 | Identification authentication system based on finger print and hand back vein |
US20110304720A1 (en) * | 2010-06-10 | 2011-12-15 | The Hong Kong Polytechnic University | Method and apparatus for personal identification using finger imaging |
CN105580051A (en) * | 2013-10-30 | 2016-05-11 | 英特尔公司 | Image capture feedback |
CN105760841A (en) * | 2016-02-22 | 2016-07-13 | 桂林航天工业学院 | Identify recognition method and identify recognition system |
CN107862282A (en) * | 2017-11-07 | 2018-03-30 | 深圳市金城保密技术有限公司 | A kind of finger vena identification and safety certifying method and its terminal and system |
CN108960068A (en) * | 2018-06-05 | 2018-12-07 | 天津大学 | For acquiring the light source brightness adjusting device and method of finger venous image |
CN110781708A (en) * | 2018-07-31 | 2020-02-11 | 贾志恒 | Finger vein image recognition system and finger vein image recognition method based on acquisition equipment |
CN109784174A (en) * | 2018-12-14 | 2019-05-21 | 深圳壹账通智能科技有限公司 | A kind of login method and equipment of user account |
CN109815869A (en) * | 2019-01-16 | 2019-05-28 | 浙江理工大学 | A kind of finger vein identification method based on the full convolutional network of FCN |
CN110119724A (en) * | 2019-05-16 | 2019-08-13 | 天津科技大学 | A kind of finger vein identification method |
CN110086999A (en) * | 2019-05-28 | 2019-08-02 | 维沃移动通信有限公司 | The feedback method and terminal device of image information |
CN110599606A (en) * | 2019-07-30 | 2019-12-20 | 广东工业大学 | Subway ticketing system based on finger vein recognition |
CN110751029A (en) * | 2019-09-12 | 2020-02-04 | 南京邮电大学 | Maximum curvature-based adaptive finger vein line extraction method |
Non-Patent Citations (11)
Title |
---|
WADMAN, MC等: "Assessment of a New Model for Femoral Ultrasound-guided Central Venous Access Procedural Training: A Pilot Study", ACADEMIC EMERGENCY MEDICINE, vol. 17, no. 1, pages 88 - 92 * |
宋丹;黄旭;: "生物识别技术及其在金融支付安全领域的应用", 信息安全研究, no. 01, pages 27 - 32 * |
崔建江等: "基于近红外图像质量评价的静脉图像采集***", 《东北大学学报(自然科学版)》 * |
崔建江等: "基于近红外图像质量评价的静脉图像采集***", 《东北大学学报(自然科学版)》, no. 08, 15 August 2009 (2009-08-15), pages 1009 - 1102 * |
王科俊;刘靖宇;马慧;李雪峰;: "手指静脉图像质量评价", 智能***学报, vol. 6, no. 04, pages 324 - 327 * |
谭营;王军;: "手指静脉身份识别技术最新进展", 智能***学报, vol. 6, no. 6, pages 471 - 482 * |
陆永刚等: "智能型手指静脉图像采集与控制***的研究", 《电子测量与仪器学报》 * |
陆永刚等: "智能型手指静脉图像采集与控制***的研究", 《电子测量与仪器学报》, no. 02, 15 February 2010 (2010-02-15), pages 184 - 189 * |
黄庆南: "静脉识别在门禁***中的应用", 《微计算机信息》 * |
黄庆南: "静脉识别在门禁***中的应用", 《微计算机信息》, no. 01, 10 January 2007 (2007-01-10), pages 289 - 291 * |
黄庆南;: "静脉识别在门禁***中的应用", 微计算机信息, no. 01, pages 289 - 291 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112200159A (en) * | 2020-12-01 | 2021-01-08 | 四川圣点世纪科技有限公司 | Non-contact palm vein identification method based on improved residual error network |
Also Published As
Publication number | Publication date |
---|---|
CN111639557B (en) | 2023-06-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Choi et al. | Automatic segmentation of latent fingerprints | |
US7072523B2 (en) | System and method for fingerprint image enhancement using partitioned least-squared filters | |
US20080013803A1 (en) | Method and apparatus for determining print image quality | |
Zanganeh et al. | Partial fingerprint matching through region-based similarity | |
CN110555382A (en) | Finger vein identification method based on deep learning and Wasserstein distance measurement | |
Short et al. | Latent fingerprint segmentation using ridge template correlation | |
Yang et al. | Superpixel based finger vein ROI extraction with sensor interoperability | |
US11475707B2 (en) | Method for extracting image of face detection and device thereof | |
WO2009029638A1 (en) | Iris recognition | |
CN109409181B (en) | Independent detection method for upper and lower edges of fingers for low-quality finger vein image | |
CN110458792B (en) | Method and device for evaluating quality of face image | |
CN111639559A (en) | Finger vein machine learning identification method and device based on topographic concave-convex characteristics | |
CN111639560B (en) | Finger vein feature extraction method and device based on dynamic fusion of vein skeleton line and topographic concave-convex characteristics | |
Kim et al. | New enhancement algorithm for fingerprint images | |
CN115223211B (en) | Identification method for converting vein image into fingerprint image | |
CN111639557B (en) | Intelligent registration feedback method for finger vein image | |
CN114943744A (en) | Edge detection method based on local Otsu thresholding | |
Krichen et al. | Specific texture analysis for iris recognition | |
CN112085683B (en) | Depth map credibility detection method in saliency detection | |
Saparudin et al. | Segmentation of fingerprint image based on gradient magnitude and coherence | |
CN111382703B (en) | Finger vein recognition method based on secondary screening and score fusion | |
CN110502996B (en) | Dynamic identification method for fuzzy finger vein image | |
Jing et al. | A novel method for shoeprints recognition and classification | |
CN115578760B (en) | Control system and control method based on topographic relief vein recognition | |
CN113408416B (en) | Fingerprint frequency estimation method and device and fingerprint information extraction method and device |
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 | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: An intelligent registration feedback method for finger vein images Granted publication date: 20230620 Pledgee: Bank of China Limited Taiyuan Binzhou sub branch Pledgor: Holy Point Century Technology Co.,Ltd. Registration number: Y2024140000011 |