Fingerprint image enhancement method, device, electronic device and storage medium

Abstract

This invention provides a fingerprint image enhancement method, apparatus, electronic device, and storage medium. The method includes: median filtering and normalization of a fingerprint image; calculating a continuous orientation pattern of the normalized fingerprint image; calculating ridge widths on the filtered fingerprint image based on the continuous orientation pattern, and calculating an average orientation pattern of the filtered fingerprint image according to a first preset size; determining a linear smoothing radius for each sub-filtered fingerprint image based on the average orientation pattern and the ridge widths corresponding to each sub-filtered fingerprint image of the first preset size; determining a linear smoothing coordinate set for each pixel in each sub-filtered fingerprint image based on the continuous orientation pattern and each radius; determining a corresponding linear smoothing path based on the linear coordinate set of each pixel; and obtaining a fingerprint enhancement image based on all pixels along the linear smoothing path of each pixel. This invention can reduce the computational load in the fingerprint image enhancement process while improving the enhancement effect.

Description

Technical Field

[0001] This invention relates to the field of fingerprint recognition technology, and in particular to a fingerprint image enhancement method, apparatus, electronic device, and storage medium. Background Technology

[0002] Biometric identification technology is a solution that uses automated technology to measure physical characteristics or behavioral traits of an individual for identity verification, comparing these characteristics with templates in a database to complete the authentication process. As the most mature and convenient member of the biometric technology family, fingerprint recognition technology has been successfully applied in various sectors of society, such as access control, attendance systems, e-commerce, and ATMs.

[0003] In automatic fingerprint recognition systems, fingerprint image enhancement is a key technology. The purposes of fingerprint image enhancement are: (1) smoothing noise, (2) improving the contrast of ridges and valleys in the fingerprint, and (3) connecting broken ridges and valleys. However, fingerprint acquisition is usually accompanied by various kinds of noise, some of which are caused by the acquisition instrument, such as dirt on the acquisition instrument or improper instrument parameter settings. Others are caused by the condition of the fingers, such as fingers that are too dry, too wet, have scars, or peel. The presence of these noises often leads to a large deviation between the extracted fingerprint features and the original fingerprint features.

[0004] Currently, commonly used fingerprint enhancement algorithms include: (1) Gabor filtering enhancement; (2) FFT-based frequency enhancement algorithm; (3) knowledge-based fingerprint enhancement algorithm; and (4) nonlinear diffusion-based model enhancement algorithm. All of these algorithms have problems to some extent, such as excessive computational load, weak noise resistance, and limited versatility. Existing technologies have also proposed a fingerprint enhancement algorithm that combines Gabor filtering with Gaussian filtering. This algorithm processes the obtained pixel horizontal and vertical gradient components using Gaussian filtering, then processes the orientation map using a Gaussian filter, then smooths the orientation map, and finally performs Gaussian filtering again. However, this fingerprint enhancement method has a high computational load and poor enhancement effect. Summary of the Invention

[0005] This invention provides a fingerprint image enhancement method, apparatus, electronic device, and storage medium to solve the problems of high computational load and poor enhancement effect in current fingerprint enhancement methods.

[0006] In a first aspect, embodiments of the present invention provide a fingerprint image enhancement method, comprising:

[0007] The fingerprint image is subjected to median filtering and normalization, and the continuous orientation pattern of the normalized fingerprint image is calculated.

[0008] Based on the orientation of each pixel in the continuous orientation map, the ridge width is counted on the filtered fingerprint image after median filtering, and the average orientation map of the filtered fingerprint image is calculated according to the first preset size.

[0009] Based on the average direction and corresponding ridge width of each sub-filtered fingerprint image of the first preset size in the average orientation map, the radius for linear smoothing of the sub-filtered fingerprint image is determined.

[0010] Based on the orientation of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image, a linear coordinate set for linearly smoothing the corresponding pixel in each sub-filtered fingerprint image is determined.

[0011] Based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, determine the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image;

[0012] Calculate the average gray value of all pixels along the linear smoothing path of each pixel in each sub-filtered fingerprint image, and use this as the linear smoothing result for that pixel to obtain the fingerprint enhancement image.

[0013] In one possible implementation, the step of calculating the ridge width on the median-filtered fingerprint image based on the orientation of each pixel in the continuous orientation map includes:

[0014] Based on the vertical direction of each pixel in the continuous orientation map, the width of the first ridge valley corresponding to each pixel in each second sub-filtered fingerprint image of the second preset size is calculated.

[0015] The average value of the first ridge width corresponding to all pixels in each second sub-filtered fingerprint image is determined as the second ridge width corresponding to each pixel in each third sub-filtered image of the third preset size; wherein, the third preset size is greater than the second preset size;

[0016] The ridge width of each pixel in the filtered fingerprint image is determined based on the second ridge width corresponding to each pixel in each third sub-filtered image.

[0017] In one possible implementation, the average orientation map of the filtered fingerprint image is calculated according to the orientation of each pixel in the continuous orientation map and a first preset size, including:

[0018] Based on the orientation of each pixel in the continuous orientation map, determine the angle of each pixel at the corresponding position in each sub-filtered fingerprint image of the first preset size;

[0019] The sine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed sine value, and the cosine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed cosine value.

[0020] Calculate the ratio of the accumulated sine value to the accumulated cosine value corresponding to each sub-filtered fingerprint image to obtain the average tangent value corresponding to each sub-filtered fingerprint image;

[0021] The angle corresponding to the average tangent value of each sub-filtered fingerprint image is determined as the average direction of that sub-filtered fingerprint image, thus obtaining the average direction map of the filtered fingerprint image.

[0022] In one possible implementation, determining the radius for linear smoothing of the filtered fingerprint image block based on the average orientation and corresponding ridge width of each sub-filtered fingerprint image block of a first preset size in the average orientation map includes:

[0023] The average angle of the filtered fingerprint image block is determined based on the average orientation of each sub-filtered fingerprint image of the first preset size in the average orientation pattern.

[0024] Based on the orientation of each pixel in the continuous orientation map, the angle of each pixel at the corresponding position in each sub-filtered fingerprint image is determined;

[0025] Calculate the standard deviation of the angle of each sub-filtered fingerprint image based on the angle of each pixel at the corresponding position in each sub-filtered fingerprint image and the average angle of that sub-filtered fingerprint image;

[0026] The radius for linear smoothing of a filtered fingerprint image block is determined based on the angular standard deviation of each block and the ridge width corresponding to that block.

[0027] In one possible implementation, determining the radius for linear smoothing of the filtered fingerprint image block based on the angular standard deviation of each sub-filtered fingerprint image block and the ridge width corresponding to that block includes:

[0028] Based on R = W * ctan(A), determine the radius for linear smoothing of the filtered fingerprint image of the block;

[0029] Where R is the radius of linear smoothing of each sub-filtered fingerprint image, W is the ridge width of each sub-filtered fingerprint image, and A is the angular standard deviation of each sub-filtered fingerprint image.

[0030] In one possible implementation, determining the linearly smoothed set of pixel coordinates for each sub-filtered fingerprint image based on the orientation of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image block includes:

[0031] Based on the orientation of each pixel in the continuous orientation map, the angle of each pixel at the corresponding position in each sub-filtered fingerprint image is determined;

[0032] A first linear coordinate set corresponding to the angle of each pixel in each sub-filtered fingerprint image is selected from the preset linear coordinate set. All coordinate points in the first linear coordinate set whose layer number is equal to the radius corresponding to the sub-filtered fingerprint image are determined as the linear coordinate set of the corresponding pixel in the sub-filtered fingerprint image and linearly smoothed.

[0033] In one possible implementation, determining the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image includes:

[0034] For each pixel in each sub-filtered fingerprint image, the origin coordinates of the linear coordinate set corresponding to the pixel are matched with the pixel to obtain the set of pixels that are linearly smoothed in the sub-filtered fingerprint image corresponding to the pixel.

[0035] Calculate the grayscale standard deviation between each pixel in the second outermost layer of the pixel set corresponding to the pixel and its adjacent pixel in the outermost layer. Take the minimum grayscale standard deviation as the first minimum grayscale standard deviation, and obtain the coordinates of the first pixel in the second outermost layer and the second pixel in the outermost layer corresponding to the first minimum grayscale standard deviation.

[0036] Following the method for obtaining the coordinates of the first pixel and the second pixel, starting from the outermost layer and moving inward, the coordinates of the pixel corresponding to the minimum grayscale standard deviation in each of the two adjacent layers are obtained sequentially from the second outermost layer in the pixel set corresponding to the pixel.

[0037] Based on the coordinates of the first pixel and the second pixel, and the coordinates of the pixel corresponding to the minimum grayscale standard deviation from the second outermost layer in the pixel set, the first linear smooth path corresponding to the pixel is determined.

[0038] Based on the method used to determine the first linear smooth path corresponding to the pixel, determine the second linear smooth path corresponding to the opposite direction of the pixel.

[0039] The linear smoothing path corresponding to the pixel is determined based on the first linear smoothing path and the second linear smoothing path corresponding to the pixel.

[0040] In a second aspect, embodiments of the present invention provide a fingerprint image enhancement device, comprising:

[0041] The preprocessing module is used to perform median filtering and normalization on the fingerprint image, and to calculate the continuous orientation pattern of the normalized fingerprint image.

[0042] The first processing module is used to calculate the ridge width on the filtered fingerprint image after median filtering based on the direction of each pixel in the continuous orientation map, and to calculate the average orientation map of the filtered fingerprint image according to the first preset size.

[0043] The second processing module is used to determine the radius for linear smoothing of the filtered fingerprint image block based on the average direction and corresponding ridge width of each sub-filtered fingerprint image block of the first preset size in the average orientation map.

[0044] The third processing module is used to determine the linear coordinate set of the corresponding position of the pixel in each sub-filtered fingerprint image by linearly smoothing it according to the direction of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image.

[0045] The fourth processing module is used to determine the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image.

[0046] The fifth processing module is used to calculate the average gray value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image, and use it as the linear smoothing result of that pixel to obtain the fingerprint enhancement image of the fingerprint image.

[0047] Thirdly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect or any possible implementation of the first aspect.

[0048] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any possible implementation thereof.

[0049] This invention provides a fingerprint image enhancement method, apparatus, electronic device, and storage medium. The method involves first performing median filtering and normalization on the fingerprint image, and then calculating the continuous orientation pattern of the normalized fingerprint image. Next, based on the orientation of each pixel in the continuous orientation pattern, the ridge widths are statistically analyzed on the filtered fingerprint image after median filtering, and the average orientation pattern of the filtered fingerprint image is calculated according to a first preset size. Then, based on the average orientation and corresponding ridge width of each sub-filtered fingerprint image of the first preset size in the average orientation pattern, a radius for linear smoothing of that sub-filtered fingerprint image is determined. Next, based on the orientation of each pixel in the continuous orientation pattern and the radius corresponding to each sub-filtered fingerprint image, a linear coordinate set for linear smoothing of the corresponding pixel in each sub-filtered fingerprint image is determined. Then, based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, a linear smoothing path corresponding to each pixel in each sub-filtered fingerprint image is determined. Finally, the average grayscale value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image is calculated as the linear smoothing result for that pixel, resulting in a fingerprint enhancement image. In this embodiment, the average orientation map of each sub-filtered fingerprint image of the first preset size is calculated based on the orientation of each pixel in the continuous orientation map to obtain the radius for linear smoothing of each sub-filtered fingerprint image. This allows each pixel in the sub-filtered fingerprint image to be linearly smoothed according to the corresponding radius, without having to determine a linear smoothing radius for each pixel in the filtered fingerprint image. This reduces the computational load in the fingerprint image enhancement process. Furthermore, since this embodiment determines the radius for linear smoothing of each pixel in the sub-filtered fingerprint image based on the average orientation map and the corresponding ridge width, the radius can be changed accordingly for each sub-filtered fingerprint image, thereby preserving the detailed features of parts with drastic orientation changes, such as cross points, and thus improving the enhancement effect of the fingerprint image. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating the implementation of the fingerprint image enhancement method provided in this embodiment of the invention;

[0052] Figure 2 This is a schematic diagram of the first linear coordinate set corresponding to the standard angle 0° in the preset linear coordinate set provided in the embodiments of the present invention;

[0053] Figure 3This is a schematic diagram of the fingerprint image enhancement device provided in an embodiment of the present invention;

[0054] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0055] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0056] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0057] See Figure 1 The document illustrates a flowchart of the fingerprint image enhancement method provided in an embodiment of the present invention, which is described in detail below:

[0058] In step 101, the fingerprint image is subjected to median filtering and normalization, and the continuous orientation pattern of the normalized fingerprint image is calculated.

[0059] Median filtering is effective at removing impulse noise, especially in protecting the edges of the signal while filtering out noise, so it is the first step to perform median filtering on the fingerprint image to remove various noises.

[0060] When normalizing the filtered fingerprint image after median filtering, the image can be first divided into M*M blocks, and then each block can be normalized. The size of M should include at least one ridge width (the width between two adjacent ridges). Generally, M should be greater than 11 pixels. For example, M can be 12.

[0061] The purpose of image normalization is to adjust the average grayscale and contrast of an image to a fixed level, thereby reducing the differences between different fingerprint images. In this embodiment, the filtered fingerprint image is first divided into M*M blocks, and then each block is normalized, which can adjust the proportion of ridges and valleys in the filtered fingerprint image to the same ratio.

[0062] The orientation map is a transformation representation of a fingerprint image, where the direction of the ridges represents the ridges. It is generally divided into point orientation maps, block orientation maps, and continuous orientation maps. The continuous orientation map is calculated based on the point orientation map, combining the advantages of the continuous transition of the point orientation map and the noise resistance of the block orientation map, while overcoming the poor noise resistance of the point orientation map and the unnatural transition in the critical region of the block orientation map. Therefore, in this embodiment, after median filtering and normalization of the fingerprint image, the continuous orientation map of the normalized fingerprint image is calculated to facilitate more accurate subsequent processing.

[0063] In step 102, based on the orientation of each pixel in the continuous orientation map, the ridge width is counted on the filtered fingerprint image after median filtering, and the average orientation map of the filtered fingerprint image is calculated according to the first preset size.

[0064] The first preset size can be the maximum value of the length for subsequent linear smoothing. Generally, the maximum radius for linear smoothing is 6 pixels, meaning the first preset size can be 12×12 pixels. This embodiment does not limit the maximum value of the linear smoothing length; the maximum value can be adjusted according to the actual needs of the fingerprint image.

[0065] In this embodiment, based on the orientation of each pixel in the continuous orientation map, the ridge width is counted on the filtered fingerprint image after median filtering, and the average orientation map of the filtered fingerprint image is calculated according to the first preset size. This allows the radius for linear smoothing of each sub-filtered fingerprint image of the first preset size to be obtained, so that each pixel in the sub-filtered fingerprint image is linearly smoothed according to the corresponding radius, without having to determine a radius for linear smoothing for each pixel in the filtered fingerprint image, thereby reducing the amount of computation in the fingerprint image enhancement process.

[0066] Optionally, based on the orientation of each pixel in the continuous orientation map, the ridge width is calculated on the filtered fingerprint image after median filtering. This can include:

[0067] Based on the vertical direction of each pixel in the continuous orientation map, the width of the first ridge valley corresponding to each pixel in each second sub-filtered fingerprint image of the second preset size is calculated.

[0068] The average value of the first ridge width corresponding to all pixels in each second sub-filtered fingerprint image is determined as the second ridge width corresponding to each pixel in each third sub-filtered image of the third preset size; wherein, the third preset size is greater than the second preset size.

[0069] The ridge width of each pixel in the filtered fingerprint image is determined based on the second ridge width corresponding to each pixel in each third sub-filtered image.

[0070] In this context, the direction perpendicular to the direction of each pixel in the continuous orientation map is used to calculate the width of adjacent ridges at the corresponding pixel position in each second sub-filtered fingerprint image according to the direction perpendicular to the direction of each pixel in the continuous orientation map. This yields the width of the first ridge valley corresponding to each pixel in each second sub-filtered fingerprint image.

[0071] Since the ridge width tends to be the same within a small area, the ridge width of the third preset size area can be calculated using an area of ​​the second preset size to reduce the computational load. For example, the area of ​​the second preset size can be the middle area of ​​the area of ​​the third preset size.

[0072] For example, the second preset size can be 48×48 pixels, and the third preset size can be 96×96 pixels. That is, the average value of the first ridge width calculated using the 48×48 pixels in the middle of the 96×96 pixels is used as the second ridge width of the 96×96 pixels.

[0073] According to this method, the second ridge width corresponding to each third sub-filtered image of the third preset size in the filtered fingerprint image can be calculated, which is the ridge width corresponding to each pixel in the filtered fingerprint image.

[0074] Optionally, calculating the average orientation map of the filtered fingerprint image according to the orientation of each pixel in the continuous orientation map and a first preset size may include:

[0075] Based on the orientation of each pixel in the continuous orientation map, determine the angle of each pixel at the corresponding position in each sub-filtered fingerprint image of the first preset size.

[0076] The sine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed sine value, and the cosine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed cosine value.

[0077] Calculate the ratio of the accumulated sine value to the accumulated cosine value for each sub-filtered fingerprint image to obtain the average tangent value for each sub-filtered fingerprint image.

[0078] The angle corresponding to the average tangent value of each sub-filtered fingerprint image is determined as the average direction of that sub-filtered fingerprint image, thus obtaining the average direction map of the filtered fingerprint image.

[0079] In this embodiment, since the orientation of each pixel in each sub-filtered fingerprint image of the first preset size is not much different, in order to reduce the amount of computation in the subsequent fingerprint image enhancement process, the average orientation pattern of the filtered fingerprint image can be calculated according to the orientation of each pixel in the continuous orientation pattern and the first preset size.

[0080] Specifically, the angles of corresponding pixels in each sub-filtered fingerprint image of the first preset size can be determined based on the orientation of each pixel in the continuous orientation map. The sine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are accumulated to obtain an accumulated sine value, and the cosine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are accumulated to obtain an accumulated cosine value. The ratio of the accumulated sine value to the accumulated cosine value for each sub-filtered fingerprint image is calculated to obtain the average tangent value for each sub-filtered fingerprint image. The angle corresponding to the average tangent value of each sub-filtered fingerprint image is determined as the average direction of that sub-filtered fingerprint image, and the average orientation map of the filtered fingerprint image is obtained based on the average direction of each sub-filtered fingerprint image.

[0081] In step 103, the radius for linear smoothing of the filtered fingerprint image block is determined based on the average direction and corresponding ridge width of each sub-filtered fingerprint image block of the first preset size in the average orientation map.

[0082] In this embodiment, the radius for linear smoothing of each sub-filtered fingerprint image is obtained based on the average orientation pattern and corresponding ridge width of each sub-filtered fingerprint image of a first preset size. This ensures that each pixel in the sub-filtered fingerprint image is linearly smoothed according to its corresponding radius, without needing to determine a linear smoothing radius for each pixel in the filtered fingerprint image. Therefore, the computational load in the fingerprint image enhancement process can be reduced. Furthermore, since this embodiment determines the radius for linear smoothing of each pixel in the filtered fingerprint image based on the average orientation pattern and corresponding ridge width of each sub-filtered fingerprint image, corresponding radius changes can be applied to each sub-filtered fingerprint image, thereby preserving the detailed features of areas with drastic directional changes, such as bifurcation points, and thus improving the fingerprint image enhancement effect.

[0083] Optionally, the radius for linearly smoothing the filtered fingerprint image block is determined based on the average orientation and corresponding ridge width of each sub-filtered fingerprint image block of a first preset size in the average orientation map. This may include:

[0084] The average angle of the filtered fingerprint image block is determined based on the average orientation of each sub-filtered fingerprint image of the first preset size in the average orientation pattern.

[0085] Based on the orientation of each pixel in the continuous orientation map, the angle of each pixel at the corresponding position in each sub-filtered fingerprint image is determined.

[0086] The standard deviation of the angle of each sub-filtered fingerprint image is calculated based on the angle of each pixel at the corresponding position in each sub-filtered fingerprint image and the average angle of that sub-filtered fingerprint image.

[0087] The radius for linear smoothing of a filtered fingerprint image block is determined based on the angular standard deviation of each block and the ridge width corresponding to that block.

[0088] The angular standard deviation of each sub-filtered fingerprint image characterizes the degree of directional difference among pixels in that sub-filtered fingerprint image. Based on the angular standard deviation and the corresponding ridge width of each sub-filtered fingerprint image, the radius for linear smoothing of that sub-filtered fingerprint image is determined. When the directional difference among pixels in each sub-filtered fingerprint image is large, the radius for linear smoothing can be reduced to preserve the detailed features of areas with drastic directional changes, such as bifurcation points, thus improving the enhancement effect of the fingerprint image. When the directional difference among pixels in each sub-filtered fingerprint image is small, a larger radius can be used for linear smoothing to reduce computational load.

[0089] Optionally, the radius for linear smoothing of the filtered fingerprint image block can be determined based on the angular standard deviation of each sub-filtered fingerprint image block and the ridge width corresponding to that sub-filtered fingerprint image block. This can include:

[0090] The radius for linear smoothing of the filtered fingerprint image of the block is determined by R = W * ctan(A).

[0091] Where R is the radius of linear smoothing of each sub-filtered fingerprint image, W is the ridge width of each sub-filtered fingerprint image, and A is the angular standard deviation of each sub-filtered fingerprint image.

[0092] According to the formula provided in this embodiment, the larger the ridge width corresponding to each sub-filtered fingerprint image and the smaller the angular standard deviation of each sub-filtered fingerprint image, the larger the radius for linear smoothing each sub-filtered fingerprint image is because the difference in the orientation of each pixel in each sub-filtered fingerprint image is small. Conversely, the smaller the ridge width corresponding to each sub-filtered fingerprint image and the larger the angular standard deviation of each sub-filtered fingerprint image, the smaller the radius for linear smoothing each sub-filtered fingerprint image is because the difference in the orientation of each pixel in each sub-filtered fingerprint image is large.

[0093] In step 104, based on the orientation of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image, a linear coordinate set for linearly smoothing the corresponding pixel in each sub-filtered fingerprint image is determined.

[0094] In this process, the radius for linear smoothing is the same for each pixel in each sub-filtered fingerprint image. However, the orientation of each pixel in each sub-filtered fingerprint image is different. Therefore, when performing linear smoothing on each pixel in each sub-filtered fingerprint image, the region for linear smoothing is determined based on the orientation of each pixel in each sub-filtered fingerprint image and the corresponding radius of each sub-filtered fingerprint image.

[0095] The linear coordinate set is the set of coordinate points for linear smoothing. It determines the linear coordinate set for linear smoothing of each pixel in each sub-filtered fingerprint image, which in turn determines the region for linear smoothing of each pixel in each sub-filtered fingerprint image.

[0096] Optionally, based on the orientation of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image, a linearly smoothed set of coordinates for the corresponding pixel in each sub-filtered fingerprint image is determined, which may include:

[0097] Based on the orientation of each pixel in the continuous orientation map, the angle of each pixel at the corresponding position in each sub-filtered fingerprint image is determined.

[0098] Select a first linear coordinate set from the preset linear coordinate set that corresponds to the angle of each pixel in each sub-filtered fingerprint image. Then, determine all coordinate points in the first linear coordinate set whose layer number is equal to the radius of the corresponding sub-filtered fingerprint image as the linear coordinate set of the corresponding pixel in the sub-filtered fingerprint image and perform linear smoothing.

[0099] The preset linear coordinate set is a collection of coordinate points that have been linearly smoothed for each standard angle. Generally, the direction of the fingerprint ridge is divided into eight directions between 0 and π, and the angle corresponding to each direction can be used as a standard angle.

[0100] The set of coordinate points for linear smoothing at each standard angle can be represented graphically as a sector. This sector has an included angle of the standard angle ± a preset angle, and a preset radius as its radius. A set of standard coordinate points is generated, and the coordinate points within this sector are used as the set of coordinate points for linear smoothing at the corresponding standard angle, which is the first linear coordinate set for each standard angle.

[0101] For example, the preset angle can be between 8° and 15°, such as 12°. The preset radius length can be the maximum value of the radius for linear smoothing, such as 6 pixels. This embodiment does not limit the specific values ​​of the preset angle and preset radius length.

[0102] In the first linear coordinate set of each standard angle, the first linear coordinate set of each standard angle can be divided into N layers according to a preset radius length, so that when performing linear smoothing, the adjacent points of adjacent layers can be used to determine the linear smoothing path.

[0103] For example, the first linear coordinate set corresponding to the standard angle 0° in the preset linear coordinate set is as follows: Figure 2 As shown in this example, since the preset radius length is 6 pixels, the first linear coordinate set corresponding to the standard angle 0° can be divided into 6 layers.

[0104] In step 105, the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image is determined based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image.

[0105] In this embodiment, based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, the linear smoothing path corresponding to each pixel in each sub-filtered fingerprint image is determined. From the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, the pixel corresponding to the coordinate that can obtain a better fingerprint image enhancement effect is selected for linear smoothing according to the length of the linear smoothing, so as to obtain a better fingerprint image enhancement effect.

[0106] Optionally, determining the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image may include:

[0107] For each pixel in each sub-filtered fingerprint image, the origin coordinates of the linear coordinate set corresponding to the pixel are matched with the pixel to obtain the set of pixels that are linearly smoothed in the sub-filtered fingerprint image corresponding to the pixel.

[0108] Calculate the grayscale standard deviation between each pixel in the second outermost layer of the pixel set corresponding to the pixel and its adjacent pixel in the outermost layer. Take the minimum grayscale standard deviation as the first minimum grayscale standard deviation, and obtain the coordinates of the first pixel in the second outermost layer and the second pixel in the outermost layer corresponding to the first minimum grayscale standard deviation.

[0109] Following the method for obtaining the coordinates of the first pixel and the second pixel, starting from the outermost layer and moving inward, the coordinates of the pixel corresponding to the minimum grayscale standard deviation in each of the two adjacent layers are obtained sequentially from the second outermost layer in the pixel set corresponding to the pixel.

[0110] Based on the coordinates of the first pixel and the second pixel, and the coordinates of the pixel corresponding to the minimum grayscale standard deviation from the second outermost layer in the pixel set, the first linear smooth path corresponding to the pixel is determined.

[0111] Based on the method used to determine the first linear smooth path corresponding to the pixel, determine the second linear smooth path corresponding to the opposite direction of the pixel.

[0112] The linear smoothing path corresponding to the pixel is determined based on the first linear smoothing path and the second linear smoothing path corresponding to the pixel.

[0113] For example, according to Figure 2 The first linear coordinate set shown illustrates the method for determining the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image in this embodiment:

[0114] Combination Figure 2 Suppose that a pixel in a sub-filtered fingerprint image has an angle of 0° and a radius of 5 pixels. Then, the set of pixels in the sub-filtered fingerprint image corresponding to this pixel, after linear smoothing, has five layers. That is, the fifth layer is the outermost layer of the pixel set corresponding to this pixel. The steps to determine the linear smoothing path corresponding to this pixel are as follows:

[0115] 1. First calculate the sum of the grayscale value and the square of the grayscale value for the fifth layer;

[0116] 2. Calculate the standard deviation between each pixel in the fourth layer and each adjacent pixel in the fifth layer, and take the minimum value, recording the position of the point with the minimum value;

[0117] 3. Calculate the standard deviation of each pixel in the third layer and each adjacent pixel in the fourth layer, and so on, up to the 0th layer, to obtain the first linear smooth path corresponding to the pixel;

[0118] 4. Then calculate the minimum standard deviation path in the opposite direction of the pixel to obtain the second linear smooth path corresponding to the pixel;

[0119] 5. The combination of the first linear smoothing path and the second linear smoothing path corresponding to the pixel is the linear smoothing path corresponding to the pixel.

[0120] In this embodiment, it can be based on The grayscale standard deviation can be calculated using the sum of squared grayscale values ​​and the cumulative sum of grayscale values. For example, the square of the grayscale value of each pixel in the fifth layer can be calculated first. When calculating the grayscale standard deviation with the corresponding pixel in the fourth layer, the square of the grayscale value of the corresponding pixel in the fourth layer can be added to obtain the sum of squared grayscale values, which can reduce the amount of computation.

[0121] Among them, with Figure 2For example, the topmost pixel of the fourth layer is adjacent to the topmost pixel of the fifth layer and the middle pixel of the fifth layer. The middle pixel of the fourth layer is adjacent to the topmost, middle, and bottommost pixels of the fifth layer. The bottommost pixel of the fourth layer is adjacent to the middle and bottommost pixels of the fifth layer.

[0122] In step 106, the average gray value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image is calculated as the linear smoothing result of that pixel, thus obtaining the fingerprint enhancement image of the fingerprint image.

[0123] In this embodiment, after determining the linear smoothing path corresponding to each pixel in each sub-filtered fingerprint image, the grayscale value of each pixel on the linear smoothing path corresponding to each pixel in each sub-filtered fingerprint image can be obtained. Then, the average grayscale value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image can be calculated as the linear smoothing result of that pixel. Thus, in the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, linear smoothing is performed using the linear smoothing path with the smallest standard deviation according to the length of the linear smoothing, so as to obtain a better fingerprint image enhancement effect.

[0124] This invention first performs median filtering and normalization on the fingerprint image, and calculates the continuous orientation pattern of the normalized fingerprint image. Then, based on the orientation of each pixel in the continuous orientation pattern, the ridge width is calculated on the filtered fingerprint image after median filtering, and the average orientation pattern of the filtered fingerprint image is calculated according to a first preset size. Then, based on the average orientation and corresponding ridge width of each sub-filtered fingerprint image of the first preset size in the average orientation pattern, the radius for linear smoothing of that sub-filtered fingerprint image is determined. Then, based on the orientation of each pixel in the continuous orientation pattern and the radius corresponding to each sub-filtered fingerprint image, the linear coordinate set for linear smoothing of the corresponding pixel in each sub-filtered fingerprint image is determined. Then, based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, the linear smoothing path corresponding to each pixel in each sub-filtered fingerprint image is determined. Finally, the average gray value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image is calculated as the linear smoothing result of that pixel, resulting in a fingerprint enhancement image. In this embodiment, the average orientation map of each sub-filtered fingerprint image of the first preset size is calculated based on the orientation of each pixel in the continuous orientation map to obtain the radius for linear smoothing of each sub-filtered fingerprint image. This allows each pixel in the sub-filtered fingerprint image to be linearly smoothed according to the corresponding radius, without having to determine a linear smoothing radius for each pixel in the filtered fingerprint image. This reduces the computational load in the fingerprint image enhancement process. Furthermore, since this embodiment determines the radius for linear smoothing of each pixel in the sub-filtered fingerprint image based on the average orientation map and the corresponding ridge width, the radius can be changed accordingly for each sub-filtered fingerprint image, thereby preserving the detailed features of parts with drastic orientation changes, such as cross points, and thus improving the enhancement effect of the fingerprint image.

[0125] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0126] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0127] Figure 3 A schematic diagram of the fingerprint image enhancement device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below:

[0128] like Figure 3 As shown, the fingerprint image enhancement device includes: a preprocessing module 31, a first processing module 32, a second processing module 33, a third processing module 34, a fourth processing module 35, and a fifth processing module 36.

[0129] The preprocessing module 31 is used to perform median filtering and normalization on the fingerprint image, and to calculate the continuous orientation pattern of the normalized fingerprint image.

[0130] The first processing module 32 is used to calculate the ridge width on the filtered fingerprint image after median filtering based on the direction of each pixel in the continuous orientation map, and to calculate the average orientation map of the filtered fingerprint image according to the first preset size.

[0131] The second processing module 33 is used to determine the radius for linear smoothing of the filtered fingerprint image block based on the average direction and corresponding ridge width of each sub-filtered fingerprint image block of the first preset size in the average orientation map.

[0132] The third processing module 34 is used to determine the linear coordinate set of the corresponding position of the pixel in each sub-filtered fingerprint image based on the direction of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image;

[0133] The fourth processing module 35 is used to determine the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image.

[0134] The fifth processing module 36 is used to calculate the average gray value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image, and use it as the linear smoothing result of that pixel to obtain the fingerprint enhancement image of the fingerprint image.

[0135] This invention first performs median filtering and normalization on the fingerprint image, and calculates the continuous orientation pattern of the normalized fingerprint image. Then, based on the orientation of each pixel in the continuous orientation pattern, the ridge width is calculated on the filtered fingerprint image after median filtering, and the average orientation pattern of the filtered fingerprint image is calculated according to a first preset size. Then, based on the average orientation and corresponding ridge width of each sub-filtered fingerprint image of the first preset size in the average orientation pattern, the radius for linear smoothing of that sub-filtered fingerprint image is determined. Then, based on the orientation of each pixel in the continuous orientation pattern and the radius corresponding to each sub-filtered fingerprint image, the linear coordinate set for linear smoothing of the corresponding pixel in each sub-filtered fingerprint image is determined. Then, based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, the linear smoothing path corresponding to each pixel in each sub-filtered fingerprint image is determined. Finally, the average gray value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image is calculated as the linear smoothing result of that pixel, resulting in a fingerprint enhancement image. In this embodiment, the average orientation map of each sub-filtered fingerprint image of the first preset size is calculated based on the orientation of each pixel in the continuous orientation map to obtain the radius for linear smoothing of each sub-filtered fingerprint image. This allows each pixel in the sub-filtered fingerprint image to be linearly smoothed according to the corresponding radius, without having to determine a linear smoothing radius for each pixel in the filtered fingerprint image. This reduces the computational load in the fingerprint image enhancement process. Furthermore, since this embodiment determines the radius for linear smoothing of each pixel in the sub-filtered fingerprint image based on the average orientation map and the corresponding ridge width, the radius can be changed accordingly for each sub-filtered fingerprint image, thereby preserving the detailed features of parts with drastic orientation changes, such as cross points, and thus improving the enhancement effect of the fingerprint image.

[0136] In one possible implementation, the first processing module 32 is used to calculate the width of the first ridge valley corresponding to each pixel in each second sub-filtered fingerprint image of the second preset size according to the vertical direction of the direction of each pixel in the continuous orientation map.

[0137] The average value of the first ridge width corresponding to all pixels in each second sub-filtered fingerprint image is determined as the second ridge width corresponding to each pixel in each third sub-filtered image of the third preset size; wherein, the third preset size is greater than the second preset size;

[0138] The ridge width of each pixel in the filtered fingerprint image is determined based on the second ridge width corresponding to each pixel in each third sub-filtered image.

[0139] In one possible implementation, the first processing module 32 is used to determine the angle of each pixel at a corresponding position in each sub-filtered fingerprint image of a first preset size based on the orientation of each pixel in the continuous orientation map.

[0140] The sine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed sine value, and the cosine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed cosine value.

[0141] Calculate the ratio of the accumulated sine value to the accumulated cosine value corresponding to each sub-filtered fingerprint image to obtain the average tangent value corresponding to each sub-filtered fingerprint image;

[0142] The angle corresponding to the average tangent value of each sub-filtered fingerprint image is determined as the average direction of that sub-filtered fingerprint image, thus obtaining the average direction map of the filtered fingerprint image.

[0143] In one possible implementation, the second processing module 33 is used to determine the average angle of the block of filtered fingerprint image based on the average orientation of each block of filtered fingerprint image of the first preset size in the average orientation pattern.

[0144] Based on the orientation of each pixel in the continuous orientation map, the angle of each pixel at the corresponding position in each sub-filtered fingerprint image is determined;

[0145] Calculate the standard deviation of the angle of each sub-filtered fingerprint image based on the angle of each pixel at the corresponding position in each sub-filtered fingerprint image and the average angle of that sub-filtered fingerprint image;

[0146] The radius for linear smoothing of a filtered fingerprint image block is determined based on the angular standard deviation of each block and the ridge width corresponding to that block.

[0147] In one possible implementation, the second processing module 33 is used to determine the radius for linear smoothing of the block-filtered fingerprint image according to R = W * ctan(A);

[0148] Where R is the radius of linear smoothing of each sub-filtered fingerprint image, W is the ridge width of each sub-filtered fingerprint image, and A is the angular standard deviation of each sub-filtered fingerprint image.

[0149] In one possible implementation, the third processing module 34 is used to determine the angle of each pixel at a corresponding position in each sub-filtered fingerprint image based on the orientation of each pixel in the continuous orientation map.

[0150] A first linear coordinate set corresponding to the angle of each pixel in each sub-filtered fingerprint image is selected from the preset linear coordinate set. All coordinate points in the first linear coordinate set whose layer number is equal to the radius corresponding to the sub-filtered fingerprint image are determined as the linear coordinate set of the corresponding pixel in the sub-filtered fingerprint image and linearly smoothed.

[0151] In one possible implementation, the fourth processing module 35 is used to, for each pixel in each sub-filtered fingerprint image, correspond the origin coordinates of the linear coordinate set corresponding to the pixel to the pixel, so as to obtain the set of pixels that are linearly smoothed in the sub-filtered fingerprint image corresponding to the pixel.

[0152] Calculate the grayscale standard deviation between each pixel in the second outermost layer of the pixel set corresponding to the pixel and its adjacent pixel in the outermost layer. Take the minimum grayscale standard deviation as the first minimum grayscale standard deviation, and obtain the coordinates of the first pixel in the second outermost layer and the second pixel in the outermost layer corresponding to the first minimum grayscale standard deviation.

[0153] Following the method for obtaining the coordinates of the first pixel and the second pixel, starting from the outermost layer and moving inward, the coordinates of the pixel corresponding to the minimum grayscale standard deviation in each of the two adjacent layers are obtained sequentially from the second outermost layer in the pixel set corresponding to the pixel.

[0154] Based on the coordinates of the first pixel and the second pixel, and the coordinates of the pixel corresponding to the minimum grayscale standard deviation from the second outermost layer in the pixel set, the first linear smooth path corresponding to the pixel is determined.

[0155] Based on the method used to determine the first linear smooth path corresponding to the pixel, determine the second linear smooth path corresponding to the opposite direction of the pixel.

[0156] The linear smoothing path corresponding to the pixel is determined based on the first linear smoothing path and the second linear smoothing path corresponding to the pixel.

[0157] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Figure 4 As shown, the electronic device 4 in this embodiment includes a processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the processor 40. When the processor 40 executes the computer program 42, it implements the steps in the various fingerprint image enhancement method embodiments described above, for example... Figure 1 Steps 101 to 106 are shown. Alternatively, when processor 40 executes computer program 42, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 3 The functions of modules / units 31 to 36 shown.

[0158] For example, computer program 42 can be divided into one or more modules / units, one or more of which are stored in memory 41 and executed by processor 40 to complete the present invention. One or more modules / units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 42 in electronic device 4. For example, computer program 42 can be divided into... Figure 3 Modules / units 31 to 36 are shown.

[0159] Electronic device 4 can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. Electronic device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic devices may also include input / output devices, network access devices, buses, etc.

[0160] The processor 40 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0161] The memory 41 can be an internal storage unit of the electronic device 4, such as a hard disk or RAM. The memory 41 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 41 can include both internal and external storage units of the electronic device 4. The memory 41 is used to store computer programs and other programs and data required by the electronic device. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0162] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to 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 embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0163] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0164] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0165] In the embodiments provided by this invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0166] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0167] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0168] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various fingerprint image enhancement method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0169] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A fingerprint image enhancement method, characterized in that, include: The fingerprint image is subjected to median filtering and normalization, and the continuous orientation pattern of the normalized fingerprint image is calculated. Based on the orientation of each pixel in the continuous orientation map, the ridge width is counted on the filtered fingerprint image after median filtering, and the average orientation map of the filtered fingerprint image is calculated according to the first preset size. Based on the average direction and corresponding ridge width of each sub-filtered fingerprint image of the first preset size in the average orientation map, the radius for linear smoothing of the sub-filtered fingerprint image is determined. Based on the orientation of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image, a linear coordinate set for linearly smoothing the corresponding pixel in each sub-filtered fingerprint image is determined. Based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image, determine the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image; Calculate the average gray value of all pixels along the linear smoothing path of each pixel in each sub-filtered fingerprint image, and use this as the linear smoothing result for that pixel to obtain the fingerprint enhancement image.

2. The fingerprint image enhancement method according to claim 1, characterized in that, The step of calculating the ridge width on the median-filtered fingerprint image based on the orientation of each pixel in the continuous orientation map includes: Based on the vertical direction of each pixel in the continuous orientation map, the width of the first ridge valley corresponding to each pixel in each second sub-filtered fingerprint image of the second preset size is calculated. The average value of the first ridge width corresponding to all pixels in each second sub-filtered fingerprint image is determined as the second ridge width corresponding to each pixel in each third sub-filtered image of the third preset size; wherein, the third preset size is greater than the second preset size; The ridge width of each pixel in the filtered fingerprint image is determined based on the second ridge width corresponding to each pixel in each third sub-filtered image.

3. The fingerprint image enhancement method according to claim 1, characterized in that, Based on the orientation of each pixel in the continuous orientation map, the average orientation map of the filtered fingerprint image is calculated according to a first preset size, including: Based on the orientation of each pixel in the continuous orientation map, determine the angle of each pixel at the corresponding position in each sub-filtered fingerprint image of the first preset size; The sine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed sine value, and the cosine values ​​corresponding to the angles of each pixel in each sub-filtered fingerprint image are summed to obtain the summed cosine value. Calculate the ratio of the accumulated sine value to the accumulated cosine value corresponding to each sub-filtered fingerprint image to obtain the average tangent value corresponding to each sub-filtered fingerprint image; The angle corresponding to the average tangent value of each sub-filtered fingerprint image is determined as the average direction of that sub-filtered fingerprint image, thus obtaining the average direction map of the filtered fingerprint image.

4. The fingerprint image enhancement method according to claim 1, characterized in that, The step of determining the radius for linear smoothing of the filtered fingerprint image block based on the average direction and corresponding ridge width of each sub-filtered fingerprint image block of the first preset size in the average orientation map includes: The average angle of the filtered fingerprint image block is determined based on the average orientation of each sub-filtered fingerprint image of the first preset size in the average orientation pattern. Based on the orientation of each pixel in the continuous orientation map, the angle of each pixel at the corresponding position in each sub-filtered fingerprint image is determined; Calculate the standard deviation of the angle of each sub-filtered fingerprint image based on the angle of each pixel at the corresponding position in each sub-filtered fingerprint image and the average angle of that sub-filtered fingerprint image; The radius for linear smoothing of a filtered fingerprint image block is determined based on the angular standard deviation of each block and the ridge width corresponding to that block.

5. The fingerprint image enhancement method according to claim 4, characterized in that, The step of determining the radius for linear smoothing of a filtered fingerprint image block based on the angular standard deviation of each sub-filtered fingerprint image block and the ridge width corresponding to that block includes: Based on R = W * ctan(A), determine the radius for linear smoothing of the filtered fingerprint image of the block; Where R is the radius of linear smoothing of each sub-filtered fingerprint image, W is the ridge width of each sub-filtered fingerprint image, and A is the angular standard deviation of each sub-filtered fingerprint image.

6. The fingerprint image enhancement method according to claim 1, characterized in that, The step of determining the linear coordinate set of corresponding pixels in each sub-filtered fingerprint image based on the orientation of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image includes: Based on the orientation of each pixel in the continuous orientation map, the angle of each pixel at the corresponding position in each sub-filtered fingerprint image is determined; A first linear coordinate set corresponding to the angle of each pixel in each sub-filtered fingerprint image is selected from the preset linear coordinate set. All coordinate points in the first linear coordinate set whose layer number is equal to the radius corresponding to the sub-filtered fingerprint image are determined as the linear coordinate set of the corresponding pixel in the sub-filtered fingerprint image and linearly smoothed.

7. The fingerprint image enhancement method according to any one of claims 1-6, characterized in that, The step of determining the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image includes: For each pixel in each sub-filtered fingerprint image, the origin coordinates of the linear coordinate set corresponding to the pixel are matched with the pixel to obtain the set of pixels that are linearly smoothed in the sub-filtered fingerprint image corresponding to the pixel. Calculate the grayscale standard deviation between each pixel in the second outermost layer of the pixel set corresponding to the pixel and its adjacent pixel in the outermost layer. Take the minimum grayscale standard deviation as the first minimum grayscale standard deviation, and obtain the coordinates of the first pixel in the second outermost layer and the second pixel in the outermost layer corresponding to the first minimum grayscale standard deviation. Following the method for obtaining the coordinates of the first pixel and the second pixel, starting from the outermost layer and moving inward, the coordinates of the pixel corresponding to the minimum grayscale standard deviation in each of the two adjacent layers are obtained sequentially from the second outermost layer in the pixel set corresponding to the pixel. Based on the coordinates of the first pixel and the second pixel, and the coordinates of the pixel corresponding to the minimum grayscale standard deviation from the second outermost layer in the pixel set, the first linear smooth path corresponding to the pixel is determined. Based on the method used to determine the first linear smooth path corresponding to the pixel, determine the second linear smooth path corresponding to the opposite direction of the pixel. The linear smoothing path corresponding to the pixel is determined based on the first linear smoothing path and the second linear smoothing path corresponding to the pixel.

8. A fingerprint image enhancement device, characterized in that, include: The preprocessing module is used to perform median filtering and normalization on the fingerprint image, and to calculate the continuous orientation pattern of the normalized fingerprint image. The first processing module is used to calculate the ridge width on the filtered fingerprint image after median filtering based on the direction of each pixel in the continuous orientation map, and to calculate the average orientation map of the filtered fingerprint image according to the first preset size. The second processing module is used to determine the radius for linear smoothing of the filtered fingerprint image block based on the average direction and corresponding ridge width of each sub-filtered fingerprint image block of the first preset size in the average orientation pattern. The third processing module is used to determine the linear coordinate set of the corresponding position of the pixel in each sub-filtered fingerprint image by linearly smoothing it according to the direction of each pixel in the continuous orientation map and the radius corresponding to each sub-filtered fingerprint image. The fourth processing module is used to determine the linear smooth path corresponding to each pixel in each sub-filtered fingerprint image based on the linear coordinate set corresponding to each pixel in each sub-filtered fingerprint image. The fifth processing module is used to calculate the average gray value of all pixels on the linear smoothing path of each pixel in each sub-filtered fingerprint image, and use it as the linear smoothing result of that pixel to obtain the fingerprint enhancement image of the fingerprint image.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory being used to store a computer program, and the processor being used to call and run the computer program stored in the memory to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7 above.

Images