Palm vein image binarization method, identification method, apparatus, device, and medium

Abstract

The present invention relates to the technical field of biometric identification, and provides a palm vein image binarization method, an identification method, an apparatus, a device, and a medium. The palm vein image binarization method comprises the following steps: acquiring a grayscale range of a palm vein image, and dividing the palm vein image into a first grayscale level region, a second grayscale level region, a third grayscale level region and a fourth grayscale level region on the basis of the grayscale range; sequentially acquiring a connected component of each grayscale level region, calculating the centroid of each connected component, and on the basis of the centroid of the connected component, determining whether the connected component needs to be fused and corrected; and sequentially performing detection on the connected component of each grayscale level region, determining a connected component comprising a palm, and using the minimum grayscale value of the grayscale level region where the connected component comprising the palm is located as a binarization segmentation threshold of the palm vein image. The present invention can effectively solve the problem of unsatisfactory binarization segmentation effect of palm vein images collected in complex outdoor environments.

Description

Palm vein image binarization method, recognition method, device, equipment and medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims priority to Chinese patent application number 2023115580403, filed with the Chinese Patent Office on November 22, 2023, entitled “A Method for Binarizing Vein Images,” the entire contents of which are incorporated by reference into this disclosure. Technical Field

[0003] The present disclosure relates to the field of biometric identification technology, and in particular to a palm vein image binarization method, identification method, device, equipment and medium. Background Art

[0004] When performing identity authentication through vein images, the vein images need to be binarized and segmented. Existing vein image binarization methods, such as OSTU, primarily rely on threshold classification based on histograms. However, due to the non-contact nature of palm vein image acquisition, palm placement is arbitrary, and the captured palm vein images may contain complex backgrounds. Furthermore, in outdoor environments, due to uneven lighting, each object reflects light to varying degrees, resulting in multiple grayscale levels in the captured vein image, similar to concentric circles, with a bright center and dark surroundings. This ultimately results in unsatisfactory binary segmentation of palm vein images captured in complex outdoor environments. Summary of the Invention

[0005] The main purpose of the present disclosure is to provide a palm vein image binarization method, recognition method, device, equipment and medium to solve the problem of unsatisfactory binary segmentation effect of palm vein images collected in complex outdoor environments and improve the accuracy of identity recognition based on palm vein images.

[0006] To achieve the above objectives, the present disclosure provides the following solutions:

[0007] In a first aspect, the present disclosure relates to a method for binarizing a palm vein image, which comprises the following steps:

[0008] Step 1. Obtaining a grayscale range of a palm vein image, and dividing the palm vein image into a first grayscale region, a second grayscale region, a third grayscale region, and a fourth grayscale region according to the grayscale range;

[0009] Step 2. Obtain the connected domains of each grayscale region in turn, calculate the centroid of each connected domain, and determine whether the connected domain needs to be fused and corrected based on the centroid of the connected domain;

[0010] Step 3. Detect the connected domains of each grayscale region in turn to determine the connected domain containing the palm, and use the minimum grayscale value of the grayscale region containing the connected domain containing the palm as the binary segmentation threshold of the palm vein image;

[0011] Optionally, step 1 specifically includes:

[0012] Step 1.1. Obtain the grayscale range of the palm vein image, that is, obtain the maximum grayscale value maxV and the minimum grayscale value minV of the palm vein image;

[0013] Step 1.2. Calculate the width range of the grayscale level using the following formula:

[0014] range = int((maxV - minV) / 4);

[0015] Step 1.3. Divide the palm vein image into a first grayscale region, a second grayscale region, a third grayscale region, and a fourth grayscale region according to the grayscale width range.

[0016] Optionally, the grayscale interval of the first grayscale area is [minV, minV+range), the grayscale interval of the second grayscale area is [minV+range, minV+range*2), the grayscale interval of the third grayscale area is [minV+range*2, minV+range*3), and the grayscale interval of the fourth grayscale area is [minV+range*3, maxV].

[0017] Optionally, before performing step 2, the first grayscale area is removed.

[0018] Optionally, step 2 specifically includes:

[0019] Step 2.1. Obtain all connected domains in the fourth grayscale region and calculate the centroid of each connected domain in turn;

[0020] Step 2.2. Obtain all connected domains in the third grayscale region and calculate the centroid of each connected domain in turn. Determine whether the connected domains meet the fusion conditions based on the centroid of the connected domains in the third grayscale region and the centroid of the connected domains in the fourth grayscale region. If so, fuse the connected domains and also fuse the grayscale regions where the connected domains are located.

[0021] Step 2.3. Obtain all connected domains in the second grayscale area, and calculate the center of gravity of each connected domain in turn. Determine whether the connected domain meets the fusion conditions based on the center of gravity of the connected domain in the second grayscale area and the center of gravity of the connected domain in the third grayscale area. If so, fuse the connected domain and fuse the grayscale areas where the connected domains are located.

[0022] Optionally, the fusion condition is: if the distance between the center points of two connected domains is within 10 pixels, then the two connected domains meet the fusion condition.

[0023] Optionally, step 3 includes:

[0024] Step 3.1. Obtain the centroid of each connected domain in the grayscale region in turn;

[0025] Step 3.2. Traverse the connected domain row by row upwards based on the center of gravity point to determine whether there is a connected domain of four fingers. If so, the connected domain is the connected domain that includes the palm.

[0026] Optionally, in step 3.1, the centroid of the connected domain within the grayscale area is obtained. If there is a connected domain formed by merging multiple connected domains, the average value of the centroids of the multiple connected domains is taken as the centroid of the fused connected domain.

[0027] Optionally, the method further includes correcting the binarization segmentation threshold, specifically by the following steps:

[0028] Step 4. Segment the palm vein image using a binary segmentation threshold to obtain an initial palm vein binary map, and calculate the initial palm area S1 in the palm vein binary map;

[0029] Step 5. Lower the binary segmentation threshold, re-segment the palm vein image using the lowered binary segmentation threshold to obtain a corrected palm vein binary map, calculate the palm area S2 and the position of the corrected center of gravity in the corrected palm vein binary map, and determine whether the binary segmentation threshold needs to be further lowered based on S2 and the position of the corrected center of gravity.

[0030] Optionally, in step 5, whether the binarization segmentation threshold needs to be further lowered is determined based on S2 and the position of the corrected center of gravity: if S2 is less than 1.5S1 and the pixel point corresponding to the corrected center of gravity in the initial palm vein binary image is in the palm area of ​​the initial palm vein binary image, then the initial binarization segmentation threshold needs to be further lowered.

[0031] In a second aspect, the present disclosure further relates to a palm vein recognition method, which comprises the following steps:

[0032] Step 10. Acquire a palm vein image of the target user, and calculate a binarization segmentation threshold of the palm vein image, wherein the binarization segmentation threshold is calculated using the palm vein image binarization method according to any one of claims 1 to 10;

[0033] Step 20. Binarize the palm vein image according to the binarization segmentation threshold to obtain a palm vein binary map;

[0034] Step 30. Obtain key feature points of the target user's palm based on the palm vein binary image;

[0035] Step 40. Acquire a target palm area of ​​the target user from the palm vein image based on the palm key feature points;

[0036] Step 50: Identify the target user based on the palm vein features of the target palm area and the template vein features.

[0037] Optionally, step 40 specifically includes:

[0038] Step 41. Correcting the palm vein image according to the palm key feature points;

[0039] Step 42: Acquire the target palm area of ​​the target user from the corrected palm vein image.

[0040] In a third aspect, the present disclosure further relates to a palm vein image binarization device, which includes the following modules:

[0041] a grayscale range acquisition module configured to acquire a grayscale range of the palm vein image and divide the palm vein image into a first grayscale level region, a second grayscale level region, a third grayscale level region, and a fourth grayscale level region according to the grayscale range;

[0042] A connected domain acquisition module is configured to sequentially acquire the connected domain of each grayscale region, calculate the centroid of each connected domain, and determine whether the connected domain needs to be fused and corrected based on the centroid of the connected domain;

[0043] The segmentation threshold determination module is configured to detect the connected domain of each grayscale area in turn, determine the connected domain containing the palm, and use the minimum grayscale value of the grayscale area where the connected domain containing the palm is located as the binary segmentation threshold of the palm vein image.

[0044] In a fourth aspect, the present disclosure further relates to a palm vein recognition device, which includes the following modules:

[0045] an image acquisition module configured to acquire a palm vein image of a target user and calculate a binarization segmentation threshold of the palm vein image, wherein the binarization segmentation threshold is calculated using the palm vein image binarization method according to any one of the first aspects;

[0046] A binarization module configured to perform binarization processing on the palm vein image according to the binarization segmentation threshold to obtain a palm vein binary map;

[0047] a feature point extraction module configured to obtain key feature points of the palm of the target user based on the palm vein binary image;

[0048] a target area extraction module configured to obtain a target area of ​​the palm of the target user from the palm vein image according to the key feature points of the palm;

[0049] The identity recognition module is configured to perform identity recognition on the target user based on the palm vein features and the template vein features of the target palm area.

[0050] In a fifth aspect, the present disclosure also relates to an electronic device, comprising: a processor, a storage medium, and a bus, wherein the storage medium stores program instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate through the bus, and the processor executes the program instructions to perform the steps of the palm vein image binarization method as described in any one of the first aspects, or the steps of the palm vein recognition method as described in any one of the second aspects.

[0051] In a sixth aspect, the present disclosure also relates to a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the palm vein image binarization method as described in any one of the first aspects, or the steps of the palm vein recognition method as described in any one of the second aspects are executed.

[0052] Compared with the existing technology, the present disclosure first divides the collected palm vein image into grayscale areas, then determines the grayscale area where the palm is located, and uses the minimum grayscale value of the grayscale area as the segmentation threshold, rather than directly determining the binary segmentation threshold based on the entire vein image. This can effectively solve the problem of unsatisfactory binary segmentation effect of palm vein images collected in complex outdoor environments, and further improve the accuracy of identity recognition based on palm vein images based on the binary segmentation threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] FIG1 is a flow chart of a palm vein image binarization method according to the present disclosure;

[0054] FIG2 is a flow chart of a palm vein recognition method according to the present disclosure;

[0055] FIG3 is a schematic structural diagram of a palm vein image binarization device according to the present disclosure;

[0056] FIG4 is a schematic structural diagram of a palm vein recognition device according to the present disclosure;

[0057] FIG5 is a schematic diagram of an electronic device involved in the present disclosure. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, the present disclosure is described in detail below with reference to embodiments and drawings, but the protection scope of the present disclosure is not limited thereto.

[0059] Referring to FIG. 1 , the present disclosure relates to a palm vein image binarization method, comprising the following steps:

[0060] Step 1. Obtain the grayscale range of the palm vein image, and divide the palm vein image into a first grayscale region, a second grayscale region, a third grayscale region, and a fourth grayscale region according to the grayscale range. The specific steps are as follows:

[0061] Step 1.1. Obtain the grayscale range of the palm vein image, that is, obtain the maximum grayscale value maxV and the minimum grayscale value minV of the palm vein image;

[0062] Step 1.2. Calculate the grayscale width range using the formula: range = int((maxV - minV) / 4);

[0063] Step 1.3. Divide the palm vein image into a first grayscale area, a second grayscale area, a third grayscale area, and a fourth grayscale area according to the grayscale width range, wherein the grayscale interval of the first grayscale area is [minV, minV+range), the grayscale interval of the second grayscale area is [minV+range, minV+range*2), the grayscale interval of the third grayscale area is [minV+range*2, minV+range*3), and the grayscale interval of the fourth grayscale area is [minV+range*3, maxV].

[0064] Step 2. Eliminate the first grayscale area, obtain the connected domain of each grayscale area in turn, calculate the center of gravity of each connected domain, and determine whether the connected domain needs to be fused and corrected based on the center of gravity of the connected domain; because the grayscale of the first grayscale area is low, that is, the image is darker, a large number of experiments have shown that the grayscale value of this grayscale area cannot be the segmentation threshold, so it is eliminated in this solution.

[0065] Step 2.1. Obtain all connected domains in the fourth grayscale region and calculate the centroid of each connected domain in turn;

[0066] Step 2.2. Obtain all connected domains in the third grayscale region and calculate the centroid of each connected domain in turn. Determine whether the connected domains meet the fusion conditions based on the centroid of the connected domains in the third grayscale region and the centroid of the connected domains in the fourth grayscale region. If so, fuse the connected domains and the grayscale regions where the connected domains are located.

[0067] Step 2.3. Obtain all connected domains in the second grayscale area and calculate the center of gravity of each connected domain in turn. Determine whether the connected domain meets the fusion condition based on the center of gravity of the connected domain in the second grayscale area and the center of gravity of the connected domain in the third grayscale area. If so, fuse the connected domains and fuse the grayscale areas where the connected domains are located. If the distance between the center of gravity of two connected domains is within 10 pixels, then the two connected domains meet the fusion condition. If the distance between the center of gravity of two connected domains is within 10 pixels, it indicates that the two connected domains are close or overlapping, so they are fused to facilitate the subsequent segmentation of the complete palm area.

[0068] Step 3. Detect the connected domains of each grayscale region in turn to determine the connected domain containing the palm, and use the minimum grayscale value of the grayscale region containing the connected domain containing the palm as the binary segmentation threshold of the palm vein image. The specific steps are as follows:

[0069] Step 3.1. Obtain the centroids of the connected domains within the grayscale region in sequence. If a connected domain exists that is fused from multiple connected domains, take the average of the centroids of the multiple connected domains as the centroid of the fused connected domain.

[0070] Step 3.2. Traverse the connected domain row by row upwards based on the center of gravity point to determine whether there is a connected domain of four fingers. If so, the connected domain is the connected domain that includes the palm.

[0071] Step 4. Segment the palm vein image using a binary segmentation threshold to obtain an initial palm vein binary map, and calculate the initial palm area S1 in the palm vein binary map;

[0072] Step 5. Lower the binarization threshold, re-segment the palm vein image using the lowered binarization threshold to obtain a corrected palm vein binary map, calculate the palm area S2 and the position of the corrected center of gravity in the corrected palm vein binary map, and determine whether the binarization threshold needs to be further lowered based on S2 and the position of the corrected center of gravity. The specific method is: if S2 is less than 1.5S1 and the pixel corresponding to the corrected center of gravity in the initial palm vein binary map is in the palm area of ​​the initial palm vein binary map, then the initial binarization threshold needs to be further lowered.

[0073] The above-mentioned palm vein image binarization method first divides the collected palm vein image into grayscale regions, then determines the grayscale region where the palm is located, and uses the minimum grayscale value of this grayscale region as the segmentation threshold, rather than directly determining the binarization segmentation threshold based on the entire vein image. This can effectively solve the problem of unsatisfactory binary segmentation effect of palm vein images collected in complex outdoor environments.

[0074] Referring to FIG. 2 , the present disclosure also relates to a palm vein recognition method, comprising the following steps:

[0075] Step 10. Capture a palm vein image of the target user and calculate a binarization segmentation threshold for the palm vein image, wherein the binarization segmentation threshold is calculated using the palm vein image binarization method described above;

[0076] Step 20. Binarize the palm vein image according to the binary segmentation threshold to obtain a palm vein binary map;

[0077] Step 30. Obtain key feature points of the target user's palm based on the palm vein binary image;

[0078] Step 40. Obtain the target palm area of ​​the target user from the palm vein image based on the key palm feature points;

[0079] Step 50: Identify the target user based on the palm vein features of the target palm area and the template vein features.

[0080] In this embodiment, the preset electronic device has the function of identifying the user through the palm vein. The preset electronic device has an image acquisition module, such as a camera. The preset electronic device can be, for example, a smart mobile terminal, a time clock, an access control device, etc.

[0081] The palm vein image of the target user to be identified is used by the image acquisition module, and the above-mentioned palm vein image binarization method is used to obtain a binary segmentation threshold of the palm vein image of the target user. The palm vein image of the target user is binarized and segmented based on the binarization threshold to obtain a palm vein binary map of the target user.

[0082] Key point recognition is performed on the palm vein binary image to determine the palm key feature points that constitute the user's palm. Based on the position of the palm key feature points in the palm vein binary image, the palm target area is intercepted from the corresponding position of the palm vein image, and palm vein features are extracted from the palm target area to determine the palm vein features of the target user.

[0083] At least one template vein feature is pre-stored or recorded in a preset electronic device, and the user corresponding to the template vein feature is a safe user or a trusted user. The palm vein feature extracted from the palm vein image is compared with the at least one template vein feature to determine whether there is a vein template feature consistent with the palm vein feature in the at least one template vein feature. If so, it is determined that the identity recognition of the target user has been successful; if not, it is determined that the identity recognition of the target user has failed.

[0084] In some embodiments, each template vein feature has corresponding identity information. If there is a vein template feature consistent with the palm vein feature in at least one template vein feature, the identity information of the matched vein template feature is displayed.

[0085] Optionally, the specific steps of step 40 are:

[0086] Step 41: Correct the palm vein image based on the key feature points of the palm;

[0087] Step 42: Acquire the target palm area of ​​the target user from the rectified palm vein image.

[0088] In this embodiment, the palm vein image is rotationally corrected based on the key feature points of the palm so that the palm vein image does not have problems such as tilt and offset. A region of interest (ROI), i.e., a target area on the palm of the target user, is captured from the corrected palm vein image.

[0089] The above palm vein recognition method has an ideal effect of performing binary segmentation on the palm vein image. Therefore, performing user identity recognition based on the binary segmentation result can improve the accuracy of identity recognition.

[0090] Referring to FIG. 3 , the present disclosure further relates to a palm vein image binarization device, comprising the following modules:

[0091] a grayscale range acquisition module 11 configured to acquire the grayscale range of the palm vein image and divide the palm vein image into a first grayscale level region, a second grayscale level region, a third grayscale level region, and a fourth grayscale level region according to the grayscale range;

[0092] The connected domain acquisition module 12 is configured to sequentially acquire the connected domain of each grayscale region, calculate the centroid of each connected domain, and determine whether the connected domain needs to be fused and corrected based on the centroid of the connected domain;

[0093] The segmentation threshold determination module 13 is configured to detect the connected domain of each grayscale area in turn, determine the connected domain containing the palm, and use the minimum grayscale value of the grayscale area where the connected domain containing the palm is located as the binary segmentation threshold of the palm vein image.

[0094] Optionally, the grayscale range acquisition module 11 is specifically configured to acquire the grayscale range of the palm vein image, that is, to acquire the maximum grayscale value maxV and the minimum grayscale value minV of the palm vein image; calculate the grayscale width range, using the formula: range = int((maxV-minV) / 4); and divide the palm vein image into a first grayscale area, a second grayscale area, a third grayscale area, and a fourth grayscale area according to the grayscale width range.

[0095] Optionally, the grayscale interval of the first grayscale area is [minV, minV+range), the grayscale interval of the second grayscale area is [minV+range, minV+range*2), the grayscale interval of the third grayscale area is [minV+range*2, minV+range*3), and the grayscale interval of the fourth grayscale area is [minV+range*3, maxV].

[0096] Optionally, the grayscale range acquisition module 11 is further configured to eliminate the first grayscale level area.

[0097] Optionally, the connected domain acquisition module 12 is specifically configured to obtain all connected domains in the fourth grayscale area, and calculate the center of gravity of each connected domain in turn; obtain all connected domains in the third grayscale area, and calculate the center of gravity of each connected domain in turn, and judge whether the connected domains meet the fusion conditions according to the center of gravity of the connected domains in the third grayscale area and the center of gravity of the connected domains in the fourth grayscale area; if so, the connected domains are fused, and the grayscale areas where the connected domains are located are fused; obtain all connected domains in the second grayscale area, and calculate the center of gravity of each connected domain in turn, and judge whether the connected domains meet the fusion conditions according to the center of gravity of the connected domains in the second grayscale area and the center of gravity of the connected domains in the third grayscale area; if so, the connected domains are fused, and the grayscale areas where the connected domains are located are fused.

[0098] Optionally, the fusion condition is: if the distance between the center points of two connected domains is within 10 pixels, then the two connected domains meet the fusion condition.

[0099] Optionally, the segmentation threshold determination module 13 is specifically configured to obtain the centroid point of each connected domain in the grayscale area in turn; traverse the connected domain row by row upward based on the centroid point to determine whether there is a connected domain of 4 fingers. If so, the connected domain is a connected domain containing the palm.

[0100] Optionally, the segmentation threshold determination module 13 is further configured to take the average value of the centroids of the multiple connected domains as the centroid of the fused connected domain if there is a connected domain fused from multiple connected domains.

[0101] Optionally, the palm vein image binarization device further includes: a correction module configured to correct the binarization segmentation threshold, the correction module being specifically configured to segment the palm vein image using the binarization segmentation threshold to obtain an initial palm vein binary map, and calculate an initial area S1 of the palm in the palm vein binary map; lower the binarization segmentation threshold, re-segment the palm vein image using the lowered binarization segmentation threshold to obtain a corrected palm vein binary map, calculate the area S2 of the palm and the position of the corrected center of gravity in the corrected palm vein binary map, and determine whether the binarization segmentation threshold needs to be further lowered based on S2 and the position of the corrected center of gravity.

[0102] Optionally, the correction module is further configured to further reduce the initial binary segmentation threshold if S2<1.5S1 and the pixel point corresponding to the correction center point in the initial palm vein binary image is in the palm area of ​​the initial palm vein binary image.

[0103] Referring to FIG. 4 , the present disclosure also relates to a palm vein recognition device, comprising the following modules:

[0104] The image acquisition module 21 is configured to acquire a palm vein image of a target user and calculate a binarization segmentation threshold of the palm vein image, wherein the binarization segmentation threshold is calculated using the palm vein image binarization method described above;

[0105] A binarization module 22 is configured to perform binarization processing on the palm vein image according to a binarization segmentation threshold value to obtain a palm vein binary map;

[0106] A feature point extraction module 23 is configured to obtain key feature points of the target user's palm based on the palm vein binary image;

[0107] a target region extraction module 24 configured to obtain a target region of the palm of the target user from the palm vein image based on key feature points of the palm;

[0108] The identity recognition module 25 is configured to perform identity recognition on the target user based on the palm vein features of the target palm area and the template vein features.

[0109] Optionally, the target area extraction module 24 is specifically configured to correct the palm vein image according to the key feature points of the palm; and obtain the target area of ​​the palm of the target user from the corrected palm vein image.

[0110] 5 , the present disclosure further relates to an electronic device, which includes a processor 31, a storage medium 32, and a bus. The storage medium 32 stores program instructions executable by the processor 31. When the electronic device 30 is running, the processor 31 communicates with the storage medium 32 via the bus, and the processor 31 executes the program instructions to perform the steps of the above-mentioned palm vein image binarization method or the steps of the palm vein recognition method.

[0111] In one possible implementation, the present disclosure further relates to a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program executes the steps of the above-mentioned palm vein image binarization method or the above-mentioned palm vein recognition method.

[0112] The embodiments described above are merely descriptions of preferred embodiments of the present disclosure and are not intended to limit the scope of the present disclosure. Without departing from the design spirit of the present disclosure, various modifications and improvements made to the technical solutions of the present disclosure by ordinary technicians in this field should fall within the scope of protection determined by the claims of the present disclosure. Industrial Applicability

[0113] Using the above scheme, the collected palm vein image is first divided into grayscale areas, and then the grayscale area where the palm is located is determined, and the minimum grayscale value of the grayscale area is used as the segmentation threshold, rather than directly determining the binary segmentation threshold based on the entire vein image. This can effectively solve the problem of unsatisfactory binary segmentation effect of palm vein images collected in complex outdoor environments, and further improve the accuracy of identity recognition based on palm vein images based on the binary segmentation threshold.

Claims

1. A palm vein image binarization method, characterized in that: The following steps are involved: Step 1. Obtaining a grayscale range of a palm vein image, and dividing the palm vein image into a first grayscale region, a second grayscale region, a third grayscale region, and a fourth grayscale region according to the grayscale range; Step 2. Obtain the connected domain of each grayscale area in turn, calculate the centroid of each connected domain, and determine whether the connected domain needs to be fused and corrected based on the centroid of the connected domain; Step 3. Detect the connected domain of each grayscale area in turn, determine the connected domain containing the palm, and use the minimum grayscale value of the grayscale area where the connected domain containing the palm is located as the binary segmentation threshold of the palm vein image.

2. The palm vein image binarization method according to claim 1, characterized in that: The step 1 specifically includes: Step 1.

1. Obtain the grayscale range of the palm vein image, that is, obtain the maximum grayscale value maxV and the minimum grayscale value minV of the palm vein image; Step 1.

2. Calculate the width range of the grayscale level using the formula: range = int((maxV - minV) / 4); Step 1.

3. Divide the palm vein image into a first grayscale area, a second grayscale area, a third grayscale area and a fourth grayscale area according to the grayscale width range.

3. The palm vein image binarization method according to claim 2, characterized in that: The grayscale interval of the first grayscale area is [minV, minV+range), the grayscale interval of the second grayscale area is [minV+range, minV+range*2), the grayscale interval of the third grayscale area is [minV+range*2, minV+range*3), and the grayscale interval of the fourth grayscale area is [minV+range*3, maxV].

4. The palm vein image binarization method according to claim 1, characterized in that: Before performing step 2, the first grayscale area is eliminated.

5. The palm vein image binarization method according to claim 4, characterized in that: The step 2 specifically includes: Step 2.

1. Obtain all connected domains in the fourth grayscale area, and calculate the centroid of each connected domain in turn; Step 2.

2. Obtain all connected domains in the third grayscale region, and calculate the centroid of each connected domain in turn, and determine whether the connected domain meets the fusion condition based on the centroid of the connected domain in the third grayscale region and the centroid of the connected domain in the fourth grayscale region. If so, merge the connected domains and merge the grayscale regions where the connected domains are located; Step 2.

3. Obtain all connected domains of the second grayscale area, and calculate the centroid of each connected domain in turn, and determine whether the connected domain meets the fusion conditions based on the centroid of the connected domain of the second grayscale area and the centroid of the connected domain of the third grayscale area. If so, merge the connected domains and merge the grayscale areas where the connected domains are located.

6. The palm vein image binarization method according to claim 5, characterized in that: The fusion condition is: if the distance between the center points of two connected domains is within 10 pixels, the two connected domains meet the fusion condition.

7. The palm vein image binarization method according to claim 5, characterized in that: The step 3 comprises: Step 3.

1. Obtain the centroid of each connected domain in the grayscale region in turn; Step 3.

2. Using the center of gravity as a reference, traverse the connected domain row by row upwards to determine whether there is a connected domain of four fingers. If so, the connected domain is the connected domain that includes the palm.

8. The palm vein image binarization method according to claim 7, characterized in that: In the step 3.1, the centroid of the connected domain in the grayscale region is obtained. If there is a connected domain formed by merging multiple connected domains, the average value of the centroids of the multiple connected domains is taken as the centroid of the fused connected domain.

9. The palm vein image binarization method according to claim 1, characterized in that: The method further includes correcting the binary segmentation threshold, the specific steps of which are: Step 4. Segment the palm vein image by using a binary segmentation threshold to obtain an initial palm vein binary map, and calculate the initial area S1 of the palm in the palm vein binary map; Step 5. Lower the binary segmentation threshold, re-segment the palm vein image using the lowered binary segmentation threshold to obtain a corrected palm vein binary image, calculate the palm area S2 and the position of the corrected center of gravity in the corrected palm vein binary image, and determine whether the binary segmentation threshold needs to be further lowered based on S2 and the position of the corrected center of gravity.

10. The palm vein image binarization method according to claim 6, characterized in that: In step 5, whether the binary segmentation threshold needs to be further lowered is judged according to S2 and the position of the correction center of gravity: if S2<1.5S1 and the pixel point corresponding to the correction center of gravity in the initial palm vein binary map is in the palm area of ​​the initial palm vein binary map, then the initial binary segmentation threshold needs to be further lowered.

11. A palm vein recognition method, characterized in that: The following steps are involved: Step 10. Acquire a palm vein image of the target user, and calculate a binarization segmentation threshold of the palm vein image, wherein the binarization segmentation threshold is calculated using the palm vein image binarization method according to any one of claims 1 to 10; Step 20. Binarize the palm vein image according to the binary segmentation threshold to obtain a palm vein binary map; Step 30. Obtain key feature points of the palm of the target user according to the palm vein binary image; Step 40. Acquire a palm target area of ​​the target user from the palm vein image according to the palm key feature points; Step 50: Identify the target user based on the palm vein features and template vein features of the target palm area.

12. The method according to claim 11, characterized in that The step 40 specifically includes: Step 41. Correcting the palm vein image according to the palm key feature points; Step 42: Acquire the target palm area of ​​the target user from the corrected palm vein image.

13. A palm vein image binarization device, characterized in that: Includes the following modules: a grayscale range acquisition module configured to acquire the grayscale range of the palm vein image, and divide the palm vein image into a first grayscale level area, a second grayscale level area, a third grayscale level area, and a fourth grayscale level area according to the grayscale range; A connected domain acquisition module is configured to sequentially acquire the connected domain of each grayscale region, calculate the centroid of each connected domain, and determine whether the connected domain needs to be fused and corrected according to the centroid of the connected domain; The segmentation threshold determination module is configured to detect the connected domain of each grayscale area in turn, determine the connected domain containing the palm, and use the minimum grayscale value of the grayscale area where the connected domain containing the palm is located as the binary segmentation threshold of the palm vein image.

14. A palm vein recognition device, characterized in that: Includes the following modules: an image acquisition module, configured to acquire a palm vein image of a target user and calculate a binarization segmentation threshold of the palm vein image, wherein the binarization segmentation threshold is calculated using the palm vein image binarization method according to any one of claims 1 to 10; A binarization module, configured to perform binarization processing on the palm vein image according to the binarization segmentation threshold to obtain a palm vein binary map; A feature point extraction module, configured to obtain key feature points of the palm of the target user according to the palm vein binary image; A target area extraction module, configured to obtain a palm target area of ​​the target user from the palm vein image according to the palm key feature points; The identity recognition module is configured to perform identity recognition on the target user according to the palm vein features and the template vein features of the palm target area.

15. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores program instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the program instructions to perform the steps of the palm vein image binarization method according to any one of claims 1 to 10, or the steps of the palm vein recognition method according to any one of claims 11 to 12.

16. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the palm vein image binarization method according to any one of claims 1 to 10, or the steps of the palm vein recognition method according to any one of claims 11 to 12 are executed.

Images