Finger vein image preprocessing method and system based on super-resolution algorithm

Abstract

The invention provides a finger vein image preprocessing method and system based on a super-resolution algorithm, and the method comprises the steps: obtaining an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image; performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; performing finger rotation alignment and finger contour boundary extraction based on the high-contrast image to obtain a finger contour image; vein region registration is carried out according to the finger contour image, a gray level image is generated, binarization processing is carried out on the gray level image, and finger vein image preprocessing is completed. The problem that the existing finger vein image recognition precision is low is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a finger vein image preprocessing method and system based on a super-resolution algorithm. Background Art

[0002] Finger vein recognition technology is a type of biometrics and has a wide range of application scenarios. The preprocessing of the captured image plays a key role in finger vein recognition because of the image resolution limitations brought by the acquisition equipment. High-resolution images make recognition more discerning, and finger vein recognition requires operations such as finger joint alignment during the preprocessing process. Finger vein recognition technology includes image acquisition, image preprocessing, feature extraction, and matching algorithms. The image acquisition process is affected by the surrounding environment, and it is easy to have problems such as unclear captured images and large differences in image brightness, which affect the recognition accuracy, so image preprocessing is needed. Among them, common algorithms for finger vein image preprocessing include image enhancement methods, common processing methods include increasing contrast and texture enhancement, etc., but small veins are often missing, and there are few related studies.

[0003] In the prior art, during the image acquisition process, due to the influence of the environment and acquisition equipment, finger vein images often contain a variety of noises, and for small veins, the edge processing is not refined enough. In subsequent applications such as segmentation, the burrs and protrusions in the vein pattern will affect the model performance, and it is not conducive to the processing of vein information in recognition applications. The existing pre-processing image enhancement method for local image degradation blur processing is based on contrast and texture distribution. Traditional image processing methods can be further improved by combining machine learning methods. Summary of the invention

[0004] The present invention provides a finger vein image preprocessing method and system based on a super-resolution algorithm, which are used to solve the problem of low recognition accuracy of existing finger vein images.

[0005] The present invention provides a finger vein image preprocessing method based on a super-resolution algorithm, comprising: Acquire an original finger vein image, input the original finger vein image into a preset super-resolution model, and output a high-resolution finger vein image; Performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; Perform finger rotation alignment and finger contour boundary extraction based on the high contrast image to obtain a finger contour image; The vein area is registered according to the finger contour image to generate a grayscale image, and the grayscale image is binarized to complete the finger vein image preprocessing.

[0006] According to a finger vein image preprocessing method based on a super-resolution algorithm provided by the present invention, the original finger vein image is obtained, the original finger vein image is input into a preset super-resolution model, and a high-resolution finger vein image is output, which specifically includes: The original finger vein image is input into a preset super-resolution model, and convolution and regularization are alternately performed through the low-frequency sub-network of the super-resolution model to output a finger vein structure feature map and its feature value of the same resolution; Based on the finger vein structure feature map, a network composed of three residual blocks in a super-resolution model is used to extract the detail part of the finger vein structure feature map to obtain detail feature value data; The feature values ​​of the finger vein structure feature map and the detail feature value data are processed by the super-resolution model to obtain a high-resolution finger vein image.

[0007] According to a finger vein image preprocessing method based on a super-resolution algorithm provided by the present invention, the feature value based on the finger vein structure feature map and the detail feature value data are processed by the super-resolution model to obtain a high-resolution finger vein image, which specifically includes: Performing a weighted summation of the characteristic value of the finger vein structure characteristic graph and the detail characteristic value data to form a complete finger vein characteristic value set; The complete finger vein feature values ​​are mapped into multiple candidate images of the original resolution size through a single-layer convolutional network in the super-resolution model; Arrange and combine multiple candidate images of original resolution sizes, and select the image with the highest definition as the output high-resolution finger vein image.

[0008] According to a finger vein image preprocessing method based on a super-resolution algorithm provided by the present invention, the contrast enhancement is performed based on the high-resolution finger vein image to generate a high-contrast image, specifically comprising: Performing histogram equalization processing on the high-resolution finger vein image to generate a balanced image; The fuzzy processing is performed based on the equalized image, the overall grayscale distribution characteristics of the image are retained, the size of the image Gaussian function is changed, the noise level of the image is adjusted, and a high-contrast image is generated.

[0009] According to a finger vein image preprocessing method based on a super-resolution algorithm provided by the present invention, the finger rotation alignment and finger contour boundary extraction are performed based on the high-contrast image to obtain a finger contour image, which specifically includes: Based on the edge extraction of the high contrast image chassis, the points on the upper and lower boundaries obtained by the edge extraction part are averaged, and the vector of the normalized direction is fitted by the least square method to generate a transformation matrix; Input the transformation matrix and the high-resolution finger vein image into the preset affine transformation function to complete the rotation alignment of the high-contrast image; Based on the high-contrast image after rotation alignment, the upper and lower contours are found, and the discontinuous pixels of the upper and lower contours are monitored. The pixels of the discontinuous part within the set threshold range are filled with lines to obtain the finger contour image.

[0010] According to a finger vein image preprocessing method based on a super-resolution algorithm provided by the present invention, the vein area is registered according to the finger contour image to generate a grayscale image, and the grayscale image is binarized to complete the finger vein image preprocessing, which specifically includes: Based on the finger contour image, determine the finger joint prominence points to perform vein area registration and generate a grayscale image; The grayscale image is binarized to be converted into a binary image with only white dots and black dots, and the background and the non-vein part of the finger image are removed to complete the finger vein image preprocessing.

[0011] The present invention also provides a finger vein image preprocessing system based on a super-resolution algorithm, the system comprising: A high-resolution processing module, used for acquiring an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image; A contrast enhancement module, used for performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; A contour acquisition module, used to perform finger rotation alignment and finger contour boundary extraction based on the high contrast image to obtain a finger contour image; The result output module is used to perform vein area registration according to the finger contour image, generate a grayscale image, perform binarization processing on the grayscale image, and complete finger vein image preprocessing.

[0012] The present invention also provides an electronic 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, a finger vein image preprocessing method based on a super-resolution algorithm as described above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for preprocessing finger vein images based on a super-resolution algorithm.

[0014] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for preprocessing finger vein images based on a super-resolution algorithm.

[0015] The present invention provides a finger vein image preprocessing method and system based on a super-resolution algorithm, which processes the original finger vein image into a high-resolution finger vein image, and then enhances the contrast and brightness distribution of the image, thereby making the vascular structure more clearly visible; in view of the rotation problem that may occur in the finger during the acquisition process, a specific finger rotation processing method is introduced to correct the rotation deviation in the image and ensure that the image is in the correct direction. Through multi-step optimization, not only the vascular details are clearly visible, but also the local missing problem can be effectively improved, which makes the subsequent finger vein feature extraction and matching process more accurate and reliable, significantly improves the performance of the finger vein recognition system, improves the accuracy of the recognition system, and reduces the false recognition rate, confirms its superiority and reliability in practical applications, and provides strong support for the further development of finger vein recognition technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a flow chart of the finger vein image preprocessing method based on super-resolution algorithm provided by the present invention.

[0018] Figure 2 It is a schematic diagram of the collection area provided by the present invention.

[0019] Figure 3 This is a flow chart of extracting image structural features by a low-frequency sub-network provided by the present invention.

[0020] Figure 4 It is a schematic diagram of the processing flow based on high-resolution finger vein images provided by the present invention.

[0021] Figure 5 This is the finger vein image after binarization provided by the present invention.

[0022] Figure 6 It is a schematic diagram of module connection of a finger vein image preprocessing system based on a super-resolution algorithm provided by the present invention.

[0023] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention.

[0024] Reference numerals: 110: high-resolution processing module; 120: contrast enhancement module; 130: contour acquisition module; 140: result output module; 710: processor; 720: communication interface; 730: memory; 740: communication bus. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Combine the following Figure 1 The present invention describes a finger vein image preprocessing method based on a super-resolution algorithm, comprising: step 100, obtaining an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image.

[0027] refer to Figure 2 In the present invention, the original finger vein image is input into a preset super-resolution model, and the low-frequency sub-network of the super-resolution model performs convolution and regularization alternately, and outputs a finger vein structure feature map and its feature value of the same resolution; wherein the low-frequency sub-network includes 4 convolution layers and 4 regularization layers. The convolution kernel size of each layer is 3*3 and the depth is 64.

[0028] refer to Figure 3 Based on the finger vein structure feature map, the details of the finger vein structure feature map are extracted through a network composed of three residual blocks in the super-resolution model to obtain detail feature value data; wherein, the high-frequency sub-network of the super-resolution model contains three residual blocks, each of which includes two 3*3 convolutional layers with a depth of 64 and a regularization layer. The input and output of each residual block are simultaneously used as the input of the next residual block, realizing the flow of information and the reuse of features. In each residual block, the current image is subjected to alternating convolution and regularization processing to generate a detail feature map of the output image.

[0029] The feature values ​​of the finger vein structure feature map and the detail feature value data are processed by the super-resolution model to obtain a high-resolution finger vein image.

[0030] The process of obtaining a high-resolution finger vein image based on the feature value of the finger vein structure feature map and the detail feature value data through the super-resolution model specifically includes: Performing a weighted summation of the characteristic value of the finger vein structure characteristic graph and the detail characteristic value data to form a complete finger vein characteristic value set; The complete finger vein feature values ​​are mapped into r×r candidate images of the original resolution size through a single-layer convolutional network in the super-resolution model; The r×r candidate images of the original resolution are arranged and combined, and the image with the highest definition is selected as the output high-resolution finger vein image.

[0031] By converting the original finger vein image into a high-resolution finger vein image, it helps in subsequent finger vein recognition and improves the recognition accuracy.

[0032] Step 200: Perform contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image.

[0033] Specifically, the method comprises: performing histogram equalization processing on the high-resolution finger vein image to generate a balanced image; The fuzzy processing is performed based on the equalized image, the overall grayscale distribution characteristics of the image are retained, the size of the image Gaussian function is changed, the noise level of the image is adjusted, and a high-contrast image is generated.

[0034] refer to Figure 4 , because the finger vein collector is affected by the light intensity when collecting images, the grayscale distribution of the finger vein image is relatively concentrated, and the image contrast is low, so the grayscale difference between the vein area and the surrounding area is not obvious. Therefore, the vein image is processed by histogram equalization to extract finger veins with better continuity and accuracy. The image is blurred to reduce the sharp changes at the black and white boundaries of the image. The overall grayscale distribution characteristics of the image are retained to reduce the noise of the image. The degree of noise in the image is adjusted by changing the kernel size of the Gaussian function. The larger the kernel size, the blurrier the image, the blurrier the boundaries, and the smaller the noise effect.

[0035] Histogram equalization is applied to the original finger vein image to enhance the contrast and brightness distribution of the image, making the vascular structure more visible. Then, Gaussian filtering technology is used to reduce noise and interference in the image, making the subsequent feature extraction and analysis more accurate and reliable.

[0036] Step 300: Perform finger rotation alignment and finger contour boundary extraction based on the high contrast image to obtain a finger contour image.

[0037] Specifically, it includes: extracting the inner edge of the high contrast image chassis, taking the average value of the points on the upper and lower boundaries obtained by the edge extraction part, fitting the vector of the normalized direction by the least square method, and generating a transformation matrix; Input the transformation matrix and the high-resolution finger vein image into the preset affine transformation function to complete the rotation alignment of the high-contrast image; Based on the high-contrast image after rotation alignment, the upper and lower contours are found, and the discontinuous pixels of the upper and lower contours are monitored. The pixels of the discontinuous part within the set threshold range are filled with lines to obtain the finger contour image.

[0038] In the present invention, during the finger rotation alignment process, three parameters of an image are first determined: center, rotation angle, and scaling ratio. Among them, the rotation angle is: the points on the upper and lower boundaries obtained by the edge extraction part are averaged, and the vector of the normalized direction is fitted by the least squares method, and finally a transformation matrix can be obtained. Referring to the following formula, the transformation matrix and the original image are input into the affine transformation function to complete the rotation of the image.

[0039] .

[0040] Aiming at the possible rotation problem of fingers during the acquisition process, a specific finger rotation processing method is introduced to correct the rotation deviation in the image and ensure that the image is in the correct direction.

[0041] Based on the high-contrast image after rotation alignment, the pixel points obtained by the storage operator, white: 1, black: 0, find the upper and lower contours, monitor the discontinuous pixels in the upper and lower parts, and if there is only one pixel difference, use line completion to obtain the finger contour image.

[0042] Step 400: perform vein area registration according to the finger contour image to generate a grayscale image, perform binarization processing on the grayscale image, and complete finger vein image preprocessing.

[0043] Specifically, it includes: determining the finger joint prominence points based on the finger contour image to perform vein area registration and generate a grayscale image; The grayscale image is binarized to be converted into a binary image with only white dots and black dots, and the background and the non-vein part of the finger image are removed to complete the finger vein image preprocessing.

[0044] In the present invention, when recording the veins, the upper and lower positions of the fingers on the instrument are uncertain. It may be biased upward or downward. Because the shape of human fingers is fixed, it is only necessary to extract the contour to complete the upper and lower registration. When recording the veins, the left and right positions of the fingers on the instrument are uncertain. It may be biased to the left or to the right. There are currently two methods: 1. Through the protruding points of the second joint and the first joint of the human finger. Thereby finding the positions of the first and second joints of each person. When processing each image, translate this joint to a specific position. 2. According to the change in curvature, find the second section of the finger.

[0045] refer to Figure 5 When binarizing the image, the original grayscale image is converted into a binary image with only white and black dots. Because the subsequent matching method is template matching, this binarization method can only retain the veins in the image and remove the influence of the background and non-vein parts on the finger. This increases the matching accuracy.

[0046] A finger vein image preprocessing method based on a super-resolution algorithm provided by the present invention processes the original finger vein image into a high-resolution finger vein image, and then enhances the contrast and brightness distribution of the image, so that the vascular structure is more clearly visible; in view of the rotation problem that may occur in the finger during the acquisition process, a specific finger rotation processing method is introduced to correct the rotation deviation in the image and ensure that the image is in the correct direction. Through multi-step optimization, not only the vascular details are clearly visible, but also the local missing problem can be effectively improved, which makes the subsequent finger vein feature extraction and matching process more accurate and reliable, significantly improves the performance of the finger vein recognition system, improves the accuracy of the recognition system, and reduces the misrecognition rate, confirms its superiority and reliability in practical applications, and provides strong support for the further development of finger vein recognition technology.

[0047] refer to Figure 6 The present invention also discloses a finger vein image preprocessing system based on a super-resolution algorithm, the system comprising: The high-resolution processing module 110 is used to obtain an original finger vein image, input the original finger vein image into a preset super-resolution model, and output a high-resolution finger vein image; A contrast enhancement module 120, configured to perform contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; A contour acquisition module 130, configured to perform finger rotation alignment and finger contour boundary extraction based on the high contrast image to obtain a finger contour image; The result output module 140 is used to perform vein area registration according to the finger contour image, generate a grayscale image, perform binarization processing on the grayscale image, and complete finger vein image preprocessing.

[0048] The method of obtaining an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image specifically includes: The original finger vein image is input into a preset super-resolution model, and convolution and regularization are alternately performed through the low-frequency sub-network of the super-resolution model to output a finger vein structure feature map and its feature value of the same resolution; Based on the finger vein structure feature map, a network composed of three residual blocks in a super-resolution model is used to extract the detail part of the finger vein structure feature map to obtain detail feature value data; The feature values ​​of the finger vein structure feature map and the detail feature value data are processed by the super-resolution model to obtain a high-resolution finger vein image.

[0049] The high-resolution finger vein image is obtained based on the feature value of the finger vein structure feature map and the detail feature value data through the super-resolution model, specifically including: Performing a weighted summation of the characteristic value of the finger vein structure characteristic graph and the detail characteristic value data to form a complete finger vein characteristic value set; The complete finger vein feature values ​​are mapped into multiple candidate images of the original resolution size through a single-layer convolutional network in the super-resolution model; Arrange and combine multiple candidate images of original resolution sizes, and select the image with the highest definition as the output high-resolution finger vein image.

[0050] Performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image specifically includes: Performing histogram equalization processing on the high-resolution finger vein image to generate a balanced image; The fuzzy processing is performed based on the equalized image, the overall grayscale distribution characteristics of the image are retained, the size of the image Gaussian function is changed, the noise level of the image is adjusted, and a high-contrast image is generated.

[0051] Finger rotation alignment and finger contour boundary extraction are performed based on the high contrast image to obtain a finger contour image, specifically including: Based on the edge extraction of the high contrast image chassis, the points on the upper and lower boundaries obtained by the edge extraction part are averaged, and the vector of the normalized direction is fitted by the least square method to generate a transformation matrix; Input the transformation matrix and the high-resolution finger vein image into the preset affine transformation function to complete the rotation alignment of the high-contrast image; Based on the high-contrast image after rotation alignment, the upper and lower contours are found, and the discontinuous pixels of the upper and lower contours are monitored. The pixels of the discontinuous part within the set threshold range are filled with lines to obtain the finger contour image.

[0052] Performing vein area registration according to the finger contour image to generate a grayscale image, performing binarization processing on the grayscale image, and completing finger vein image preprocessing specifically includes: Based on the finger contour image, determine the finger joint prominence points to perform vein area registration and generate a grayscale image; The grayscale image is binarized to be converted into a binary image with only white dots and black dots, and the background and the non-vein part of the finger image are removed to complete the finger vein image preprocessing.

[0053] A finger vein image preprocessing system based on a super-resolution algorithm provided by the present invention processes the original finger vein image into a high-resolution finger vein image, and then enhances the contrast and brightness distribution of the image, so that the vascular structure is more clearly visible; in view of the rotation problem that may occur in the finger during the acquisition process, a specific finger rotation processing method is introduced to correct the rotation deviation in the image and ensure that the image is in the correct direction. Through multi-step optimization, not only the vascular details are clearly visible, but also the local missing problem can be effectively improved, which makes the subsequent finger vein feature extraction and matching process more accurate and reliable, significantly improves the performance of the finger vein recognition system, improves the accuracy of the recognition system, and reduces the misrecognition rate, confirms its superiority and reliability in practical applications, and provides strong support for the further development of finger vein recognition technology.

[0054] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute a finger vein image preprocessing method based on a super-resolution algorithm, the method comprising: acquiring an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image; performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; performing finger rotation alignment and finger contour boundary extraction based on the high-contrast image to obtain a finger contour image; performing vein area registration according to the finger contour image to generate a grayscale image, performing binarization processing on the grayscale image, and completing finger vein image preprocessing.

[0055] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0056] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a finger vein image preprocessing method based on a super-resolution algorithm provided by the above methods, the method including: obtaining an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image; performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; performing finger rotation alignment and finger contour boundary extraction based on the high-contrast image to obtain a finger contour image; performing vein area registration according to the finger contour image to generate a grayscale image, binarizing the grayscale image, and completing finger vein image preprocessing.

[0057] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute a finger vein image preprocessing method based on a super-resolution algorithm provided by the above-mentioned methods, the method comprising: acquiring an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image; performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; performing finger rotation alignment and finger contour boundary extraction based on the high-contrast image to obtain a finger contour image; performing vein area registration according to the finger contour image to generate a grayscale image, binarizing the grayscale image, and completing finger vein image preprocessing.

[0058] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0059] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these 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 invention.

Claims

1. A finger vein image preprocessing method based on super-resolution algorithm, characterized in that: include: Acquire an original finger vein image, input the original finger vein image into a preset super-resolution model, and output a high-resolution finger vein image; Performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; Perform finger rotation alignment and finger contour boundary extraction based on the high contrast image to obtain a finger contour image; The vein area is registered according to the finger contour image to generate a grayscale image, and the grayscale image is binarized to complete the finger vein image preprocessing.

2. The finger vein image preprocessing method based on super-resolution algorithm according to claim 1 is characterized in that: The obtaining of the original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image specifically includes: The original finger vein image is input into a preset super-resolution model, and convolution and regularization are alternately performed through the low-frequency sub-network of the super-resolution model to output a finger vein structure feature map and its feature value of the same resolution; Based on the finger vein structure feature map, a network composed of three residual blocks in a super-resolution model is used to extract the detail part of the finger vein structure feature map to obtain detail feature value data; The feature values ​​of the finger vein structure feature map and the detail feature value data are processed by the super-resolution model to obtain a high-resolution finger vein image.

3. The finger vein image preprocessing method based on super-resolution algorithm according to claim 2 is characterized in that: The feature value based on the finger vein structure feature map and the detail feature value data are processed by the super-resolution model to obtain a high-resolution finger vein image, specifically including: Performing a weighted summation of the characteristic value of the finger vein structure characteristic graph and the detail characteristic value data to form a complete finger vein characteristic value set; The complete finger vein feature values ​​are mapped into multiple candidate images of the original resolution size through a single-layer convolutional network in the super-resolution model; Arrange and combine multiple candidate images of original resolution sizes, and select the image with the highest definition as the output high-resolution finger vein image.

4. The finger vein image preprocessing method based on super-resolution algorithm according to claim 1, characterized in that: The step of performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image specifically includes: Performing histogram equalization processing on the high-resolution finger vein image to generate a balanced image; The fuzzy processing is performed based on the equalized image, the overall grayscale distribution characteristics of the image are retained, the size of the image Gaussian function is changed, the noise level of the image is adjusted, and a high-contrast image is generated.

5. The finger vein image preprocessing method based on super-resolution algorithm according to claim 1, characterized in that: The step of performing finger rotation alignment and finger contour boundary extraction based on the high contrast image to obtain a finger contour image specifically includes: Based on the edge extraction of the high contrast image chassis, the points on the upper and lower boundaries obtained by the edge extraction part are averaged, and the vector of the normalized direction is fitted by the least square method to generate a transformation matrix; Input the transformation matrix and the high-resolution finger vein image into the preset affine transformation function to complete the rotation alignment of the high-contrast image; Based on the high-contrast image after rotation alignment, the upper and lower contours are found, and the discontinuous pixels of the upper and lower contours are monitored. The pixels of the discontinuous part within the set threshold range are filled with lines to obtain the finger contour image.

6. The finger vein image preprocessing method based on super-resolution algorithm according to claim 1, characterized in that: The finger vein image is registered according to the finger contour image to generate a grayscale image, and the grayscale image is binarized to complete the finger vein image preprocessing, specifically including: Based on the finger contour image, determine the finger joint prominence points to perform vein area registration and generate a grayscale image; The grayscale image is binarized to be converted into a binary image with only white dots and black dots, and the background and the non-vein part of the finger image are removed to complete the finger vein image preprocessing.

7. A finger vein image preprocessing system based on super-resolution algorithm, characterized in that: The system comprises: A high-resolution processing module, used for acquiring an original finger vein image, inputting the original finger vein image into a preset super-resolution model, and outputting a high-resolution finger vein image; A contrast enhancement module, used for performing contrast enhancement based on the high-resolution finger vein image to generate a high-contrast image; A contour acquisition module, used to perform finger rotation alignment and finger contour boundary extraction based on the high contrast image to obtain a finger contour image; The result output module is used to perform vein area registration according to the finger contour image, generate a grayscale image, perform binarization processing on the grayscale image, and complete finger vein image preprocessing.

8. An electronic 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 finger vein image preprocessing method based on the super-resolution algorithm as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the finger vein image preprocessing method based on the super-resolution algorithm as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the finger vein image preprocessing method based on the super-resolution algorithm as claimed in any one of claims 1 to 6 is implemented.

Images