Palm vein feature identification method and device

Abstract

The invention discloses a palm vein feature recognition method and device, and the method comprises the steps: obtaining a palm vein image, extracting a target contour of a palm object in the palm vein image, recognizing key feature points in the target contour, and enabling the key feature points to be used for positioning the edge of a palm vein region, therefore, the palm vein region in the palm vein image can be quickly positioned based on the key feature points used for positioning the edge of the palm vein region, and after the palm vein region is determined, the palm vein feature recognition can be performed in the region including the palm vein feature, so that the recognition accuracy is improved. The efficiency and accuracy of identifying the palm vein features in the palm vein image are effectively improved.

Description

Technical Field

[0001] The present application belongs to the field of biometric identification technology, and in particular, relates to a palm vein feature identification method and device. Background Art

[0002] Palm vein recognition technology is a biometric technology that uses the difference in near-infrared light absorption rate between deoxyhemoglobin in palm veins and other physiological tissues to collect palm vein images using near-infrared reflectance imaging, identify palm vein features from the palm vein images, and use the palm vein features as the basis for identity authentication.

[0003] However, the current process of collecting palm vein images is easily affected by factors such as the palm position, direction, and angle of the fingers, resulting in the collected palm vein images having low quality, thereby reducing the accuracy and efficiency of palm vein feature recognition through the collected palm vein images. Summary of the invention

[0004] The embodiments of the present application provide a palm vein feature recognition method and device, aiming to solve the problem that the accuracy and efficiency of palm vein feature recognition through palm vein images are reduced due to the low quality of the existing palm vein images.

[0005] In a first aspect, an embodiment of the present application provides a method for identifying palm vein features, the method comprising:

[0006] Acquire a palm vein image, and extract a target contour of a palm object in the palm vein image;

[0007] Identifying key feature points in the target contour; wherein the key feature points are used to locate the edge of the palm vein area;

[0008] Based on the key feature points, the palm vein area is determined, and the palm vein features in the palm vein area are identified; wherein the palm vein area is a region in the palm vein image that includes the palm vein features.

[0009] In a possible implementation manner of the first aspect, extracting a target contour of a palm object in the palm vein image includes:

[0010] Determining a grayscale threshold according to image features of the palm vein image, and binarizing the palm vein image based on the grayscale threshold to obtain a binary image;

[0011] The target contour of the palm object is extracted from the binary image.

[0012] In a possible implementation manner of the first aspect, the key feature points include finger joint feature points for describing finger joint positions in a palm object, and identifying the key feature points in the target contour includes:

[0013] Identify a finger contour in the target contour, and determine at least two target line segments based on the finger contour; wherein the target line segment is a line segment tangent to the finger contour, and the finger contour is a contour of a finger in the palm object;

[0014] According to the target contour and the target line segment, a feature point range corresponding to the target line segment in the target contour is determined, and finger joint feature points are identified from the feature point range; wherein the feature point range is determined by the intersection of the target line segment and the target contour, and different target line segments correspond to different finger joint feature points.

[0015] In a possible implementation manner of the first aspect above, the finger gap feature point includes a first feature point and a second feature point, the first feature point is a feature point where an index finger contour and a middle finger contour intersect in the finger contour, and the second feature point is a feature point where a ring finger contour and a little finger contour intersect in the finger contour. Before determining the palm vein area based on the key feature points, the method further includes:

[0016] Determine a test line segment according to the first feature point and the second feature point; wherein the direction of the test line segment is the direction from the first feature point to the second feature point;

[0017] When the angle between the direction of the test line segment and the contour direction of the target contour is not within the first angle range, the first feature point and the second feature point are reselected from the feature point range until the angle between the direction of the reselected test line segment and the contour direction is within the first angle range, and the step of determining the palm vein area based on the key feature points is performed based on the reselected first feature point and the reselected second feature point.

[0018] In a possible implementation manner of the first aspect, the key feature points further include edge feature points, and determining the palm vein area based on the key feature points includes:

[0019] Based on the finger joint feature points, edge feature points of the target contour are determined; wherein the edge feature points are feature points that describe the edge of the palm in the palm object;

[0020] The palm vein area is determined according to the edge feature points.

[0021] In a possible implementation manner of the first aspect, the finger joint feature point further includes a third feature point, where the third feature point is a feature point where a thumb contour and an index finger contour intersect in the finger contour, and determining the edge feature point of the target contour based on the finger joint feature point includes:

[0022] Determine a first extension line according to the first feature point and the second feature point, and determine a first edge feature point and a second edge feature point where the first extension line intersects the target contour; wherein the first edge feature point is a feature point located in the contour of the index finger, and the second edge feature point is a feature point located outside the contour of the little finger;

[0023] Determine a second extension line according to the first edge feature point and the third feature point, and determine a third edge feature point where the second extension line intersects the target contour; wherein the third edge feature point is a feature point outside the thumb contour;

[0024] A fourth edge feature point in the target contour is determined according to the first edge feature point, the second edge feature point, and the third edge feature point.

[0025] In a possible implementation manner of the first aspect, determining a fourth edge feature point in the target contour according to the first edge feature point, the second edge feature point, and the third edge feature point includes:

[0026] Determine an angle bisector between the first extension line and the second extension line;

[0027] A fourth edge feature point where the angle bisector intersects the target contour is determined.

[0028] In a possible implementation manner of the first aspect above, before determining the palm vein area according to the edge feature points, the method further includes:

[0029] According to the fourth edge feature point, determining a feature range to be determined; wherein the feature range to be determined is determined based on an intersection point where a target circular area intersects with the target contour, and the target circular area is a circular area with the fourth edge feature point as the center and the target length as the diameter;

[0030] Determine a fifth edge feature point in the undetermined feature range, so as to perform the step of determining the palm vein area according to the edge feature points based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point; wherein the fifth edge feature point is a feature point in the undetermined feature range whose sum of distances to the first edge feature point, the second edge feature point and the third edge feature point is the largest.

[0031] In a possible implementation manner of the first aspect, determining the fifth edge feature point in the undetermined feature range includes:

[0032] Determine any feature point in the undetermined feature range as a undetermined feature point, and determine a first distance between the undetermined feature point and the first edge feature point, determine a second distance between the undetermined feature point and the second edge feature point, and determine a third distance between the undetermined feature point and the third edge feature point;

[0033] Determine the sum of distances of the undetermined feature points according to the first distance, the second distance and the third distance;

[0034] According to the sum of the distances of each of the undetermined feature points, the undetermined feature point with the largest sum of the distances is determined as the fifth edge feature point.

[0035] In a second aspect, the present invention provides a palm vein feature recognition device, the device comprising:

[0036] An acquisition module, used to acquire a palm vein image and extract a target contour of a palm object in the palm vein image;

[0037] A recognition module, used to recognize key feature points in the target contour; wherein the key feature points are used to locate the edge of the palm vein area;

[0038] A determination module is used to determine the palm vein area based on the key feature points and identify the palm vein features in the palm vein area; wherein the palm vein area is the area in the palm vein image that includes the palm vein features.

[0039] In a third aspect, an embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the palm vein feature recognition method provided in the first aspect above is implemented.

[0040] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the palm vein feature recognition method provided in the first aspect above is implemented.

[0041] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program. When the computer program is run on a computer, the computer executes the palm vein feature recognition method provided in the first aspect above.

[0042] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0043] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0044] In the embodiment of the present invention, a palm vein image is acquired, and a target contour of a palm object in the palm vein image is extracted, and key feature points in the target contour are identified, and the key feature points are used to locate the edge of the palm vein area. Then, based on the key feature points used to locate the edge of the palm vein area, the palm vein area in the palm vein image can be quickly located, and after the palm vein area is determined, the palm vein features can be identified in the area including the palm vein features, thereby effectively improving the efficiency and accuracy of identifying the palm vein features in the palm vein image. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flowchart of a method for identifying palm vein features provided by an embodiment of the present application;

[0046] Figure 2 is a schematic diagram of a target contour in a binarized image provided by an embodiment of the present application;

[0047] Figure 3 is a schematic diagram of a target contour of a palm object provided by an embodiment of the present application;

[0048] Figure 4 is a schematic diagram of a binary image of a palm object provided by an embodiment of the present application;

[0049] Figure 5 is a schematic diagram of a target line segment provided by an embodiment of the present application;

[0050] Figure 6 is a schematic diagram of a palm vein area provided by an embodiment of the present application;

[0051] Figure 7 is a schematic diagram of a contour direction of a target contour provided by an embodiment of the present application;

[0052] Figure 8 is a flowchart of another palm vein feature recognition method provided by an embodiment of the present application;

[0053] Fig. 9 is a schematic diagram of edge feature points in a target contour provided by an embodiment of the present application;

[0054] Fig.10 It is a flowchart of a palm vein feature recognition method provided by an embodiment of the present application;

[0055] Fig.11 It is a structural schematic diagram of a palm vein feature recognition device provided by an embodiment of the present application;

[0056] Fig.12 It is a structural block diagram of a terminal device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] Palm vein recognition technology is a biometric technology that uses the difference in the absorption rate of near-infrared light between deoxyhemoglobin in palm veins and other physiological tissues to collect palm vein images using near-infrared reflective imaging, that is, to collect palm images, and the palm images are images that include the internal veins of the palm, and then the palm vein features can be identified from the palm vein images and converted into digital signals, which are used as the basis for identity authentication. Palm vein recognition technology is currently widely used in government services, financial transactions and other fields because it is highly anti-counterfeit, has obvious features, stable characteristics and is easy to use.

[0059] However, currently palm vein images can be collected in a non-contact manner, such as using a near-infrared camera to collect palm vein images. However, the images collected by the near-infrared camera may have a lot of background interference, which reduces the quality of the collected palm vein images, thereby reducing the accuracy of palm vein feature analysis based on low-quality palm vein images.

[0060] Moreover, when performing image acquisition in a non-contact manner, the position and direction of the palm, the angle of opening of the fingers in the palm, and the distance between the palm and the camera may be different each time the image is acquired, so that the palm vein image acquired each time may have different degrees of rotation, offset, tilt, and texture deformation, further reducing the quality of the acquired palm vein image, and further reducing the accuracy of palm vein feature determination based on low-quality palm vein images, and increasing the difficulty of identifying palm vein features in palm vein images acquired from different angles and positions, thereby reducing the efficiency of palm vein feature determination based on low-quality palm vein images.

[0061] Based on this, the embodiments of the present application provide a palm vein feature recognition method and device, which obtains a palm vein image, extracts the target contour of the palm object in the palm vein image, and identifies key feature points in the target contour, and the key feature points are used to locate the edge of the palm vein area. Then, based on the key feature points used to locate the edge of the palm vein area, the palm vein area in the palm vein image can be quickly located, and after the palm vein area is determined, the palm vein features can be recognized in the area including the palm vein features, thereby effectively improving the efficiency and accuracy of recognizing the palm vein features in the palm vein image.

[0062] See also Figure 1 , Figure 1 A flowchart of a palm vein feature recognition method provided by an embodiment of the present invention is shown, which may specifically include the following steps:

[0063] Step 101: Acquire a palm vein image and extract a target contour of a palm object in the palm vein image.

[0064] The palm vein image may be a palm image captured from a palm, and the palm image may include features of palm veins inside the palm, the palm veins may be all veins within the palm, the palm object may be an object representing the palm in the palm vein image, and the target contour may be a contour describing the palm object in the palm vein image.

[0065] In practical applications, a collection device using near-infrared imaging technology can be used to obtain a palm vein image of the palm, and a palm object representing the palm in the palm vein image can be identified, and then the contour of the palm object can be extracted as the target contour.

[0066] In a specific implementation, the outline of the palm object in the palm vein image can be identified by determining the pixel value in the palm vein image and calculating the gradient change of the pixel value, such as using algorithms such as the Sobel operator, the Prewitt operator, and the Roberts Cross operator to identify the outline of the palm object in the palm vein image. The outline of the palm object in the palm vein image can also be identified by creating and training a corresponding convolutional neural network model through a convolutional neural network.

[0067] In an embodiment of the present application, step 101 may include steps 1011 to 1012:

[0068] Step 1011, determining a grayscale threshold according to the image features of the palm vein image, and binarizing the palm vein image based on the grayscale threshold to obtain a binary image.

[0069] The image feature may be a characteristic attribute used to distinguish different objects in the palm vein image, and the image feature may include grayscale features, color features, brightness features, and the like. The grayscale threshold may be used to distinguish the foreground and background in the palm vein image. The binarization process may be an operation of binarizing the pixel value of each pixel in the palm vein image. The binarized image may be an image after the palm vein image is binarized. The pixel value of each pixel in the binarized image is one of two extreme pixel values, and the extreme pixel value corresponds to the number of bits of the palm vein image. For example, when the number of bits of the palm vein image is 8 bits, the corresponding extreme pixel values ​​are 0 and 255, and when the number of bits of the palm vein image is 16 bits, the corresponding extreme pixel values ​​are 0 and 65535.

[0070] After the palm vein image is obtained, the image features of the palm vein image may be acquired, and then the grayscale threshold of the palm vein image may be determined based on the color features and / or grayscale features of the palm vein image.

[0071] In practical applications, the grayscale histogram of the palm vein image can be calculated according to the color features, or the grayscale histogram of the palm vein image can be calculated according to the grayscale features, and then the inter-class variance is calculated by the maximum inter-class variance method and the determined grayscale histogram to obtain the optimal grayscale threshold, that is, the optimal grayscale threshold is used to distinguish the foreground and background in the palm vein image, and the inter-class variance between the distinguished foreground and background is the largest.

[0072] After the grayscale threshold is obtained, the palm vein image can be binarized based on the grayscale threshold to obtain a binary image.

[0073] In practical applications, the grayscale value corresponding to each pixel in the palm vein image can be compared with the grayscale threshold in turn, and the grayscale value of the pixel whose grayscale value is less than the grayscale threshold is adjusted to the grayscale value corresponding to black, and the grayscale value of the pixel whose grayscale value is greater than or equal to the grayscale threshold is adjusted to the grayscale value corresponding to white, and then the grayscale value of all pixels in the palm vein image is binarized to obtain a binary image displayed only in black and white.

[0074] Exemplarily, when the number of bits of the palm vein image is 8 bits, the grayscale values ​​of pixels in the palm vein image whose grayscale values ​​are less than the grayscale threshold can be adjusted to 0, and the grayscale values ​​of pixels in the palm vein image whose grayscale values ​​are greater than or equal to the grayscale threshold can be adjusted to 255. When the number of bits of the palm vein image is 16 bits, the grayscale values ​​of pixels in the palm vein image whose grayscale values ​​are less than the grayscale threshold can be adjusted to 0, and the grayscale values ​​of pixels in the palm vein image whose grayscale values ​​are greater than or equal to the grayscale threshold can be adjusted to 65535.

[0075] Step 1012: extracting the target contour of the palm object from the binary image.

[0076] After obtaining the binary image, the grayscale value of each pixel in the binary image can be traversed to detect one or more pixels whose grayscale values ​​have changed, and the outline of the palm object can be drawn according to all the pixels whose grayscale values ​​have changed, that is, the boundary where the grayscale value in the binary image changes is determined to be the target outline. For example, the grayscale value of each pixel in the binary image can be any one of 0 or 255, and then the boundary where the grayscale value changes from 0 to 255 can be determined to be the target outline.

[0077] See also Figure 2 , Figure 2 FIG. 1 is a schematic diagram showing a target contour in a binary image provided by an embodiment of the present invention, such as Figure 2 As shown, the contour A may be a boundary where the gray value in the binary image changes, that is, the target contour.

[0078] Step 102, identifying key feature points in the target contour.

[0079] Among them, the key feature points can be used to locate the edge of the palm vein area, and the key feature points can include finger gap feature points used to describe the finger gap position in the palm object. The finger gap position can be the gap position between adjacent fingers in the palm object, and the finger gap feature points can be feature points at the finger gap position in the target contour. The finger gap feature points can include a first feature point, a second feature point, and a third feature point. The first feature point can be a feature point where the index finger contour and the middle finger contour intersect in the finger contour, the second feature point can be a feature point where the ring finger contour and the little finger contour intersect in the finger contour, and the third feature point can be a feature point where the thumb contour and the index finger contour intersect in the finger contour. The finger contour can be the contour of the finger in the palm object, and specifically can be the contour of the finger in the target contour representing the palm object. Since the palm object can include the index finger, the middle finger, the ring finger, the little finger, and the thumb, the finger contour can include the index finger contour representing the index finger, the middle finger contour representing the middle finger, the ring finger contour representing the ring finger, the little finger contour representing the little finger, and the thumb contour representing the thumb. The key feature points may also include edge feature points, and the edge feature points may be feature points describing the edge of the palm in the palm object.

[0080] See also Figure 3 , Figure 3 FIG. 4 is a schematic diagram showing a target contour of a palm object provided by an embodiment of the present invention. Figure 3 As shown, contour A can be represented as the target contour of a palm object, the contour in area A1 can be the index finger contour, the contour in area A2 can be the middle finger contour, the contour in area A3 can be the ring finger contour, the contour in area A4 can be the little finger contour, and the contour in area A5 can be the thumb contour.

[0081] After obtaining the target contour, key feature points in the target contour for locating the edge of the palm vein area may be identified.

[0082] In practical applications, the finger contours of the palm object can be identified from the target contour, and then the finger gap positions between every two adjacent finger contours can be determined, and any point from the finger gap positions can be identified as a finger gap feature point.

[0083] Specifically, the interdigital gaps between the fingers are usually the locations of the finger webs in the palm, which are specifically represented as the locations where the depressions appear in the palm object. Then, the locations where the depressions appear can be determined as the interdigital gap locations from the target contour of the palm object.

[0084] In a specific implementation, any point in the gap position can be determined as a finger gap feature point, or the point where each two adjacent finger contours intersect in the gap position can be determined as a finger gap feature point, thereby obtaining the first feature point where the index finger contour intersects with the middle finger contour, the second feature point where the ring finger contour intersects with the little finger contour, and the third feature point where the thumb contour intersects with the index finger contour.

[0085] See also Figure 4 , Figure 4 FIG. 1 is a schematic diagram showing a binary image of a palm object provided by an embodiment of the present invention, such as Figure 4 As shown, the contour B may be the target contour of the palm object, the region 1 may be the finger gap position between the index finger contour and the middle finger contour, the region 2 may be the finger gap position between the ring finger contour and the middle finger contour, and the region 3 may be the finger gap position between the index finger contour and the thumb contour. The feature point B1 may be the first feature point between the index finger contour and the middle finger contour, the feature point B2 may be the second feature point between the ring finger contour and the little finger contour, and the feature point B3 may be the third feature point between the index finger contour and the thumb contour.

[0086] It should be understood that in practical applications, the palm vein image after binary processing is processed and recognized, that is, the binary image is processed and recognized, and Figures 3 to 7 as well as Fig. 9 These are all exemplary illustrations, used to explain the target contour of the palm object in the palm vein image and the key feature points in the target contour.

[0087] In an embodiment of the present application, step 102 may include steps 1021 to 1022:

[0088] Step 1021 , identifying a finger contour in the target contour, and determining at least two target line segments based on the finger contour.

[0089] The target line segment may be a line segment tangent to the finger contour.

[0090] After the target contour is obtained, the finger contours of the palm object are identified from the target contour, and a target line segment tangent to each two adjacent finger contours is determined to obtain at least two target line segments.

[0091] See also Figure 5 , Figure 5 FIG. 4 shows a schematic diagram of a target line segment provided by an embodiment of the present invention, such as Figure 5 As shown, contour C may be the target contour of the palm object, and the target contour may include the contour of the index finger, the contour of the middle finger, the contour of the ring finger, the contour of the little finger, and the contour of the thumb. Among them, the contour of the index finger and the contour of the middle finger, the contour of the middle finger and the contour of the ring finger, the contour of the ring finger and the contour of the little finger, and the contour of the thumb and the contour of the index finger are all two adjacent finger contours. Line segment 1 may be a target line segment tangent to the contour of the index finger and the contour of the middle finger, respectively, line segment 2 may be a target line segment tangent to the contour of the ring finger and the contour of the middle finger, respectively, line segment 3 may be a target line segment tangent to the contour of the ring finger and the contour of the little finger, respectively, and line segment 4 may be a target line segment tangent to the contour of the index finger and the contour of the thumb, respectively.

[0092] Specifically, based on the finger contour, at least a target line segment tangent to the index finger contour and the middle finger contour, and a target line segment tangent to the ring finger contour and the little finger contour can be determined, thereby obtaining at least two target line segments.

[0093] Step 1022, according to the target contour and the target line segment, determine the contour range corresponding to the target line segment in the target contour, and identify the finger joint feature points from the contour range.

[0094] The contour range may be determined by the intersection of the target line segment and the target contour, and different target line segments may correspond to different finger gap feature points.

[0095] After obtaining the target line segment, the intersection point of the target line segment and the target contour can be determined. Generally, the number of intersection points between the target line segment and the target contour is 2, and then the contour range between the determined intersection points can be determined from the target contour as the feature point range corresponding to the target line segment.

[0096] For example, Figure 5 The line segment 1 in the figure can be tangent to the outline of the index finger and the outline of the middle finger, so that the intersection point C1 where the line segment 1 is tangent to the outline of the index finger and the intersection point C2 where the line segment 1 is tangent to the outline of the middle finger can be determined, and the contour range 5 between the intersection point C1 and the intersection point C2 in the target contour can be determined, which is the contour range corresponding to the line segment 1.

[0097] After obtaining the contour range corresponding to the target line segment, the finger joint feature points can be identified from the contour range.

[0098] In practical applications, any feature point in the contour range can be determined as a finger gap feature point.

[0099] Step 103: determine the palm vein area based on the key feature points, and identify the palm vein features in the palm vein area.

[0100] The palm vein area may be an area in the palm vein image that includes palm vein features.

[0101] After the key feature points are obtained, all the key feature points may be connected to obtain a palm vein region, and the palm vein features of the palm vein region may be identified to obtain the palm vein features in the palm vein region.

[0102] In a specific implementation, since the key feature points are usually finger joint feature points, and the area determined based on the finger joint feature points is relatively small, that is, the determined area is difficult to include all the palm vein features in the palm object, and the palm vein features identified from the determined area are difficult to apply to the required field.

[0103] Based on this, auxiliary intersections can also be determined based on key feature points, so as to determine the palm vein area based on each key feature point and the auxiliary intersection point, so as to reduce the situation where the palm vein features in the determined area are small and difficult to apply. Specifically, any two key feature points can be connected to obtain auxiliary line segments, and the perpendicular bisector of the auxiliary line segments can be determined, and the intersection of the perpendicular bisector and the edge of the palm in the target contour can be determined to obtain auxiliary intersections, and then each key feature point and the auxiliary intersection point can be connected to obtain a polygonal area, which is the palm vein area.

[0104] Specifically, the palm edge may be the outline of the palm area representing the palm object in the target outline.

[0105] Generally, the key feature points may include a first feature point, a second feature point and a third feature point, and the palm vein area may be a quadrilateral area. When the key feature points also include a finger gap feature point between the middle finger contour and the ring finger contour, the palm vein area may be a pentagonal area.

[0106] See also Figure 6 , Figure 6 FIG. 4 shows a schematic diagram of a palm vein region provided by an embodiment of the present invention, such as Figure 6 As shown, the contour D can be the target contour of the palm object, the line segment 1 can be an auxiliary line segment determined based on the first feature point D1 and the second feature point D2, the line segment 2 can be the perpendicular bisector of the line segment 1, the line segment 2 intersects with the edge of the palm in the target contour, and the intersection point is the auxiliary intersection point D4, and then the first feature point D1, the second feature point D2, the third feature point D3 and the auxiliary intersection point D4 can be connected to obtain a quadrilateral area, which is the palm vein area.

[0107] In an embodiment of the present application, before step 103, the following steps may also be included:

[0108] A test line segment is determined based on the first feature point and the second feature point. When the angle between the direction of the test line segment and the contour direction of the target contour is not within the first angle range, the first feature point and the second feature point are reselected from the contour range until the angle between the direction of the reselected test line segment and the contour direction is within the first angle range, and based on the reselected first feature point and the reselected second feature point, a step of determining the palm vein area based on the key feature points is performed.

[0109] Among them, the test line segment can be a line segment connecting the first feature point and the second feature point, the direction of the test line segment can be the direction from the first feature point to the second feature point, the contour direction can be expressed as the direction of the target contour, specifically the direction determined based on the finger contour in the target contour, the first angle range can be a user-defined range, and the first angle range can be used to determine the degree of angular deviation of the palm object in the palm vein image.

[0110] After obtaining the first feature point and the second feature point, the first feature point and the second feature point may be connected to obtain a test line segment, and the direction of the test line segment is determined to be the direction from the first feature point to the second feature point.

[0111] At the same time, any two adjacent finger joint feature points can be connected to obtain a first intermediate line segment for determining the contour direction, and the perpendicular bisector of the first intermediate line segment is determined, and the direction from the palm area to the finger in the perpendicular bisector is determined as the contour direction. Exemplarily, the first feature point between the index finger contour and the middle finger contour can be determined, and the finger joint feature point adjacent to the first feature point can be determined, such as the finger joint feature point or the third feature point between the middle finger contour and the ring finger contour, and then the finger joint feature point between the middle finger contour and the ring finger contour can be connected with the first feature point to obtain the first intermediate line segment for determining the contour direction, and the perpendicular bisector of the first intermediate line segment is determined, and the direction from the palm area to the finger in the perpendicular bisector is determined as the contour direction.

[0112] In one embodiment of the present invention, the tangent points of the contours of two adjacent fingers in the finger contour and the corresponding target line segment can also be determined. For example, the tangent points of the target line segment and the contour of the index finger can be determined, and the tangent points of the same target line segment and the contour of the middle finger can be determined. Then, the tangent points of the finger contour and the target line segment can be connected with the finger joint feature points corresponding to the target line segment, respectively, to obtain two second intermediate line segments for determining the contour direction, namely, the second intermediate line segment where the tangent point of the target line segment and the contour of the index finger is connected with the finger joint feature points corresponding to the target line segment, and the second intermediate line segment where the tangent point of the target line segment and the contour of the middle finger is connected with the finger joint feature points corresponding to the target line segment, and the angle bisector between the second intermediate line segments is determined, which is the contour direction.

[0113] See also Figure 7 , Figure 7 A schematic diagram showing the contour direction of a target contour provided by an embodiment of the present invention is shown as follows: Figure 7As shown, the contour E can be the target contour of the palm object, the target line segment 1 can be the target line segment tangent to the index finger contour and the middle finger contour, the tangent point E1 can be the tangent point where the target line segment 1 is tangent to the index finger contour, the tangent point E2 can be the tangent point where the target line segment 1 is tangent to the middle finger contour, the finger joint feature point E3 can be the finger joint feature point corresponding to the target line segment 1, the second intermediate line segment 2 can be the line segment connecting the tangent point E1 and the finger joint feature point E3, the second intermediate line segment 3 can be the line segment connecting the tangent point E2 and the finger joint feature point E3, the angle bisector 4 can be the angle bisector between the second intermediate line segment 2 and the second intermediate line segment 3, and the direction of the angle bisector 4 can be defined as the direction from the palm area to the finger, that is, the contour direction of the target contour E.

[0114] After obtaining the direction of the test line segment and the contour direction, the angle between the direction of the test line segment and the contour direction can be determined, and the determined angle can be compared with the first angle range. When the determined angle is not within the first angle range, it can be determined that the angle deviation of the palm object in the palm vein image is large, that is, the quality of the palm vein image collected this time is greatly affected by the angle deviation, and it is difficult to identify the palm vein area from the palm vein image collected this time. Even if it can be identified, it is difficult to identify the correct palm vein features from the identified palm vein area, and there is a problem of low accuracy in palm vein feature recognition.

[0115] Based on this, when the determined angle is not within the first angle range, the first feature point and the second feature point can be reselected from the contour range until the angle between the direction of the reselected test line segment and the contour direction is within the first angle range, and based on the reselected first feature point and the reselected second feature point, the step of determining the palm vein area based on the key feature points is performed.

[0116] When the determined angle is within the first angle range, it can be determined that the angle deviation of the palm object in the palm vein image is small, that is, the quality of the palm vein image acquired this time is less affected by the angle deviation, and the palm vein area can be directly identified from the palm vein image acquired this time, and the palm vein features can be accurately identified from the identified palm vein area, and there is no need to reselect the first feature point and the second feature point, that is, there is no need to process the first feature point and the second feature point, and the first feature point and the second feature point can be directly used to perform the step of determining the palm vein area based on the key feature points.

[0117] Exemplarily, the first feature point and the second feature point may be reselected in the following manner:

[0118] Determine a first adjustment range of the first feature point and a second adjustment range of the second feature point, select any one feature point from the first adjustment range as the re-selected first feature point, and select any one feature point from the second adjustment range as the re-selected second feature point, until the angle between the direction of the re-selected first feature point pointing to the re-selected second feature point and the contour direction is within the first angle range.

[0119] The first adjustment range may be a range for reselecting the first feature point from the contour range, and the second adjustment range may be a range for reselecting the second feature point from the contour range.

[0120] When the determined angle is not within the first angle range, a first circular area can be determined with the first feature point as the center and a predetermined length as the diameter, and a first adjustment range within the first circular area within the contour range can be determined. Similarly, a second circular area can be determined with the second feature point as the center and a predetermined length as the diameter, and a second adjustment range within the second circular area within the contour range can be determined. Then, the first feature point can be reselected from the first adjustment range, and the second feature point can be reselected from the second adjustment range. Based on the reselected first feature point and the reselected second feature point, the angle between the direction of the test line segment and the contour direction can be determined until the angle between the direction of the reselected first feature point pointing to the reselected second feature point and the contour direction after reselection is within the first angle range.

[0121] In the embodiment of the present invention, a palm vein image is acquired, and a target contour of a palm object in the palm vein image is extracted, and key feature points in the target contour are identified, and the key feature points are used to locate the edge of the palm vein area. Then, based on the key feature points used to locate the edge of the palm vein area, the palm vein area in the palm vein image can be quickly located, and after the palm vein area is determined, the palm vein features can be identified in the area including the palm vein features, thereby effectively improving the efficiency and accuracy of identifying the palm vein features in the palm vein image.

[0122] See also Figure 8 , Figure 8 A flowchart of another palm vein feature recognition method provided by an embodiment of the present invention is shown, which may specifically include the following steps:

[0123] Step 801, acquiring a palm vein image, and extracting a target contour of a palm object in the palm vein image.

[0124] Step 802: Identify a finger contour in a target contour, and determine at least two target line segments based on the finger contour.

[0125] Step 803: Determine the contour range corresponding to the target line segment in the target contour according to the target contour and the target line segment, and identify the finger joint feature points from the contour range.

[0126] For the relevant descriptions of steps 801 to 803, please refer to steps 101 to 102 and will not be repeated here.

[0127] Step 804: determining edge feature points of the target contour based on the finger joint feature points.

[0128] The edge feature points may be feature points describing the edge of the palm in the palm object.

[0129] After the finger gap feature points are obtained, feature points in the target contour used to describe the edge of the palm in the palm object, namely, edge feature points, can be determined based on the finger gap feature points.

[0130] In an embodiment of the present application, step 804 may include steps 8041 to 8043:

[0131] Step 8041, determining a first extension line according to the first feature point and the second feature point, and determining a first edge feature point and a second edge feature point where the first extension line intersects the target contour.

[0132] Among them, the first extension line can be an extension line connecting the first feature point and the second feature point, the first edge feature point can be a feature point in the index finger outline, and the second edge feature point can be a feature point outside the little finger outline. The outside of the little finger outline can be represented as the side of the little finger outline away from the palm area.

[0133] After obtaining the first feature point and the second feature point, the first feature point and the second feature point may be connected to obtain a first extension line, and a first edge feature point and a second edge feature point where the first extension line intersects the target contour may be determined.

[0134] See also Fig. 9 , Fig. 9 FIG. 4 is a schematic diagram showing edge feature points in a target contour provided by an embodiment of the present invention. Fig. 9 As shown, the contour F may be the target contour of the palm object, the first extension line 1 may be obtained by connecting the first feature point F1 and the second feature point F2, and the first edge feature point F4 and the second edge feature point F5 may be the intersection points where the first extension line 1 and the contour F intersect.

[0135] Step 8042: Determine a second extension line based on the first edge feature point and the third feature point, and determine a third edge feature point where the second extension line intersects the target contour.

[0136] The second extension line may be an extension line connecting the first edge feature point and the third edge feature point, the third edge feature point may be a feature point outside the thumb contour, and the outside of the thumb contour may be represented as the side of the thumb contour away from the palm.

[0137] After obtaining the first edge feature point, the first edge feature point and the third feature point may be connected to obtain a second extension line, and a third edge feature point where the second extension line intersects the target contour may be determined.

[0138] like Fig. 9 As shown, the second extension line 2 may be an extension line connecting the third feature point F3 and the first edge feature point F4, and the third edge feature point F6 may be an intersection point where the second extension line 2 and the contour F intersect.

[0139] Step 8043: Determine a fourth edge feature point in the target contour according to the first edge feature point, the second edge feature point and the third edge feature point.

[0140] The fourth edge feature point may be a feature point at the edge of the palm.

[0141] After obtaining the first edge feature point, the second edge feature point and the third edge feature point, any point in the palm edge of the target contour can be determined as a fourth edge feature point based on the first edge feature point, the second edge feature point and the third edge feature point.

[0142] Exemplarily, any point in the palm edge of the target contour can be determined as a pending feature point, and the determined pending feature point can be connected with the first edge feature point, the second edge feature point, and the third edge feature point in sequence to obtain a quadrilateral region, and the area of ​​the quadrilateral region can be determined, thereby obtaining the area of ​​the quadrilateral region corresponding to each pending feature point in the palm edge of the target contour, and then the pending feature point corresponding to the quadrilateral region with the largest area can be determined as the fourth edge feature point.

[0143] In an embodiment of the present application, step 8043 may include the following steps:

[0144] An angle bisector between the first extension line and the second extension line is determined, and a fourth edge feature point where the angle bisector intersects the target contour is determined.

[0145] After obtaining the first extension line and the second extension line, the angle bisector between the first extension line and the second extension line may be determined, and the intersection point where the angle bisector intersects the target contour may be determined as the fourth edge feature point.

[0146] like Fig. 9 As shown, the angle bisector 3 may be the angle bisector between the first extension line 1 and the second extension line 2 , and the fourth edge feature point F7 may be the intersection point where the angle bisector 3 and the contour F intersect.

[0147] Step 805: determine the palm vein area according to the edge feature points, and identify the palm vein features in the palm vein area.

[0148] After the edge feature points are obtained, all edge feature points may be connected in sequence to obtain a palm vein region, and then the palm vein features of the palm vein region may be identified to obtain the palm vein features in the palm vein region.

[0149] like Fig. 9 As shown, line segment 5 can be a line segment connecting the third edge feature point F6 and the fourth edge feature point F7, and line segment 6 can be a line segment connecting the second edge feature point F5 and the fourth edge feature point F7. Then the palm vein area determined by the first edge feature point F4, the second edge feature point F5, the third edge feature point F6 and the fourth edge feature point F7 can be an area divided by line segment 5, line segment 6, the line segment in the first extension line 1 between the first edge feature point F4 and the second edge feature point F5, and the line segment in the first extension line 2 between the first edge feature point F4 and the third edge feature point F6.

[0150] In an embodiment of the present application, before step 805, the following steps may also be included:

[0151] According to the fourth edge feature point, a feature range to be determined is determined, and a fifth edge feature point in the feature range to be determined is determined, so as to perform the step of determining the palm vein area according to the edge feature points based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point.

[0152] Among them, the undetermined feature range can be determined based on the intersection point where the target circular area intersects the target contour, and the target circular area can be a circular area with the fourth edge feature point as the center and the target length as the diameter. The target length can be represented by the number of pixels, and the target length can be user-defined or determined based on experience. The target length can usually be set to 30 pixels. The fifth edge feature point can be any point in the undetermined feature range, and specifically can be a feature point with the largest sum of distances to the first edge feature point, the second edge feature point, and the third edge feature point in the undetermined feature range.

[0153] After obtaining the fourth edge feature point, the target circular area can be determined with the fourth edge feature point as the center and the target length as the diameter, and then the intersection point where the target circular area intersects with the target contour can be determined, and the contour range in the target contour between the intersection points can be determined as the to-be-determined feature range.

[0154] It should be understood that, since the number of points where the target circular area intersects with the target contour is usually at least two, the contour range between the intersection points can be determined as the undetermined feature range.

[0155] like Fig. 9 As shown, the target circular area 4 can be centered on the fourth edge feature point F7 and determined by the target length as the diameter, the intersection point F8 and the intersection point F9 can be the intersection points where the target circular area 4 intersects with the target contour F, and the to-be-determined feature range can be the contour range in the target contour F between the intersection point F8 and the intersection point F9, that is, the contour of the target contour F in the target circular area 4.

[0156] After obtaining the undetermined feature range, any point in the undetermined feature range can be determined as the fifth edge feature point, and then the step of determining the palm vein area according to the edge feature points can be performed based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point, that is, the palm vein area is determined based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point.

[0157] In an embodiment of the present application, the range of features to be determined may be determined in the following manner:

[0158] Determine any feature point in the range of pending features as a pending feature point, and determine a first distance between the pending feature point and the first edge feature point, determine a second distance between the pending feature point and the second edge feature point, and determine a third distance between the pending feature point and the third edge feature point; determine the sum of distances of the pending feature points based on the first distance, the second distance, and the third distance; and determine the pending feature point with the largest distance sum as the fifth edge feature point based on the sum of distances of each pending feature point.

[0159] Among them, the pending feature point can be any feature point in the pending feature range, the first distance can be the distance between the pending feature point and the first edge feature point, the second distance can be the distance between the pending feature point and the second edge feature point, and the third distance can be the distance between the pending feature point and the third edge feature point.

[0160] After determining the range of the pending feature, any feature point in the range of the pending feature can be determined as the pending feature point, and then the first distance between each pending feature point and the first edge feature point, the second distance between the pending feature point and the second edge feature point, and the third distance between the pending feature point and the third edge feature point can be determined, that is, the first distance, the second distance and the third distance corresponding to each pending feature point can be obtained, and for each pending feature point, the first distance, the second distance and the third distance corresponding to the pending feature point can be added to obtain the sum of the distances corresponding to the pending feature point, and then the sum of the distances corresponding to each pending feature point can be obtained.

[0161] After obtaining the distance sum corresponding to each pending feature point, the distance sum corresponding to each pending feature point can be compared, so as to determine the pending feature point with the largest distance sum from all the distance sums, and determine the pending feature point as the fifth edge feature point, so as to perform the step of determining the palm vein area according to the edge feature points based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point.

[0162] It should be understood that although the palm vein area determined based on the first edge feature point, the second edge feature point, the third edge feature point and the fourth edge feature point is the area with the largest area determined based on the first edge feature point, the second edge feature point and the third edge feature point in the target contour, there may be a situation where some palm vein features are missed in the determined palm vein area due to the angle offset of the palm object in the palm vein image, the distance offset from the acquisition device and the texture deformation, thereby reducing the accuracy of palm vein feature recognition for the palm vein area.

[0163] By determining the fifth edge feature point in the pending feature range and determining the palm vein area based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point, the situation in which some palm vein features are missed in the determined palm vein area can be reduced, thereby improving the accuracy of palm vein feature identification for the palm vein area.

[0164] In an embodiment of the present invention, a palm vein image is acquired, and a target contour of a palm object in the palm vein image is extracted, a finger contour in the target contour is identified, and at least two target line segments are determined based on the finger contour, and a contour range corresponding to the target line segments in the target contour is determined according to the target contour and the target line segments, and finger joint feature points are identified from the contour range, thereby realizing identification of finger joint feature points for locating the edge of a palm vein area from the target contour, and determining edge feature points of the target contour based on the finger joint feature points, and determining a palm vein area based on the edge feature points, thereby being able to quickly locate the palm vein area in the palm vein image based on the edge feature points for locating the edge of the palm vein area, and after determining the palm vein area, being able to identify the palm vein feature in the area including the palm vein feature, thereby effectively improving the efficiency and accuracy of identifying the palm vein feature in the palm vein image.

[0165] Reference Fig.10 , Fig.10 A schematic flow chart of a palm vein feature recognition method provided by an embodiment of the present invention is shown, which may specifically include the following steps:

[0166] Step 1001, collecting a palm vein image;

[0167] Step 1002, binarizing the palm vein image to obtain a binarized image;

[0168] Step 1003, extracting the target contour of the palm object from the binary image;

[0169] Step 1004, identifying a first feature point, a second feature point, and a third feature point in the target contour;

[0170] Step 1005, determining a first edge feature point and a second edge feature point according to the first feature point and the second feature point;

[0171] Step 1006, determining a third edge feature point according to the first feature point and the third feature point;

[0172] Step 1007, determining a fourth edge feature point according to the first edge feature point, the second edge feature point and the third edge feature point;

[0173] Step 1008 , determining a palm vein region based on the first edge feature point, the second edge feature point, the third edge feature point, and the fourth edge feature point, and identifying palm vein features in the palm vein region.

[0174] In an embodiment of the present invention, a palm vein image is collected, a binarization process is performed on the palm vein image to obtain a binarized image, a target contour of a palm object is extracted from the binarized image, a first feature point, a second feature point, and a third feature point in the target contour are identified, a first edge feature point and a second edge feature point are determined according to the first feature point and the second feature point, a third edge feature point is determined according to the first feature point and the third feature point, a fourth edge feature point is determined according to the first edge feature point, the second edge feature point, and the third edge feature point, and a palm vein region is determined based on the first edge feature point, the second edge feature point, the third edge feature point, and the fourth edge feature point, thereby enabling the palm vein region in the palm vein image to be quickly located based on the key feature points used to locate the edge of the palm vein region, and the palm vein features in the palm vein region to be identified, and after the palm vein region is determined, the palm vein features can be identified in the region including the palm vein features, thereby effectively improving the efficiency and accuracy of identifying the palm vein features in the palm vein image.

[0175] Reference Fig.11 , Fig.11 The structure diagram of a palm vein feature recognition device provided by an embodiment of the present invention is shown, which may specifically include the following modules:

[0176] An acquisition module 1101 is used to acquire a palm vein image and extract a target contour of a palm object in the palm vein image;

[0177] The recognition module 1102 is used to recognize key feature points in the target contour; wherein the key feature points are used to locate the edge of the palm vein area;

[0178] The determination module 1103 is used to determine the palm vein area based on the key feature points, and identify the palm vein features in the palm vein area; wherein the palm vein area is the area in the palm vein image that includes the palm vein features.

[0179] In one implementation, the acquisition module 1101 may be specifically used for:

[0180] Determining a grayscale threshold according to image features of the palm vein image, and binarizing the palm vein image based on the grayscale threshold to obtain a binary image;

[0181] Extract the target contour of the palm object from the binary image.

[0182] In one implementation, the key feature points include finger joint feature points used to describe finger joint positions in the palm object. The recognition module 1102 may be specifically used to:

[0183] Identify the finger contour in the target contour, and determine at least two target line segments based on the finger contour; wherein the target line segment is a line segment tangent to the finger contour, and the finger contour is the contour of the finger in the palm object;

[0184] According to the target contour and the target line segment, the contour range corresponding to the target line segment in the target contour is determined, and the finger joint feature points are identified from the contour range; wherein the contour range is determined by the intersection of the target line segment and the target contour, and different target line segments correspond to different finger joint feature points.

[0185] In one implementation, the finger gap feature point includes a first feature point and a second feature point, the first feature point is a feature point where the index finger contour and the middle finger contour intersect in the finger contour, and the second feature point is a feature point where the ring finger contour and the little finger contour intersect in the finger contour. The above-mentioned determination module 1103 can also be used for;

[0186] Before determining the palm vein area based on the key feature point, determining a test line segment according to the first feature point and the second feature point; wherein the direction of the test line segment is the direction from the first feature point to the second feature point;

[0187] When the angle between the direction of the test line segment and the contour direction of the target contour is not within the first angle range, the first feature point and the second feature point are reselected from the contour range until the angle between the direction of the reselected test line segment and the contour direction is within the first angle range, and the step of determining the palm vein area based on the key feature points is performed based on the reselected first feature point and the reselected second feature point.

[0188] In one implementation, the key feature points also include edge feature points, and the above-mentioned determination module 1103 can be specifically used for:

[0189] Based on the finger joint feature points, edge feature points of the target contour are determined; wherein the edge feature points are feature points that describe the edge of the palm in the palm object;

[0190] The palm vein area is determined based on the edge feature points.

[0191] In one implementation, the finger joint feature point further includes a third feature point, and the third feature point is a feature point where the thumb contour and the index finger contour intersect in the finger contour. The above-mentioned determination module 1103 can be specifically used for:

[0192] According to the first feature point and the second feature point, a first extension line is determined, and a first edge feature point and a second edge feature point where the first extension line intersects the target contour are determined; wherein the first edge feature point is a feature point located in the contour of the index finger, and the second edge feature point is a feature point located outside the contour of the little finger;

[0193] Determine a second extension line according to the first edge feature point and the third feature point, and determine a third edge feature point where the second extension line intersects the target contour; wherein the third edge feature point is a feature point outside the thumb contour;

[0194] A fourth edge feature point in the target contour is determined according to the first edge feature point, the second edge feature point, and the third edge feature point.

[0195] In one implementation, the determination module 1103 may be specifically used to:

[0196] Determine the angle bisector of the first extension line and the second extension line;

[0197] Determine the fourth edge feature point where the angle bisector intersects the target contour.

[0198] In one implementation, the determination module 1103 may also be used to:

[0199] Before determining the palm vein area according to the edge feature point, determine the undetermined feature range according to the fourth edge feature point; wherein the undetermined feature range is determined based on the intersection point where the target circular area intersects with the target contour, and the target circular area is a circular area with the fourth edge feature point as the center and the target length as the diameter;

[0200] Determine a fifth edge feature point in the feature range to be determined, so as to perform the step of determining the palm vein area according to the edge feature points based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point; wherein the fifth edge feature point is a feature point in the feature range to be determined with the maximum sum of distances from the first edge feature point, the second edge feature point and the third edge feature point.

[0201] In one implementation, the determination module 1103 may be specifically used to:

[0202] Determine any feature point in the undetermined feature range as a undetermined feature point, and determine a first distance between the undetermined feature point and a first edge feature point, determine a second distance between the undetermined feature point and a second edge feature point, and determine a third distance between the undetermined feature point and a third edge feature point;

[0203] Determine the sum of distances of the undetermined feature points according to the first distance, the second distance, and the third distance;

[0204] According to the sum of the distances of each undetermined feature point, the undetermined feature point with the largest sum of the distances is determined as the fifth edge feature point.

[0205] In the embodiment of the present invention, a palm vein image is acquired, and a target contour of a palm object in the palm vein image is extracted, and key feature points in the target contour are identified, and the key feature points are used to locate the edge of the palm vein area. Then, based on the key feature points used to locate the edge of the palm vein area, the palm vein area in the palm vein image can be quickly located, and after the palm vein area is determined, the palm vein features can be identified in the area including the palm vein features, thereby effectively improving the efficiency and accuracy of identifying the palm vein features in the palm vein image.

[0206] It should be noted that the information interaction, execution process and other contents between the above-mentioned devices are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0207] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0208] Reference Fig.12 , Fig.12 A structural block diagram of a terminal device provided by an embodiment of the present application is shown as follows: Fig.12 As shown, an embodiment of the present application also provides a terminal device 121, which includes: at least one processor 1212, a memory 1212, and a computer program 12121 stored in the memory 1212 and executable on at least one processor 1212, and when the processor 1212 executes the computer program 12121, the steps in any of the above-mentioned method embodiments are implemented.

[0209] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments can be implemented.

[0210] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0211] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device capable of carrying the computer program code to the camera device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium.

[0212] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A palm vein feature recognition method, characterized in that: The method comprises: Acquire a palm vein image, and extract a target contour of a palm object in the palm vein image; Identifying key feature points in the target contour; wherein the key feature points are used to locate the edge of the palm vein area; Based on the key feature points, the palm vein area is determined, and the palm vein features in the palm vein area are identified; wherein the palm vein area is a region in the palm vein image that includes the palm vein features.

2. The palm vein feature recognition method according to claim 1, characterized in that: The step of extracting a target contour of a palm object in the palm vein image comprises: Determining a grayscale threshold according to image features of the palm vein image, and binarizing the palm vein image based on the grayscale threshold to obtain a binary image; The target contour of the palm object is extracted from the binary image.

3. The palm vein feature recognition method according to claim 1 or 2, characterized in that: The key feature points include finger joint feature points for describing finger joint positions in a palm object, and the identifying key feature points in the target contour includes: Identify a finger contour in the target contour, and determine at least two target line segments based on the finger contour; wherein the target line segment is a line segment tangent to the finger contour, and the finger contour is a contour of a finger in the palm object; According to the target contour and the target line segment, a feature point range corresponding to the target line segment in the target contour is determined, and finger joint feature points are identified from the feature point range; wherein the feature point range is determined by the intersection of the target line segment and the target contour, and different target line segments correspond to different finger joint feature points.

4. The palm vein feature recognition method according to claim 3, characterized in that: The finger gap feature points include a first feature point and a second feature point, the first feature point is a feature point where the index finger contour and the middle finger contour intersect in the finger contour, and the second feature point is a feature point where the ring finger contour and the little finger contour intersect in the finger contour. Before determining the palm vein area based on the key feature points, the method further includes: Determine a test line segment according to the first feature point and the second feature point; wherein the direction of the test line segment is the direction from the first feature point to the second feature point; When the angle between the direction of the test line segment and the contour direction of the target contour is not within the first angle range, the first feature point and the second feature point are reselected from the feature point range until the angle between the direction of the reselected test line segment and the contour direction is within the first angle range, and the step of determining the palm vein area based on the key feature points is performed based on the reselected first feature point and the reselected second feature point.

5. The palm vein feature recognition method according to claim 4, characterized in that: The key feature points also include edge feature points, and determining the palm vein area based on the key feature points includes: Based on the finger joint feature points, edge feature points of the target contour are determined; wherein the edge feature points are feature points that describe the edge of the palm in the palm object; The palm vein area is determined according to the edge feature points.

6. The palm vein feature recognition method according to claim 5, characterized in that: The finger gap feature points further include a third feature point, which is a feature point where the thumb contour and the index finger contour intersect in the finger contour. The determining of the edge feature points of the target contour based on the finger gap feature points includes: Determine a first extension line according to the first feature point and the second feature point, and determine a first edge feature point and a second edge feature point where the first extension line intersects the target contour; wherein the first edge feature point is a feature point located in the contour of the index finger, and the second edge feature point is a feature point located outside the contour of the little finger; Determine a second extension line according to the first edge feature point and the third feature point, and determine a third edge feature point where the second extension line intersects the target contour; wherein the third edge feature point is a feature point outside the thumb contour; A fourth edge feature point in the target contour is determined according to the first edge feature point, the second edge feature point, and the third edge feature point.

7. The palm vein feature recognition method according to claim 6, characterized in that: The step of determining a fourth edge feature point in the target contour according to the first edge feature point, the second edge feature point, and the third edge feature point comprises: Determine an angle bisector between the first extension line and the second extension line; A fourth edge feature point where the angle bisector intersects the target contour is determined.

8. The palm vein feature recognition method according to claim 6, characterized in that: Before determining the palm vein area according to the edge feature points, the method further includes: According to the fourth edge feature point, determining a feature range to be determined; wherein the feature range to be determined is determined based on an intersection point where a target circular area intersects with the target contour, and the target circular area is a circular area with the fourth edge feature point as the center and the target length as the diameter; Determine a fifth edge feature point in the undetermined feature range, so as to perform the step of determining the palm vein area according to the edge feature points based on the first edge feature point, the second edge feature point, the third edge feature point and the fifth edge feature point; wherein the fifth edge feature point is a feature point in the undetermined feature range whose sum of distances to the first edge feature point, the second edge feature point and the third edge feature point is the largest.

9. The palm vein feature recognition method according to claim 8, characterized in that: The determining of the fifth edge feature point in the undetermined feature range includes: Determine any feature point in the undetermined feature range as a undetermined feature point, and determine a first distance between the undetermined feature point and the first edge feature point, determine a second distance between the undetermined feature point and the second edge feature point, and determine a third distance between the undetermined feature point and the third edge feature point; Determine the sum of distances of the undetermined feature points according to the first distance, the second distance and the third distance; According to the sum of the distances of each of the undetermined feature points, the undetermined feature point with the largest sum of the distances is determined as the fifth edge feature point.

10. A palm vein feature recognition device, characterized in that: The device comprises: An acquisition module, used to acquire a palm vein image and extract a target contour of a palm object in the palm vein image; A recognition module, used to recognize key feature points in the target contour; wherein the key feature points are used to locate the edge of the palm vein area; A determination module is used to determine the palm vein area based on the key feature points and identify the palm vein features in the palm vein area; wherein the palm vein area is the area in the palm vein image that includes the palm vein features.

Images