CN114882539B - Vein image ROI extraction method and device - Google Patents
Vein image ROI extraction method and device Download PDFInfo
- Publication number
- CN114882539B CN114882539B CN202210807554.7A CN202210807554A CN114882539B CN 114882539 B CN114882539 B CN 114882539B CN 202210807554 A CN202210807554 A CN 202210807554A CN 114882539 B CN114882539 B CN 114882539B
- Authority
- CN
- China
- Prior art keywords
- finger
- edge
- confidence
- vein image
- straight line
- 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
- G06V40/1353—Extracting features related to minutiae or pores
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
A vein image ROI extraction method and a vein image ROI extraction device are provided, wherein the extraction method comprises the following steps: collecting a finger vein image of a finger of a user; acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of the finger vein image by adopting an edge detection algorithm; performing straight line detection on the middle areas of one-third and two-thirds rows of the finger vein image to obtain a first calibration straight line and a second calibration straight line; correcting the first confidence upper finger contour edge and the first confidence lower finger contour edge based on the first calibration straight line and the second calibration straight line respectively to obtain a second confidence upper finger contour edge and a second confidence lower finger contour edge; and respectively obtaining an internal tangent of the edge of the finger contour on the second confidence and an internal tangent of the edge of the finger contour under the second confidence, and performing ROI extraction on the finger vein image based on the internal tangent. By the method, the problem that the acquisition of the finger contour edge of the low-quality image is inaccurate can be effectively solved, so that the extraction of the finger vein image ROI is more effective and accurate.
Description
Technical Field
The invention belongs to a biological identification technology in the field of information security, and particularly relates to a vein image ROI extraction method and device.
Background
The traditional authentication mode based on passwords, passwords and tokens can not meet the industry requirements gradually due to the fact that the traditional authentication mode is easy to forget and not easy to carry. The vein recognition technology is used as a natural in-vivo characteristic living body recognition technology, the characteristic that deoxyhemoglobin in venous blood absorbs specific near infrared rays is utilized, fingers, palms or the backs of the hands and the like are irradiated by the near infrared rays, the vein distribution diagram in the fingers, the palms or the backs of the hands is shot by using infrared cameras with corresponding wavelengths, and counterfeiting and tampering are extremely difficult.
In the process of finger vein identity authentication, the collected finger vein image can seriously affect the result of finger vein identification due to the deviation of a user finger or the existence of noise in the area around the finger, so in the process of finger vein identity authentication, the ROI of the finger vein image is generally extracted first. The existing ROI extraction method generally obtains a finger contour edge based on edge detection, but the image quality of a finger vein is easily influenced by ambient light, finger thickness, user behavior and the like to generate phenomena of inclination, light spots and blurring, and the low-quality image cannot obtain a complete and correct finger contour edge by directly utilizing the edge detection.
Disclosure of Invention
The invention aims to provide a vein image ROI extraction method and device, which can completely and accurately acquire the edge of a finger contour so as to accurately extract the ROI of a vein image. In order to achieve the purpose, the technical scheme provided by the invention is as follows:
a vein image ROI extraction method comprises the following steps: collecting a finger vein image of a finger of a user; acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of the finger vein image by adopting an edge detection algorithm; performing straight line detection on the middle areas of one-third and two-thirds rows of the finger vein image to obtain a first calibration straight line and a second calibration straight line; correcting the first confidence upper finger contour edge and the first confidence lower finger contour edge based on the first calibration straight line and the second calibration straight line respectively to obtain a second confidence upper finger contour edge and a second confidence lower finger contour edge; and respectively obtaining internal tangent lines of the edge of the finger contour under the second confidence and the edge of the finger contour under the second confidence, and performing ROI extraction on the finger vein image based on the internal tangent lines.
Preferably, the performing line detection on the middle area of one-third and two-thirds of the finger vein image to obtain a first calibration line and a second calibration line includes: and performing straight line detection on the middle areas of one-third columns and two-thirds columns of the finger vein image to obtain two longest straight lines which are respectively used as a first calibration straight line and a second calibration straight line.
Preferably, the correcting the first confidence upper finger contour edge and the first confidence lower finger contour edge based on the first calibration straight line and the second calibration straight line respectively to obtain a second confidence upper finger contour edge and a second confidence lower finger contour edge includes: acquiring a first pseudo edge point on the contour edge of the first confidence upper finger, and correcting the first pseudo edge point based on the first calibration straight line; and acquiring a second pseudo edge point on the edge of the finger contour under the first confidence, and correcting the second pseudo edge point based on the second correction line.
Preferably, the acquiring a first pseudo-edge point on the edge of the first confidence upper finger contour includes: traversing pixel points on the contour edge of the first confidence upper finger, sequentially calculating the difference value of the lines of two adjacent pixel points, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to the first calibration straight line, wherein the pixel point with the larger distance to the first calibration straight line is a first pseudo edge point; the obtaining a second pseudo edge point on the edge of the finger contour under the first confidence comprises: traversing pixel points on the contour edge of the finger under the first confidence, sequentially calculating the difference value of the lines where two adjacent pixel points are located, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to the second calibration straight line, wherein the pixel point with the larger distance to the second calibration straight line is a second pseudo edge point.
Preferably, the correcting the first pseudo edge point based on the first calibration straight line includes: substituting the column coordinates of the first pseudo edge points into the first calibration straight line to calculate corrected row coordinates; the correcting the second pseudo edge point based on the second correction line includes: and substituting the column coordinates of the second pseudo edge points into the second calibration straight line to calculate corrected row coordinates.
Preferably, the acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of the finger vein image by using an edge detection algorithm includes: performing convolution on the finger vein image by adopting an edge detection operator to obtain a gradient map of the finger vein image; dividing the gradient map into an upper gradient map and a lower gradient map with the geometric center of the gradient map; traversing pixel points of the upper gradient map from bottom to top, wherein the pixel point of each row with the first gradient value larger than or equal to the second threshold value is the edge point of the outline of the first confidence upper finger in the row, and if the gradient value of each pixel point in the row is smaller than the second threshold value, the pixel point with the maximum gradient value in the row is taken as the edge point of the outline of the first confidence upper finger; and traversing the pixel points of the lower gradient map from top to bottom, wherein the pixel point of which the first gradient value is greater than or equal to the second threshold value in each row is the edge point of the finger contour under the first confidence in the row, and if the gradient value of each pixel point in one row is smaller than the second threshold value, the pixel point of which the gradient value is the largest in the row is taken as the edge point of the finger contour under the first confidence.
Preferably, before the obtaining the internal tangents of the second confidence upper finger-contour edge and the second confidence lower finger-contour edge respectively, further comprises: fitting a finger central line according to the second confidence upper finger contour edge and the second confidence lower finger contour edge, and performing rotation correction on the finger vein image based on an included angle between the finger central line and a horizontal axis, wherein the fitting of the finger central line comprises the steps of respectively obtaining the middle points of each row of the second confidence upper finger contour edge and the second confidence lower finger contour edge, and then fitting the middle points into the finger central line.
Preferably, the performing ROI extraction on the finger vein image based on the inscribed line includes: constructing an extraction window by taking the distance of the internal tangent as a width; sliding the extraction window on the finger vein image along a horizontal axis, and calculating the quality score of the finger vein image in the extraction window; and extracting the region corresponding to the window with the highest quality score as the ROI of the finger vein image.
Preferably, the quality score is a weighted value of the sharpness of the finger vein image and the gray level of the finger vein image in the extraction window, and the gray level of the finger vein image is the gray level of the left and right edge blocks of the finger vein image in the extraction window.
In order to solve the above technical problem, the present application further provides a vein image ROI extraction device, including: the acquisition module is used for acquiring finger vein images of fingers of a user; the edge detection module is used for acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of the finger vein image; the calibration straight line acquisition module is used for performing straight line detection on the middle areas of one-third and two-thirds rows of the finger vein image to obtain a first calibration straight line and a second calibration straight line; the calibration module is used for respectively calibrating the first upper confidence finger contour edge and the first lower confidence finger contour edge based on the first calibration straight line and the second calibration straight line to obtain a second upper confidence finger contour edge and a second lower confidence finger contour edge; and the ROI extraction module is used for respectively acquiring internal tangent lines of the edge of the finger contour under the second confidence and extracting the ROI of the finger vein image based on the internal tangent lines.
By adopting the technical scheme provided by the application, the first confidence upper finger contour edge and the first confidence lower finger contour edge are corrected based on the first calibration straight line and the second calibration straight line respectively to obtain the second confidence upper finger contour edge and the second confidence lower finger contour edge, so that the problem of inaccurate acquisition of the low-quality image finger contour edge can be effectively solved, and the extraction of the finger vein image ROI is more effective and accurate.
Drawings
In order to more clearly illustrate the embodiments of the present application, the drawings needed for the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings can be obtained by those skilled in the art without inventive effort.
Fig. 1 is a flowchart of a vein image ROI extraction method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a vein image ROI extraction device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without any creative effort belong to the protection scope of the present application.
Referring to fig. 1, the invention relates to a vein image ROI extraction method, which includes the following steps:
1) collecting a finger vein image of a finger of a user through a finger vein collecting device; it should be noted that, the connecting line between the finger tip and the finger root of the finger vein image collected in the present application is in the horizontal axis direction, and the following scheme is also a processing scheme for the connecting line between the finger tip and the finger root of the finger vein image in the horizontal axis direction; if the connecting line of the finger tip and the finger root of the collected finger vein image is on the vertical axis, the finger vein image can be rotated to the horizontal axis.
2) Acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of a finger vein image by adopting an edge detection algorithm, wherein the method comprises the following steps:
2.1) carrying out convolution on the finger vein image by adopting an edge detection operator to obtain a gradient image of the finger vein image;
2.2) dividing the gradient map into an upper gradient map and a lower gradient map by the geometric center of the gradient map;
2.3) traversing pixel points of the upper gradient map from bottom to top, wherein the pixel point of each row with the first gradient value larger than or equal to the second threshold value is the edge point of the outline of the first confidence upper finger in the row, and if the gradient value of each pixel point in the row is smaller than the second threshold value, the pixel point with the maximum gradient value in the row is taken as the edge point of the outline of the first confidence upper finger;
and 2.4) traversing pixel points of the lower gradient map from top to bottom, wherein the pixel point of which the first gradient value is greater than or equal to the second threshold value in each row is the edge point of the finger contour under the first confidence in the row, and if the gradient value of each pixel point in each row is smaller than the second threshold value, the pixel point with the maximum gradient value in the row is taken as the edge point of the finger contour under the first confidence.
It should be noted that the edge detection operator may be a commonly used edge detection operator such as Sobel, Roberts, Canny, and the second threshold is a preset segmentation threshold. The prior art adopts an edge detection algorithm to obtain a finger contour edge of a finger vein image, and generally comprises the following steps: after the gradient image of the finger vein image is obtained, the threshold value is directly used for judging or searching the maximum gradient value as the edge point of the finger contour. According to the method and the device, the gradient map is divided into the upper gradient map and the lower gradient map, the upper gradient map is traversed from bottom to top, the lower gradient map is traversed from top to bottom to find the pixel point of the first gradient value larger than or equal to the second threshold value, and the situation that the contour edge point is judged wrongly due to the fact that an abnormal object exists at the top or the bottom in the collected finger vein image can be effectively avoided.
3) The method for detecting the straight line of the middle area of one-third column and two-thirds column of the finger vein image to obtain a first calibration straight line and a second calibration straight line comprises the following steps: and performing straight line detection on the middle areas of one-third columns and two-thirds columns of the finger vein image to obtain two longest straight lines which are respectively used as a first calibration straight line and a second calibration straight line.
The conventional line detection method may be a hough line detection method, an lsd line detection method, an fld line detection method, an lsm line detection method, or the like. Because the collected finger vein image may have light spots, exposure, light leakage and the like around the finger, the obtained finger contour edge is inaccurate, and therefore the finger vein image needs to be corrected. Meanwhile, the image quality of the general middle area of the collected finger vein image is stable, and the contour edge obtained through the middle area is relatively accurate, so that the middle areas of one-third and two-thirds rows of the finger vein image are firstly intercepted, and then the straight line detection is carried out on the middle areas, wherein two longest straight lines are respectively close to the contour edge of the upper hand gesture and the contour edge of the lower finger, and are respectively used as a first calibration straight line and a second calibration straight line.
4) Correcting the first confidence upper finger contour edge and the first confidence lower finger contour edge based on the first calibration straight line and the second calibration straight line respectively to obtain a second confidence upper finger contour edge and a second confidence lower finger contour edge, comprising:
4.1) obtaining a first pseudo-edge point on the edge of the finger contour on the first confidence, comprising: traversing pixel points on the contour edge of the first confidence upper finger, sequentially calculating the difference value of the lines of two adjacent pixel points, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to a first calibration straight line, wherein the pixel point with the larger distance to the first calibration straight line is a first pseudo edge point; acquiring a second pseudo edge point on the edge of the finger contour under the first confidence, wherein the method comprises the following steps: and traversing pixel points on the contour edge of the finger under the first confidence, sequentially calculating the difference value of the lines of two adjacent pixel points, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to a second calibration straight line, wherein the pixel point with the larger distance to the second calibration straight line is a second pseudo-edge point.
It should be noted that the edge of the normal finger contour should be approximately a straight line, the difference between the rows of the pixels in the adjacent columns should be very small, a first threshold is preset, and if the difference between the rows of the pixels in the adjacent columns is greater than the preset first threshold, it can be determined that a false edge point exists. And simultaneously, respectively calculating the distances from the two pixel points to the calibration straight line, wherein the pixel point with the larger distance is the pseudo edge point.
4.2) correcting the first pseudo edge point based on the first correction alignment line, comprising: substituting the column coordinates of the first pseudo edge points into a first calibration straight line to calculate and obtain corrected row coordinates; correcting the second pseudo edge point based on the second correction line, comprising: and substituting the column coordinates of the second pseudo edge points into a second calibration straight line to calculate and obtain corrected row coordinates.
5) Rotationally correcting the finger vein image, comprising: and fitting a finger central line according to the second confidence upper finger contour edge and the second confidence lower finger contour edge, and performing rotation correction on the finger vein image based on an included angle between the finger central line and a horizontal axis, wherein the fitting of the finger central line comprises the steps of respectively obtaining the middle point of each column of the second confidence upper finger contour edge and the second confidence lower finger contour edge, and then fitting the middle point into the finger central line.
The fitting method includes a least square method, a gradient descent method, and the like, and is not limited herein. Through rotating and correcting the finger vein image, the ROI in the later period can be conveniently intercepted.
6) ROI interception comprising:
6.1) respectively obtaining internal tangents of the edge of the finger profile at the second confidence upper position and the edge of the finger profile at the second confidence lower position;
6.2) constructing an extraction window by taking the distance h between the two internal tangents as the height, and presetting the width of the extraction window as w;
6.3) sliding the extraction window on the finger vein image along a horizontal axis, and calculating the mass fraction G of the finger vein image in the extraction window, wherein the mass fraction G is the weighted value of the definition D of the finger vein image in the extraction window and the gray S of the left and right edge blocks of the finger vein image in the extraction window, and the calculation formula is as follows:
in the formula, G is the quality fraction of the finger vein image in the extraction window, D is the definition of the finger vein image in the extraction window, S is the gray scale of the left and right edge blocks of the finger vein image in the extraction window, a is the weighted proportion of the definition of the finger vein image in the extraction window, b is the weighted proportion of the gray scale of the left and right edge blocks of the finger vein image in the extraction window,、respectively the columns where the left and right edges of the extraction window are located,is the image gray value and x is the width of the edge block.
And 6.4) extracting the region corresponding to the window with the highest quality score as the ROI of the finger vein image.
It should be noted that, common methods for calculating the sharpness of the graph include Laplacian gradient function, brenner gradient function, and the like, which are not limited herein. Generally, the finger vein image acquired is brighter in finger joint position, and the finger vein ROI interception also needs to include the positions of two finger joints, so that the gray values of the left and right edge blocks of the finger vein image in the extraction window are calculated instead of the gray values of the finger vein image in the extraction window, and the calculation speed can be effectively improved.
Based on the above method for extracting ROI from vein image provided in the embodiment of the present application, the present application further provides a corresponding device for extracting ROI from vein image, where the device may be a module, a program segment, or a code on an electronic device. It should be understood that the apparatus corresponds to the above method embodiment, and can perform the steps related to the above method embodiment, the specific functions of the apparatus can be referred to the above description, and the detailed description is appropriately omitted here to avoid redundancy. As shown in fig. 2, a vein image ROI extraction device includes: the acquisition module is used for acquiring finger vein images of fingers of a user; the edge detection module is used for acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of the finger vein image; the calibration straight line acquisition module is used for carrying out straight line detection on the middle areas of one-third and two-thirds rows of the finger vein image to obtain a first calibration straight line and a second calibration straight line; the calibration module is used for respectively calibrating the first confidence upper finger contour edge and the first confidence lower finger contour edge based on the first calibration straight line and the second calibration straight line to obtain a second confidence upper finger contour edge and a second confidence lower finger contour edge; and the ROI extraction module is used for respectively obtaining the internal tangent of the edge of the finger contour on the second confidence and the internal tangent of the edge of the finger contour under the second confidence and extracting the ROI from the finger vein image based on the internal tangent. It should be noted that each functional module in this embodiment may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
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 (6)
1. A vein image ROI extraction method is characterized by comprising the following steps: collecting a finger vein image of a finger of a user; acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of the finger vein image by adopting an edge detection algorithm; performing straight line detection on the middle areas of one-third and two-thirds rows of the finger vein image to obtain two longest straight lines which are respectively used as a first calibration straight line and a second calibration straight line; correcting the first confidence upper finger contour edge and the first confidence lower finger contour edge based on the first calibration straight line and the second calibration straight line respectively to obtain a second confidence upper finger contour edge and a second confidence lower finger contour edge, including: (1) obtaining a first pseudo-edge point on the first confidence upper finger contour edge, comprising: traversing pixel points on the contour edge of the first confidence upper finger, sequentially calculating the difference value of the lines of two adjacent pixel points, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to the first calibration straight line, wherein the pixel point with the larger distance to the first calibration straight line is a first pseudo edge point; acquiring a second pseudo edge point on the edge of the finger contour under the first confidence, wherein the method comprises the following steps: traversing pixel points on the contour edge of the finger under the first confidence, sequentially calculating the difference value of the lines where two adjacent pixel points are located, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to the second calibration straight line, wherein the pixel point with the larger distance to the second calibration straight line is a second pseudo edge point; (2) correcting the first pseudo edge point based on the first calibration straight line, including: substituting the column coordinates of the first pseudo edge points into the first calibration straight line to calculate corrected row coordinates; correcting the second pseudo edge point based on the second correction line, including: substituting the column coordinates of the second pseudo edge points into the second calibration straight line to calculate corrected row coordinates; and respectively obtaining internal tangent lines of the edge of the finger contour under the second confidence and the edge of the finger contour under the second confidence, and performing ROI extraction on the finger vein image based on the internal tangent lines.
2. The vein image ROI extraction method according to claim 1, wherein: the method for acquiring the first confidence upper finger contour edge and the first confidence lower finger contour edge of the finger vein image by adopting the edge detection algorithm comprises the following steps: performing convolution on the finger vein image by adopting an edge detection operator to obtain a gradient map of the finger vein image; dividing the gradient map into an upper gradient map and a lower gradient map with the geometric center of the gradient map; traversing pixel points of the upper gradient map from bottom to top, wherein the pixel point of each row with the first gradient value larger than or equal to the second threshold value is the edge point of the outline of the first confidence upper finger in the row, and if the gradient value of each pixel point in the row is smaller than the second threshold value, the pixel point with the maximum gradient value in the row is taken as the edge point of the outline of the first confidence upper finger; and traversing the pixel points of the lower gradient map from top to bottom, wherein the pixel point of which the first gradient value is greater than or equal to the second threshold value in each row is the edge point of the finger contour under the first confidence in the row, and if the gradient value of each pixel point in one row is smaller than the second threshold value, the pixel point of which the gradient value is the largest in the row is taken as the edge point of the finger contour under the first confidence.
3. The vein image ROI extraction method according to claim 1, wherein: before the obtaining the internal tangents of the second confidence upper finger contour edge and the second confidence lower finger contour edge respectively, further comprising: fitting a finger central line according to the second confidence upper finger contour edge and the second confidence lower finger contour edge, and performing rotation correction on the finger vein image based on an included angle between the finger central line and a horizontal axis, wherein the fitting of the finger central line comprises the steps of respectively obtaining the middle points of each row of the second confidence upper finger contour edge and the second confidence lower finger contour edge, and then fitting the middle points into the finger central line.
4. The vein image ROI extraction method according to claim 1, wherein: the ROI extraction of the finger vein image based on the internal tangent line comprises the following steps: constructing an extraction window by taking the distance of the internal tangent as high; sliding the extraction window on the finger vein image along a horizontal axis, and calculating the quality score of the finger vein image in the extraction window; and extracting the region corresponding to the window with the highest quality score as the ROI of the finger vein image.
5. The vein image ROI extraction method according to claim 4, wherein: the quality score is a weighted value of the definition of the finger vein image and the gray level of the finger vein image in the extraction window, and the gray level of the finger vein image is the gray level of the left edge block and the right edge block of the finger vein image in the extraction window.
6. A vein image ROI extraction device, comprising: the acquisition module is used for acquiring finger vein images of fingers of a user; the edge detection module is used for acquiring a first confidence upper finger contour edge and a first confidence lower finger contour edge of the finger vein image; the calibration straight line acquisition module is used for carrying out straight line detection on the middle areas of one-third and two-thirds rows of the finger vein image to obtain two longest straight lines which are respectively used as a first calibration straight line and a second calibration straight line; a calibration module, configured to correct the first confidence upper finger contour edge and the first confidence lower finger contour edge based on the first calibration straight line and the second calibration straight line, respectively, to obtain a second confidence upper finger contour edge and a second confidence lower finger contour edge, including: (1) obtaining a first pseudo-edge point on the first confidence upper finger contour edge, comprising: traversing pixel points on the contour edge of the first confidence upper finger, sequentially calculating the difference value of the lines of two adjacent pixel points, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to the first calibration straight line, wherein the pixel point with the larger distance to the first calibration straight line is a first pseudo edge point; acquiring a second pseudo edge point on the edge of the finger contour under the first confidence, wherein the method comprises the following steps: traversing pixel points on the contour edge of the finger under the first confidence, sequentially calculating the difference value of the lines of two adjacent pixel points, if the difference value is greater than a first threshold value, respectively calculating the distance from the two pixel points to the second calibration straight line, wherein the pixel point with the larger distance to the second calibration straight line is a second pseudo-edge point; (2) correcting the first pseudo edge point based on the first calibration straight line, including: substituting the column coordinates of the first pseudo edge points into the first calibration straight line to calculate corrected row coordinates; correcting the second pseudo edge point based on the second correction line, including: substituting the column coordinates of the second pseudo edge points into the second calibration straight line to calculate corrected row coordinates; and the ROI extraction module is used for respectively acquiring internal tangent lines of the edge of the finger contour under the second confidence and extracting the ROI of the finger vein image based on the internal tangent lines.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210807554.7A CN114882539B (en) | 2022-07-11 | 2022-07-11 | Vein image ROI extraction method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210807554.7A CN114882539B (en) | 2022-07-11 | 2022-07-11 | Vein image ROI extraction method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114882539A CN114882539A (en) | 2022-08-09 |
CN114882539B true CN114882539B (en) | 2022-09-30 |
Family
ID=82683587
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210807554.7A Active CN114882539B (en) | 2022-07-11 | 2022-07-11 | Vein image ROI extraction method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114882539B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116884048B (en) * | 2023-09-08 | 2023-12-12 | 江苏圣点世纪科技有限公司 | Abnormal vein image detection method based on edge morphology |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN110751059A (en) * | 2019-09-27 | 2020-02-04 | 五邑大学 | Least square method-based finger vein ROI extraction method, device and storage medium |
CN111639556A (en) * | 2020-05-15 | 2020-09-08 | 圣点世纪科技股份有限公司 | Finger axis rotation finger vein image correction method based on non-uniform interpolation |
CN114529950A (en) * | 2020-10-30 | 2022-05-24 | 北京眼神智能科技有限公司 | Finger vein identification method, device, computer readable storage medium and equipment |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013192234A1 (en) * | 2012-06-18 | 2013-12-27 | The Research Foundation For The State University Of New York | Systems and methods for identifying historical vasculature cases |
-
2022
- 2022-07-11 CN CN202210807554.7A patent/CN114882539B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN110751059A (en) * | 2019-09-27 | 2020-02-04 | 五邑大学 | Least square method-based finger vein ROI extraction method, device and storage medium |
CN111639556A (en) * | 2020-05-15 | 2020-09-08 | 圣点世纪科技股份有限公司 | Finger axis rotation finger vein image correction method based on non-uniform interpolation |
CN114529950A (en) * | 2020-10-30 | 2022-05-24 | 北京眼神智能科技有限公司 | Finger vein identification method, device, computer readable storage medium and equipment |
Also Published As
Publication number | Publication date |
---|---|
CN114882539A (en) | 2022-08-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108960068B (en) | Light source brightness adjusting device and method for collecting finger vein image | |
CN102663393B (en) | Method for extracting region of interest of finger vein image based on correction of rotation | |
CN108764229B (en) | Water gauge image automatic identification method based on computer vision technology | |
CN103310196B (en) | The finger vein identification method of area-of-interest and direction element | |
CN109426814B (en) | Method, system and equipment for positioning and identifying specific plate of invoice picture | |
CN104933432B (en) | Processing method of finger abdomen crease and finger vein image | |
WO2021027364A1 (en) | Finger vein recognition-based identity authentication method and apparatus | |
CN106096569B (en) | A kind of finger vein identification method | |
CN103955674B (en) | Palm print image acquisition device and palm print image positioning and segmenting method | |
JPH0471079A (en) | Positioning method for image | |
CN109409181B (en) | Independent detection method for upper and lower edges of fingers for low-quality finger vein image | |
CN110909631B (en) | Finger vein image ROI extraction and enhancement method | |
CN113936307A (en) | Vein image recognition method and device based on thin film sensor | |
CN114882539B (en) | Vein image ROI extraction method and device | |
CN109308462B (en) | Finger vein and knuckle print region-of-interest positioning method | |
CN115223211B (en) | Identification method for converting vein image into fingerprint image | |
CN107480678A (en) | A kind of chessboard recognition methods and identifying system | |
WO2018196521A1 (en) | Quality score calculation method and device for finger vein image | |
CN112926516B (en) | Robust finger vein image region-of-interest extraction method | |
CN113936303A (en) | Method for determining maximum inscribed rectangle of hand image and image identification method | |
Peng et al. | An effective preprocessing method for finger vein recognition | |
Dong et al. | Eye detection based on integral projection and hough round transform | |
CN115359249B (en) | Palm image ROI region extraction method and system | |
CN115082972B (en) | Living body detection method based on RGB image and vein gray level image | |
CN116309780A (en) | Water gauge water level identification method based on target detection |
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 |