CN109902586A - Palmprint extraction method, device, storage medium, and server - Google Patents

Abstract

The present invention relates to bio-identification, palmprint recognition technology field, a kind of palmmprint extracting method provided by the embodiments of the present application, comprising: obtain palm image, palm image is pre-processed, palmprint image to be modified is obtained；The calculating of curvature value is carried out to each feature pixel in palmprint image to be modified, the curvature value based on the feature pixel is modified palmprint image to be modified, obtains amendment image；Binary conversion treatment is carried out to the amendment image, obtains palmprint image.In embodiment provided by the present application, image is pre-processed, easier palmmprint is extracted so that subsequent, even if the palmmprint for obtaining palm more highlights, or remove noise in palm image, the extraction for avoiding influence of noise palmmprint is more clear the display effect of image, improves the subsequent accuracy extracted to palm palmmprint.Curvature value is carried out by pixel each in palmprint image to be modified to screen pixel, further improves the success rate that palmmprint lines extracts.

Description

Palmprint extraction method, device, storage medium, and server

technical field

The invention relates to the technical field of biometric identification and palmprint identification, and in particular to a palmprint extraction method, device, storage medium and server.

Background technique

In recent years, the industry and academia have been continuously working on improving the verification effect of identity information to meet the stringent requirements for identifying people's identities in different fields such as access control, aviation security, and electronic banking. Methods based on biometric identification are attracting more and more attention, and palmprint recognition is one of the most representative biometric identification methods. The palmprint recognition method has many advantages, such as high discrimination, strong robustness, and user-friendliness. Palm print refers to the skin texture on the surface of the palm. The palm print of the hand is divided into normal and abnormal lines due to various reasons, and abnormal lines may appear "ten"-shaped lines, "well"-shaped lines, "rice"-shaped lines and other lines. Palm print features are immutable, permanent, and unique to human individuals.

At present, there are many methods for extracting palm prints, but the effect is general, mainly limited by imaging factors, and is greatly affected by lighting conditions, that is, the resistance to light is poor, and the resistance to people with unclear palm prints is poor. .

SUMMARY OF THE INVENTION

In order to overcome the above technical problems, especially in the process of palmprint extraction, the resistance to light is poor and the resistance to people with unclear palmprints is poor, the following technical solutions are proposed:

A palmprint extraction method provided by an embodiment of the present invention includes:

Obtain the palm image, preprocess the palm image, and obtain the palm print image to be corrected;

Calculate the curvature value of each characteristic pixel point in the palmprint image to be corrected, and correct the palmprint image to be corrected based on the curvature value of the characteristic pixel point to obtain a corrected image;

Binarization is performed on the corrected image to obtain a palmprint image.

Optionally, calculating the curvature value of each feature pixel in the palmprint image to be corrected, and modifying the palmprint image to be corrected based on the curvature value of the feature pixel to obtain a corrected image, include:

Cut the palmprint image to be corrected according to the preset cutting direction and the preset pixel interval to obtain n cutting lines, where n is a positive integer;

Calculate the curvature value of each pixel point on each of the cutting lines, and correct the palmprint image to be corrected based on the curvature value to obtain a corrected image, and each pixel point on the cutting line is the characteristic pixel point.

Optionally, the correction of the palmprint image to be corrected based on the curvature value to obtain a corrected image includes:

Determining the pixel point with the curvature value greater than zero as the evaluation pixel point, and determining the area where the continuous evaluation pixel point is located as the local palmprint area;

Obtain the width of the local palmprint area, multiply the width by the curvature value of each of the evaluation pixels in the local palmprint area, and obtain the width of each evaluation pixel in the local palmprint area. assessment score;

Adjust the pixel value of the evaluation pixel point according to the evaluation score, obtain the modified pixel value of each evaluation pixel point, modify the palmprint image to be modified based on the modified pixel value, and obtain the modified image .

Optionally, after obtaining the corrected image, the method further includes:

For each pixel in the corrected image, obtain the pixel value of the first adjacent pixel on one side of the pixel and adjacent to the pixel, and the first interval pixel that is separated from the pixel by one pixel The pixel value of the point; wherein, the first interval pixel point and the first adjacent pixel point are located on the same side of the pixel point;

Obtain the pixel value of the second adjacent pixel point on the other side of the pixel point and adjacent to the pixel point, and the pixel value of the second interval pixel point separated from the pixel point by one pixel point; Two spaced pixels and the second adjacent pixel are located on the same side of the pixel, and the one side is opposite to the other side;

The pixel value of the pixel point is modified according to the pixel value of the one side and the other side.

Optionally, the preset cutting directions include at least 2 directions; and performing binarization processing on the corrected image to obtain a palmprint image includes:

Taking the corrected image obtained according to each of the cutting directions as the image to be synthesized;

Comparing the pixel values of the same pixel in each of the images to be synthesized, the maximum pixel value of the pixel is determined as the synthesized pixel value of the pixel in the synthesized image;

obtaining a composite image according to the composite pixel value of each pixel;

Binarization processing is performed on the composite image to obtain the palmprint image.

Optionally, the described palm image is preprocessed to obtain the palm print image to be corrected, including:

Traverse the palm image, obtain the RGB component value of each pixel, and perform grayscale processing on the RGB component value of each pixel through a grayscale processing rule to obtain a grayscale image;

The grayscaled image is inverted, and the inverted grayscaled image is processed by filtering transformation to obtain the palmprint image to be corrected.

The embodiment of the present application also provides a palmprint extraction device, including:

The palmprint image acquisition module to be corrected is used to acquire the palm image, and preprocess the palm image to obtain the palmprint image to be corrected;

A corrected image obtaining module is used to calculate the curvature value of each characteristic pixel point in the palmprint image to be corrected, and correct the palmprint image to be corrected based on the curvature value of the characteristic pixel point to obtain a corrected image ;

The binarization processing module is configured to perform binarization processing on the corrected image to obtain a palmprint image.

Optionally, the modified image obtaining module includes:

a cutting unit, configured to cut the palmprint image to be corrected according to a preset cutting direction and a preset pixel interval to obtain n cutting lines, where n is a positive integer;

The curvature calculation unit is used to calculate the curvature value of each pixel point on each of the cutting lines, and based on the curvature value, the palmprint image to be corrected is corrected to obtain a corrected image, and each pixel point on the cutting line is the feature pixels.

Embodiments of the present invention further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the program is executed by a processor, the palmprint extraction method described in any one of the technical solutions is implemented.

The embodiment of the present invention also provides a server, including:

one or more processors;

memory;

one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, the one or more application programs are configured to execute The steps of the palmprint extraction method according to any technical solution.

Compared with the prior art, the present invention has the following beneficial effects:

1. A palmprint extraction method provided by an embodiment of the present application includes: acquiring a palm image, preprocessing the palm image to obtain a palmprint image to be corrected; Calculate the rate value, correct the palmprint image to be corrected based on the curvature value of the characteristic pixel point to obtain a corrected image; perform binarization processing on the corrected image to obtain a palmprint image. In the embodiments provided in this application, the image is preprocessed to make it easier to extract the palm prints later, that is, to make the palm prints of the palm more prominently displayed, or to remove the noise in the palm image, so as to avoid the noise affecting the palm prints. extract. Perform grayscale inversion processing on the grayscale image to obtain the palm image after grayscale inversion, thereby reducing the amount of original image data and improving the calculation efficiency in subsequent processing calculations; then grayscale the grayscaled image. Inversion processing makes the display effect of the image clearer and improves the accuracy of subsequent palm print extraction. Furthermore, in order to improve the success rate of palmprint pattern extraction, it is also necessary to calculate the curvature value of each pixel in the palmprint image to be corrected, and then correct the palmprint image to be corrected based on the curvature value of each pixel to obtain a correction. image. Correspondingly, the curvature is calculated based on the preset direction, so that there is a corresponding benchmark in the calculation process of the curvature value, and then it is convenient to screen pixels based on the benchmark to obtain pixels with suitable pixels, or based on the curvature of some pixels. The value is to correct the pixels of other pixels in the palm print image to be corrected to make the palm print more prominent, improve the palm print image's resistance to light, and improve the palm print image's resistance to people with unclear palm prints. Obtaining an image that is convenient for extraction and correction of palm lines, further improving the success rate of palm line extraction.

2. A palmprint extraction method provided by an embodiment of the present application, wherein the calculation of the curvature value of each characteristic pixel point in the palmprint image to be corrected is performed, and the Correcting the palmprint image for correction to obtain a corrected image includes: cutting the palmprint image to be corrected according to a preset cutting direction and a preset pixel interval to obtain n cutting lines, where n is a positive integer; The curvature value of each pixel point on the cutting line is used to correct the palmprint image to be corrected based on the curvature value to obtain a corrected image. For each cutting line obtained by cutting the palmprint image to be corrected, the curvature value of each pixel on the cutting line is calculated, and the curvature value is used to judge whether the pixel belongs to the pixel point on the palmprint, and then it is convenient to perform the following steps based on the curvature value. The above-mentioned process of selecting and correcting the pixel points further realizes the correction of the pixels of other pixel points in the palmprint image to be corrected, so as to obtain a corrected image.

Additional aspects and advantages of the present invention will be set forth in part in the following description, which will be apparent from the following description, or may be learned by practice of the present invention.

Description of drawings

The above and/or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of embodiments taken in conjunction with the accompanying drawings, wherein:

1 is a schematic flowchart of an embodiment of a typical embodiment of a palmprint extraction method of the present invention;

2 is a schematic diagram of a palmprint image to be corrected in a typical embodiment of the palmprint extraction method of the present invention;

3 is a schematic diagram of a palmprint image in a typical embodiment of the palmprint extraction method of the present invention;

4 is a schematic structural diagram of a typical embodiment of a palmprint extraction device of the present invention;

FIG. 5 is a schematic structural diagram of an embodiment of the server of the present invention.

Detailed ways

The following describes in detail the embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, but not to be construed as a limitation of the present invention.

It will be understood by those skilled in the art that the singular forms "a", "an", "the" and "the" as used herein can include the plural forms as well, unless expressly stated otherwise. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of the stated features, integers, steps, operations, but does not exclude the presence or addition of one or more other features, integers, steps, operations.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It should also be understood that terms, such as those defined in a general dictionary, should be understood to have meanings consistent with their meanings in the context of the prior art and, unless specifically defined as herein, should not be interpreted in idealistic or overly formal meaning to explain.

Those skilled in the art should understand that the concepts of "application", "application program", "application software" and similar expressions in the present invention are the same concepts known to those skilled in the art, and refer to a series of computer instructions and related concepts. Data resources are organically constructed computer software suitable for electronic execution. Unless otherwise specified, the naming itself is not limited by the type or level of programming language, nor by the operating system or platform on which it runs. Of course, such concepts are also not limited by any form of terminal.

The palmprint extraction provided by the embodiments of the present application is applied to a scenario including a server and a client, wherein the server and the client are connected through a network, the client is used to collect palm images, and the collected palm images Sent to the server, the client can be, but not limited to, a camera, camera, scanner or palm image acquisition device with other photographing functions; the server is used to extract the palm pattern from the palm image, and the server can use an independent A server or a server cluster consisting of multiple servers is implemented. The image processing method provided by the embodiment of the present application is applied to the server.

The palmprint extraction method provided in the embodiment of the present application, in one implementation manner, as shown in FIG. 1 , includes: S100 , S200 , and S300 .

S100: Obtain a palm image, preprocess the palm image, and obtain a palm print image to be corrected;

S200: Calculate the curvature value of each feature pixel in the palmprint image to be corrected, and modify the palmprint image to be corrected based on the curvature value of the feature pixel to obtain a corrected image;

S300: Perform binarization processing on the corrected image to obtain a palmprint image.

In the embodiment provided in this application, the palm image is obtained, and the palm image may be the image of the entire palm, or may be the palm image excluding the fingers. The palm prints in part of the palms are extracted, and of course the entire palm prints can also be extracted. In order to improve the recognition rate of palmprints in the palm image and avoid noise affecting the extraction of palmprints, the palm image needs to be preprocessed, and the preprocessed image is the palmprint image to be corrected. In the embodiments provided in this application, the image is preprocessed to make it easier to extract the palm prints later, that is, to make the palm prints of the palm more prominently displayed, or to remove the noise in the palm image, so as to avoid the noise affecting the palm prints. extract. For example: grayscale the collected color palmprint image: read a palm image in RGB mode (three-element (red, green, blue) mode), and use rgb2gray() to convert the grayscale image to the image mode. It is a grayscale image; make a grayscale histogram of the image, find the valley value T between the two peaks of the grayscale histogram as the threshold value of the binarization segmentation, and convert the grayscale image into a binary image; use 3*3 The square median filter window of the image is median filtered; the contour of the palm image is extracted using the boundary tracking algorithm. Grayscale the collected color palmprint images. On the basis of the above, the grayscale image is binarized. The grayscale image binarization is to select a grayscale image with 256 grayscale levels (the value range of grayscale value is 0 to 255) through an appropriate threshold, and reset the grayscale value of the pixel on the image to 0. Or 255, which renders the entire image in black and white. The grayscale histogram of the grayscale image is made by the imhist() function. The horizontal axis of the grayscale histogram represents the grayscale value range of the grayscale image, ranging from 0 to 255, and the vertical axis represents the number of times a certain grayscale value appears on the image. Because the palm grayscale image has a single foreground (only the palm) and a simple background, the grayscale histogram distribution of the palm grayscale image shows a significant bimodal characteristic. Therefore, any horizontal axis value corresponding to the middle part of the valley between the two peaks of the grayscale histogram is selected as the threshold value of the binarization segmentation, and the point on the grayscale image whose grayscale value is greater than the threshold value is reset to 255 (white), the points smaller than this threshold are reset to 0 (black), so that the grayscale image is converted into a binary image, that is, an image with a black background and a white foreground palm; a smooth image. Since the boundary of the binary image obtained after the grayscale image is binarized is often very unsmooth, in order to obtain a smoother palm contour line, the binarized image needs to be smoothed to make the edge of the image sharp. The "glitch" is flattened. Since the collected palm image has simple content and few details, a simple median filtering method is adopted, and a 3*3 square median filtering window is used, which can filter out noise and keep the edges from being blurred. The contour of the binary image, namely the palm contour, is extracted using the contour tracking algorithm. Then, the image surrounded by the contour of the palm and after the above-mentioned processing is used as the palm print image to be corrected (as shown in FIG. 2 ).

Combined with the image preprocessing process of the previous example, the detailed process of preprocessing the palm image is as follows:

Further, on the basis of the foregoing, the pixels in the palm image are traversed to obtain the RGB component value of each pixel.

Specifically, the pixels in the palm image are traversed according to a preset traversal method, and the RGB component value of each pixel is obtained, wherein R, G, and B represent the colors of red, green, and blue channels, respectively.

Wherein, the preset traversal method may specifically take the upper left pixel point of the palm image as the starting point, and perform line-by-line traversal in the order from top to bottom and from left to right, or it may be traversed from the position of the center line of the palm image to both sides at the same time, It can also be other traversal methods, which are not limited here.

Further, on the basis of the foregoing, according to the RGB component values of the pixel points, the palm image is subjected to grayscale processing according to formula (1) to obtain a grayscale image:

g(x,y)=k ₁ *R(x,y)+k ₂ *G(x,y)+k ₃ *B(x,y) (1)

Among them, x and y are the abscissa and ordinate of each pixel in the palm image, g(x,y) is the grayscale value of the pixel (x,y) after grayscale processing, R(x,y) is the color component of the R channel of the pixel point (x, y), G(x, y) is the color component of the G channel of the pixel point (x, y), and B(x, y) is the pixel point (x, y) The color components k ₁ , k ₂ , and k ₃ of the B channel are the ratio parameters corresponding to the R channel, the G channel, and the B channel, respectively. The formula (1) is the aforementioned gray scale rule.

In the embodiment of the present invention, in order to achieve accurate extraction of the information content in the palm image, it is first necessary to perform grayscale processing on the palm image, wherein the parameter values of k ₁ , k ₂ , k ₃ and σ can be determined according to the actual application It needs to be set, which is not limited here. By adjusting the value range of k ₁ , k ₂ , and k ₃ , the ratio of R channel, G channel and B channel can be adjusted respectively.

The RGB model is a commonly used way of expressing color information. It uses the brightness of the three primary colors of red, green and blue to quantitatively represent the color. This model, also known as the additive color mixing model, is a method in which RGB three-color lights are superimposed on each other to achieve color mixing, so it is suitable for the display of luminous bodies such as monitors.

Grayscale means that in the RGB model, if R=G=B, the color represents only one grayscale color, and the value of R=G=B is called the grayscale value. Therefore, each pixel of the grayscale image has only one grayscale color. One byte is required to store the grayscale value, and the grayscale range is 0-255.

It should be noted that, in the embodiment of the present invention, the gray value is weighted by formula (1). There are no restrictions here.

Further, on the basis of the foregoing, grayscale inversion processing is performed on the grayed image to obtain a palm image after grayscale inversion.

Specifically, in the aforementioned process, traverse each pixel in the acquired graying image, obtain the pixel value of each pixel, perform grayscale inversion processing on the graying image, and convert the pixel in the graying image The pixel value range of [0, 255] is transformed to [255, 0], that is, the pixel value of the pixel point is adjusted from 0 to 255, and the pixel value of the pixel point is adjusted from 255 to 0, so as to make the original grayscale image The white pixels become black pixels, and the original black pixel tops become white pixels. After grayscale inversion processing, a palm image after grayscale inversion is obtained, that is, the palm print image to be corrected.

It should be noted that, in order to facilitate the calculation in different environments, the value range of the pixel point can be further compressed from [0, 255] to [0, 1], that is, the pixel value of each pixel point is divided by 255 to get The compressed pixel value, for example, the pixel value of a pixel with a pixel value of 1 is 1/255, the pixel value of a pixel with a pixel value of 254 is 254/255, and the pixel values of other pixels are transformed by And so on.

For example: in combination with the above description, in the MATLAB tool, you can directly call the imadjust function to perform grayscale inversion processing on the grayed image, and transform the pixel value range in the image from the original [0, 255] to [255, 0], and then The compression transformation is [1,0] to generate a palm image whose grayscale is opposite to that of the grayed image, that is, the palmprint image to be corrected.

In this embodiment, by traversing the pixel points in the palm image and obtaining the RGB component value of the corresponding pixel point, according to the obtained RGB component value of each pixel point, the palm image is grayed by using formula (1), Set the pixel value range of pixel points in the image between 0 and 255, thereby reducing the amount of original image data and improving the computational efficiency in subsequent processing calculations; and then performing grayscale inversion processing on the grayscaled image. , so that the display effect of the image is clearer and the accuracy of subsequent palm print extraction is improved.

After the palmprint image to be corrected is obtained, the curvature value of each pixel in the palmprint image to be corrected is calculated, and then the palmprint image to be corrected is corrected based on the curvature value of each pixel to obtain a corrected image. In the specific curvature calculation process, the curvature is calculated based on the preset direction, so that there is a corresponding benchmark in the calculation process of the curvature value, and then it is convenient to screen pixels based on the benchmark to obtain pixels with suitable pixels, or Pixels of other pixels in the palmprint image to be corrected are corrected based on the curvature values of some pixels to obtain a corrected image that is convenient for palmprint extraction. Furthermore, in order to improve the success rate of palmprint extraction, the present application provides In this embodiment, the corrected image also needs to be binarized. After this process, the palm pattern in the image can be clearly distinguished from other parts of the palm, and the palm pattern is obtained as shown in FIG. 3 . Correspondingly, in order to extract the palm texture, in order to make the pixel value of the pixel in the image only show 0 or 255, that is, the image only shows two colors of black or white, the enhanced image needs to be further binarized. After the above-mentioned processing, it can be seen from the comparison between Fig. 2 and Fig. 3 that the curvature of the palm print image to be corrected is screened and corrected by the pixel point curvature value. The palm print in Fig. 3 is more prominent than Fig. Image resistance to light, and improved palmprint image resistance to people with unclear palm prints.

Binarization is to set the pixel value of the pixel on the image to 0 or 255, that is, to present the entire image with an obvious visual effect of only black and white.

Specifically, scan and correct each pixel in the image, if the pixel value of the pixel is less than the preset pixel threshold, the pixel value of the pixel is set to 0, that is, the pixel becomes black; The pixel value of the pixel is greater than or equal to the pixel threshold of the preset value, then the pixel value of the pixel is set to 255, that is, the pixel becomes white, and a binarized image is obtained.

Optionally, calculating the curvature value of each feature pixel in the palmprint image to be corrected, and modifying the palmprint image to be corrected based on the curvature value of the feature pixel to obtain a corrected image, include:

Cut the palmprint image to be corrected according to the preset cutting direction and the preset pixel interval to obtain n cutting lines, where n is a positive integer;

Calculate the curvature value of each pixel point on each of the cutting lines, and correct the palmprint image to be corrected based on the curvature value to obtain a corrected image, and each pixel point on the cutting line is the characteristic pixel point.

In the embodiment of the present invention, the preset cutting direction may be horizontal cutting, vertical cutting, or cutting in other directions, which may be specifically set according to actual application requirements, which is not limited here. The preset pixel interval refers to a preset number of pixels as an interval, which can be 1 pixel as an interval or 5 pixels as an interval, and can also be set according to the needs of actual applications. , there is no restriction here. In order to better understand this step, a specific example is used below to illustrate. For example, if the obtained palm image is a rectangle, in the corrected image, the preset cutting direction is the vertical direction, and the preset pixel interval is 5 pixels. pixels, you will get 399 vertical cutting lines.

After the palmprint image to be corrected is cut according to the preset direction, the curvature value of each pixel on each cutting line is calculated. In the embodiment provided by this application, the calculation of the curvature value is performed by the following formula (2):

Among them, z is the pixel point on the cutting line, K(z) is the curvature value of the pixel point z, P _f (z) is the pixel point of the pixel point z, is the value of the second derivative of P _f (z), is the value of the first derivative of P _f (z).

Specifically, for each cutting line obtained by cutting the palmprint image to be corrected, calculate the curvature value of each pixel on the cutting line according to formula (2), and use the curvature value to judge whether the pixel belongs to the pixel on the palmprint , which facilitates the process of screening and correction of the aforementioned pixels based on the curvature value, and then realizes the correction of the pixels of other pixels in the palmprint image to be corrected, so as to obtain a corrected image. Correcting the palmprint image to obtain the corrected image will be described in detail later, and will not be repeated here.

Optionally, the correction of the palmprint image to be corrected based on the curvature value to obtain a corrected image includes:

Determining the pixel point with the curvature value greater than zero as the evaluation pixel point, and determining the area where the continuous evaluation pixel point is located as the local palmprint area;

Obtain the width of the local palmprint area, multiply the width by the curvature value of each of the evaluation pixels in the local palmprint area, and obtain the width of each evaluation pixel in the local palmprint area. assessment score;

Adjust the pixel value of the evaluation pixel point according to the evaluation score, obtain the modified pixel value of each evaluation pixel point, modify the palmprint image to be modified based on the modified pixel value, and obtain the modified image .

Combined with the above description, if the curvature value of the pixel point on the cutting line is greater than 0, it means that the pixel point is a pixel point on the palm print, and it is used as an evaluation pixel point. If the curvature value of the pixel point is less than or equal to 0, it means that the pixel point is This pixel does not belong to the pixel on the palm print. In addition, the local palmprint region is composed of continuous pixels with a curvature value greater than 0, that is, continuous evaluation pixels.

In this embodiment of the present invention, since the local palmprint region is composed of continuous pixels with a curvature value greater than 0, its width may be the number of continuous pixels with a curvature value greater than 0. For example, if the continuous pixels with a curvature value greater than 0 If the number of pixels is 5, the width of the local palmprint area is 5.

For each evaluation pixel, multiply the width of the local palmprint region including the evaluation pixel by the curvature value of the evaluation pixel, and use the multiplication result as the evaluation score of the evaluation pixel.

Specifically, the evaluation score of the evaluation pixel is calculated by formula (3):

S _r (z _i )=k(z _i )*W _r (3)

Among them, _zi is the ith evaluation pixel, i is a positive number greater than 0, S _r (z _i ) is the evaluation score of the ith evaluation pixel, k(z _i ) is the ith evaluation pixel Curvature value, W _r is the width of the local palmprint region containing _zi .

In the embodiment of the present invention, for each evaluation pixel point, the original pixel value of each evaluation pixel point and its corresponding evaluation score are added, and the obtained sum is used as the corrected pixel value of the evaluation pixel point. The corrected pixel value of the pixel point, the corrected pixel value is obtained after adjusting the pixel value of each evaluation pixel point, and the pixel value of the same pixel point in the palmprint image to be corrected is updated based on the corrected pixel value, so that the palmprint The points on the area become more obvious, the recognition degree of the palmprint area is improved, and the palmprint area and the non-palmprint area can be better identified, and the corrected image as described above is obtained.

Specifically, the corrected pixel value of the evaluation pixel is calculated by formula (4):

V _a '(x,y)=V _a (x,y)+S _r (z _a ) (4)

Among them, x and y are the abscissa and ordinate of the a-th evaluation pixel in the palmprint image to be corrected, a is a positive number greater than 0, za is the _a -th evaluation pixel, V _a '(x,y ) is the corrected pixel value of the a-th evaluation pixel, V _a (x, y) is the pixel value of the a-th evaluation pixel, and S _r (z _a ) is the evaluation score of the a-th evaluation pixel.

It should be noted that, if the calculated corrected pixel value of the evaluation pixel exceeds the maximum pixel value, the corrected pixel value is set as the maximum pixel value.

Optionally, after obtaining the corrected image, the method further includes:

For each pixel in the corrected image, obtain the pixel value of the first adjacent pixel on one side of the pixel and adjacent to the pixel, and the first interval pixel that is separated from the pixel by one pixel The pixel value of the point; wherein, the first interval pixel point and the first adjacent pixel point are located on the same side of the pixel point;

Obtain the pixel value of the second adjacent pixel point on the other side of the pixel point and adjacent to the pixel point, and the pixel value of the second interval pixel point separated from the pixel point by one pixel point; The spaced pixel points and the second adjacent pixel points are located on the same side of the pixel point, and the one side is opposite to the other side;

The pixel value of the pixel point is modified according to the pixel value of the one side and the other side.

In the embodiment of the present invention, the pixel value of each pixel is corrected according to formula (5):

C(x,y)=min{max(V(x+1,y),V(x+2,y)),max(V(x-1,y),V(x-2,y)) } (5)

Among them, x and y are the abscissa and ordinate of each pixel in the palm image, V(x,y) is the pixel value of the pixel (x,y) in the corrected image, and C(x,y) is the pixel (x,y) The corrected pixel value.

Specifically, select the two adjacent pixels (x-1, y) and (x-2, y) on the left side of the pixel point (x, y) in the corrected image and the two adjacent pixels on the right side (x+ 1, y), (x+2, y), if (x, y) is the same as the pixel value of the pixels on both sides, it will not be processed; If the pixel values of the two pixel points are different, select the larger pixel value among the pixel values of the two pixel points on the left, and then select the larger pixel value among the pixel values of the two pixel points on the right, and finally compare the larger pixel value on the left side. The pixel value of , and the larger pixel value on the right, select the smaller pixel value of the two to correct the pixel point (x, y).

It should be noted that, if the pixel point is the pixel point of the image boundary, only the pixel value of one side is compared. For example, if the pixel point is located on the left boundary of the image, select the two adjacent pixels on the right side of the pixel point. The larger pixel value among the pixel values, the pixel value of the pixel point is corrected; if the pixel point is located on the right edge of the image, the larger pixel value among the pixel values of the two adjacent pixel points on the left side of the pixel point is selected, and the pixel value of the pixel point is selected. The pixel value of the pixel point is corrected.

In this embodiment, if the pixel value of the pixel point (x, y) is small and the pixel values on both sides are large, the pixel value of the pixel point (x, y) is increased by formula (5), so that the pixel value of the pixel point (x, y) is increased. The point and the pixels on both sides can be connected to form a texture; if the pixel value of the pixel point (x, y) is large and the pixel value on both sides is small, the pixel is considered to be noise. The extraction of the palm print causes interference, and the pixel value of the pixel point (x, y) is reduced by formula (5) to eliminate the noise in the palm print image, so that the palm print area becomes more obvious and improves the palm print image. Discrimination, but also improve the accuracy of subsequent palmprint extraction.

Optionally, the preset cutting directions include at least 2 directions; and performing binarization processing on the corrected image to obtain a palmprint image includes:

Taking the corrected image obtained according to each of the cutting directions as the image to be synthesized;

Comparing the pixel values of the same pixel in each of the images to be synthesized, the maximum pixel value of the pixel is determined as the synthesized pixel value of the pixel in the synthesized image;

obtaining a composite image according to the composite pixel value of each pixel;

Binarization processing is performed on the composite image to obtain the palmprint image.

In one embodiment of the examples provided in this application, the preset cutting directions include at least two directions, that is, the palmprint image to be corrected can be processed by image processing based on two or more different cutting directions to obtain the corrected palmprint image. image. The preset cutting direction can specifically include 4 directions of 45°, 90°, 135° and 180°, but it is not limited to this. It can also include other directions, which can be set according to the needs of practical applications. limit.

In this embodiment of the present invention, for each specific cutting direction, an image cut according to the preset cutting direction according to the corrected image is used as the image to be synthesized. For example, if the preset cutting direction includes 45°, 90°, 135° and 180° in a total of 4 directions, the corrected image obtained with a cutting direction of 45° is one image to be synthesized, the corrected image obtained with a cutting direction of 90° is another image to be synthesized, and so on for other directions. Four images to be synthesized are obtained.

Specifically, according to the to-be-synthesized image obtained by the foregoing process, by comparing the pixel values of the pixels at the same position in each to-be-synthesized image, the largest pixel value is selected as the pixel value of the pixel at the corresponding position of the synthesized image, and a composite image is obtained. image.

On the basis of the composite image, in order to make the pixel values of the pixels in the image only show 0 or 255, that is, the image only shows two colors of black or white, it is necessary to further binarize the composite image to obtain palm prints. image.

Specifically, scan each pixel in the synthesized image, if the pixel value of the pixel is less than the preset pixel threshold, the pixel value of the pixel is set to 0, that is, the pixel turns black; If the pixel value of the point is greater than or equal to the pixel threshold value of the preset value, the pixel value of the pixel point is set to 255, that is, the pixel point becomes white, and the palmprint image is obtained.

In this embodiment, different images to be synthesized are obtained according to different cutting directions, the pixel values of each pixel at the same position in each image to be synthesized are compared, and the maximum pixel value of each pixel is selected as the synthesized image The pixel value of the pixel at the corresponding position in the image is synthesized, and finally the synthesized image is binarized to obtain the palmprint image. Since there may be errors in palmprint extraction from the corrected image obtained in only one cutting direction, the palmprint image obtained by binarizing the synthesized image is synthesized by synthesizing the images to be synthesized in multiple cutting directions. Texture extraction can effectively reduce errors, achieve accurate palmprint extraction, and improve the accuracy of palmprint extraction.

Optionally, the preprocessing of the palm image includes:

Gabor filter transformation is performed on the palm image to obtain the palm print image to be corrected.

In the embodiment of the present invention, in order to further improve the quality of the palm image obtained, the image is enhanced by the method of Gabor filter transformation, and finally a processed enhanced image is obtained.

Specifically, a convolution operation is performed on the palm image according to the Gabor filter function, and an enhanced image is obtained through the result of the convolution operation. Among them, the convolution operation refers to using a convolution kernel to perform a series of operations on each pixel in the palm image. The convolution kernel is a preset matrix template, which is used for operation with the palm image. A square grid structure, such as a 3*3 matrix, each element in the matrix has a preset weight value. When using the convolution kernel for calculation, place the center of the convolution kernel on the target to be calculated On the pixel point, the product between the weight value of each element in the convolution kernel and the pixel value of the image pixel point it covers is calculated and summed, and the result obtained is the new pixel value of the target pixel point.

The Gabor filter transform belongs to the windowed Fourier transform, and the Gabor function can extract the relevant features of the image in different scales and directions in the frequency domain to achieve the enhancement effect on the image. Specifically, the palm image processing process through Gabor filtering is as follows: according to formula (6), the palm image is subjected to Gabor filtering transformation:

x'=x cosθ+y sinθ

y'=-x sinθ+y cosθ (6)

in, is the Gabor filter function, x and y are the abscissa and ordinate of the pixel in the palm image, λ is the preset wavelength, θ is the preset direction, is the phase offset, σ is the standard deviation of the Gaussian factor of the gabor function, γ is the aspect ratio, U(x,y) is the enhanced image, I(x,y) is the palm image, For the tensor product operation, x' and y' are the abscissa and ordinate of the pixel point (x, y) in the palm image rotated according to θ.

Specifically, using the preset wavelength and the preset direction, the palm image is transformed by the Gabor filter function of formula (5), so as to filter out the high-frequency wave of the palm image, leaving only the low-frequency part, in the preset direction The low-frequency wave is filtered out, leaving only the high-frequency part, and finally the image becomes highlighted, that is, the enhanced image obtained after the Gabor filter transformation.

Wherein, the preset wavelength λ may be 1, and may also be set according to actual requirements, which is not limited here. The preset direction θ can be selected as 0, For these 8 directions, other directions can also be selected, which can be selected according to actual application needs, and there is no limitation here.

In this embodiment, the Gabor filter transformation is performed on the palm image by formula (6), the image can be quickly highlighted, and the effect of image enhancement can be achieved, thereby improving the image quality of the palm image and identifying the lines in the palm image. In order to achieve accurate positioning when palm pattern extraction is performed on low-quality palm images collected by low-end palmprint collection equipment, the accuracy of palmprint extraction can be improved, and the applicability of different palmprint collection equipment can also be improved. .

In the embodiments provided in the present application, combined with the foregoing technical solutions, it can be achieved that the palmprint image to be corrected is obtained by performing Gabor filter transformation on the palm image, wherein the palmprint image to be corrected is the enhanced image after Gabor filter transformation, Cut the enhanced image and obtain n cutting lines, calculate the curvature value of each pixel on each cutting line, obtain pixels with a curvature value greater than 0 as evaluation pixels, and obtain continuous pixels with a curvature value greater than zero The area where it is located is regarded as the local palmprint area, and the product of the curvature value of the evaluation pixel and the width of the local palmprint area where the evaluation pixel is located is used to calculate the evaluation score for each evaluation pixel, and then use the evaluation score to evaluate the evaluation score. The pixel value of the pixel is adjusted, the corrected pixel value of each evaluation pixel is obtained, and the pixel on the enhanced image is updated, and finally the updated enhanced image is binarized to obtain a palmprint image. On the one hand, the image quality of the palm image is improved through Gabor filter transformation, so that the accuracy of palmprint recognition can be improved when extracting palmprints, so as to realize palmprint identification of low-quality palm images collected by low-end palmprint collection equipment. Accurate positioning can effectively improve the extraction accuracy of palm prints in palm images, and the applicability to a variety of different palm print collection devices; Recognition efficiency; and by calculating the evaluation score, the palmprint area and the non-palmprint area can be further accurately distinguished, thereby further improving the accuracy of palmprint extraction.

The embodiment of the present invention also provides a palmprint extraction device. In one embodiment, as shown in FIG. 4 , it includes: a palmprint image acquisition module 100 to be corrected, a corrected image acquisition module 200, and a binarization processing module 300:

The to-be-corrected palmprint image acquisition module 100 is configured to acquire a palm image, preprocess the palm image, and obtain the to-be-corrected palmprint image;

The corrected image obtaining module 200 is used for calculating the curvature value of each characteristic pixel in the palmprint image to be corrected, and correcting the palmprint image to be corrected based on the curvature value of the characteristic pixel to obtain a corrected image. image;

The binarization processing module 300 is configured to perform binarization processing on the corrected image to obtain a palmprint image.

Further, as shown in FIG. 4 , a palmprint extraction device provided in an embodiment of the present invention further includes: a cutting unit 210, configured to cut the palmprint image to be corrected according to a preset cutting direction and a preset pixel interval, Obtain n cutting lines, where n is a positive integer; the curvature calculation unit 220 is used to calculate the curvature value of each pixel point on each of the cutting lines, and based on the curvature value, the palmprint image to be corrected is corrected, A corrected image is obtained, and each pixel on the cutting line is the characteristic pixel. The local palmprint area determination unit 221 is used to determine the pixel point with the curvature value greater than zero as the evaluation pixel point, and the area where the continuous described evaluation pixel point is located is determined as the local palmprint area; the evaluation score obtaining unit 222, for obtaining the width of the local palmprint area, and multiplying the width by the curvature value of each of the evaluation pixels in the local palmprint area to obtain each evaluation pixel in the local palmprint area The evaluation score of the point; the adjustment unit 223 is configured to adjust the pixel value of the evaluation pixel point according to the evaluation score, obtain the modified pixel value of each evaluation pixel point, and modify the pixel value based on the modified pixel value. The palmprint image to be corrected is obtained, and the corrected image is obtained. The first pixel value acquisition unit 230 is configured to, for each pixel in the corrected image, acquire the pixel value of the first adjacent pixel on one side of the pixel and adjacent to the pixel, and the pixel value corresponding to the pixel. The pixel value of the first interval pixel point whose point is separated by one pixel point; wherein, the first interval pixel point and the first adjacent pixel point are located on the same side of the pixel point; the second pixel value obtaining unit 240, using for obtaining the pixel value of the second adjacent pixel point on the other side of the pixel point and adjacent to the pixel point, and the pixel value of the second interval pixel point separated from the pixel point by one pixel point; The two spaced pixels and the second adjacent pixel are located on the same side of the pixel, and the one side is opposite to the other side; the correction unit 250 is configured to adjust the pixel value according to the pixel value of the one side and the other side Correct the pixel value of the pixel. The to-be-combined image obtaining unit 310 is used to use the corrected image obtained according to each of the cutting directions as the to-be-combined image; the comparison unit 320 is used to compare the pixel values of the same pixel in each of the to-be-combined images, The maximum pixel value of the pixel is determined as the composite pixel value of the pixel in the composite image; the composite image obtaining unit 330 is configured to obtain a composite image according to the composite pixel value of each pixel; the binarization processing unit 340, for performing binarization processing on the composite image to obtain the palmprint image. The grayscale unit 110 is configured to traverse the palm image, obtain the RGB component value of each pixel, and perform grayscale processing on the RGB component value of each pixel through a grayscale processing rule to obtain a grayscale The transforming unit 120 inverts the grayscale image, and processes the inverted grayscale image by filtering transformation to obtain the palmprint image to be corrected.

The palmprint extraction method and apparatus provided in the embodiments of the present invention can implement the above palmprint extraction method embodiments. For specific function implementation, please refer to the descriptions in the method embodiments, which will not be repeated here.

An embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the program is executed by a processor, the palmprint extraction method described in any one of the technical solutions is implemented. Wherein, the computer-readable storage medium includes but is not limited to any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM (Read-Only Memory, read-only memory), RAM (Random Access Memory, random access memory), EPROM (EraSable Programmable Read-Only Memory), EEPROM (Electrically EraSable Programmable Read-Only Memory), flash memory, magnetic card or light card. That is, a storage device includes any medium that stores or transmits information in a readable form by a device (eg, a computer, a cell phone), and may be a read-only memory, a magnetic disk or an optical disk, and the like.

A computer-readable storage medium provided by the embodiment of the present invention can implement the above-mentioned embodiments of the palmprint extraction method. In this application, by preprocessing the image, it is easier to extract the palmprint subsequently, that is, the palm print is obtained. The palm print is more prominently displayed, or the noise in the palm image is removed, so as to avoid the noise affecting the palm print extraction, so that the image display effect is clearer, and the accuracy of the subsequent palm print extraction is improved. Screening the pixels by the curvature value of each pixel in the palmprint image to be corrected further improves the success rate of palmprint extraction; a palmprint extraction method provided by an embodiment of the present application includes: acquiring a palm image, The image is preprocessed to obtain a palmprint image to be corrected; curvature values are calculated for each feature pixel in the palmprint image to be corrected, and the palmprint image to be corrected is performed based on the curvature value of the feature pixel. Correction is performed to obtain a corrected image; the corrected image is subjected to binarization processing to obtain a palmprint image. In the embodiment provided in this application, the palm image is obtained, and the palm image may be the image of the entire palm, or may be the palm image excluding the fingers. The palm prints in part of the palms are extracted, and of course the entire palm prints can also be extracted. In order to improve the recognition rate of palmprints in the palm image and avoid noise affecting the extraction of palmprints, the palm image needs to be preprocessed, and the preprocessed image is the palmprint image to be corrected. In the embodiments provided in this application, the image is preprocessed to make it easier to extract the palm prints later, that is, to make the palm prints of the palm more prominently displayed, or to remove the noise in the palm image, so as to avoid the noise affecting the palm prints. extract. Further, on the basis of the foregoing, the pixels in the palm image are traversed to obtain the RGB component value of each pixel. Specifically, the pixels in the palm image are traversed according to a preset traversal method, the RGB component value of each pixel is obtained, and the palm image is grayed according to the RGB component value of the pixel to obtain a grayed image. ; Perform grayscale inversion processing on the grayscale image to obtain a palm image after grayscale inversion, so that the original white pixels in the grayscale image become black pixels, and the top of the original black pixels become white pixels. After grayscale inversion processing, a palm image after grayscale inversion is obtained, that is, the palm print image to be corrected. In this embodiment, the pixel value range of the pixel points in the image is set between 0 and 255, which reduces the amount of original data of the image and improves the calculation efficiency in the subsequent processing calculation; The degree inversion processing makes the display effect of the image clearer and improves the accuracy of subsequent palm print extraction. After the palmprint image to be corrected is obtained, the curvature value of each pixel in the palmprint image to be corrected is calculated, and then the palmprint image to be corrected is corrected based on the curvature value of each pixel to obtain a corrected image. In the specific curvature calculation process, the curvature is calculated based on the preset direction, so that there is a corresponding benchmark in the calculation process of the curvature value, and then it is convenient to screen pixels based on the benchmark to obtain pixels with suitable pixels, or Pixels of other pixels in the palmprint image to be corrected are corrected based on the curvature values of some pixels to obtain a corrected image that is convenient for palmprint extraction. Furthermore, in order to improve the success rate of palmprint extraction, the present application provides In the embodiment of the invention, the corrected image also needs to be binarized. After this process, the palm pattern in the image can be clearly distinguished from other parts of the palm to obtain the palm pattern. In addition, in another embodiment, the present invention further provides a server, as shown in FIG. 5 , the server processor 503 , the memory 505 , the input unit 507 , and the display unit 509 and other devices. Those skilled in the art can understand that the structural components shown in FIG. 5 do not constitute a limitation to all servers, and may include more or less components than those shown, or combine some components. The memory 505 can be used to store the application program 501 and various functional modules, and the processor 503 executes the application program 501 stored in the memory 505 to execute various functional applications and data processing of the device. Memory 505 may be internal memory or external memory, or include both. Internal memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, or random access memory. External storage may include hard disks, floppy disks, ZIP disks, U disks, magnetic tapes, and the like. The memory disclosed herein includes, but is not limited to, these types of memory. The memory 505 disclosed in the present invention is only an example and not a limitation.

The input unit 507 is used to receive input of signals, as well as personal information and related physical condition information input by the user. The input unit 507 may include a touch panel and other input devices. The touch panel can collect the touch operations of the customer on or near it (such as the customer's operation on or near the touch panel with fingers, stylus, etc. The program drives the corresponding connection device; other input devices may include, but are not limited to, one or more of physical keyboards, function keys (such as playback control keys, switch keys, etc.), trackballs, mice, and joysticks. The display unit 509 may be used to display information input by the customer or information provided to the customer and various menus of the computer device. The display unit 509 may take the form of a liquid crystal display, an organic light emitting diode, or the like. The processor 503 is the control center of the computer equipment, using various interfaces and lines to connect various parts of the entire computer, by running or executing the software programs and/or modules stored in the memory 503, and calling the data stored in the memory. Various functions and processing data. The one or more processors 503 shown in FIG. 5 can execute and realize the functions of the palmprint image acquisition module 100 to be corrected, the functions of the corrected image acquisition module 200 and the functions of the binarization processing module 300 shown in FIG. 4 . , the function of the cutting unit 210, the function of the curvature calculation unit 220, the function of the local palmprint area determination unit 221, the function of the evaluation score obtaining unit 222, the function of the adjustment unit 223, the function of the first pixel value obtaining unit 230, the second function Function of pixel value acquisition unit 240, function of correction unit 250, function of to-be-synthesized image acquisition unit 310, function of comparison unit 320, function of combined image acquisition unit 330, function of binarization processing unit 340, grayscale unit Function of 110, function of transformation unit 120.

In one embodiment, the server includes one or more processors 503, and one or more memories 505, one or more application programs 501, wherein the one or more application programs 501 are stored in the memory 505 and configured to be executed by the one or more processors 503, and the one or more application programs 301 are configured to execute the palmprint extraction method described in the above embodiments.

A server provided by an embodiment of the present invention can implement the above embodiments of the palmprint extraction method. In this application, by preprocessing the image, it is easier to extract the palmprint subsequently, that is, the palmprint of the palm is more easily extracted. Highlight or remove the noise in the palm image, so as to avoid the noise affecting the palm print extraction, so that the display effect of the image is clearer, and the accuracy of the subsequent palm print extraction is improved. Screening the pixels by the curvature value of each pixel in the palmprint image to be corrected further improves the success rate of palmprint extraction; a palmprint extraction method provided by an embodiment of the present application includes: acquiring a palm image, The image is preprocessed to obtain a palmprint image to be corrected; curvature values are calculated for each feature pixel in the palmprint image to be corrected, and the palmprint image to be corrected is performed based on the curvature value of the feature pixel. Correction is performed to obtain a corrected image; the corrected image is subjected to binarization processing to obtain a palmprint image. In the embodiment provided in this application, the palm image is obtained, and the palm image may be the image of the entire palm, or may be the palm image excluding the fingers. The palm prints in part of the palms are extracted, and of course the entire palm prints can also be extracted. In order to improve the recognition rate of palmprints in the palm image and avoid noise affecting the extraction of palmprints, the palm image needs to be preprocessed, and the preprocessed image is the palmprint image to be corrected. In the embodiments provided in this application, the image is preprocessed to make it easier to extract the palm prints later, that is, to make the palm prints of the palm more prominently displayed, or to remove the noise in the palm image, so as to avoid the noise affecting the palm prints. extract. Further, on the basis of the foregoing, the pixels in the palm image are traversed to obtain the RGB component value of each pixel. Specifically, the pixels in the palm image are traversed according to a preset traversal method, the RGB component value of each pixel is obtained, and the palm image is grayed according to the RGB component value of the pixel to obtain a grayed image. ; Perform grayscale inversion processing on the grayscale image to obtain a palm image after grayscale inversion, so that the original white pixels in the grayscale image become black pixels, and the top of the original black pixels become white pixels. After grayscale inversion processing, a palm image after grayscale inversion is obtained, that is, the palm print image to be corrected. In this embodiment, the pixel value range of the pixel points in the image is set between 0 and 255, which reduces the amount of original data of the image and improves the calculation efficiency in the subsequent processing calculation; The degree inversion processing makes the display effect of the image clearer and improves the accuracy of subsequent palm print extraction. After the palmprint image to be corrected is obtained, the curvature value of each pixel in the palmprint image to be corrected is calculated, and then the palmprint image to be corrected is corrected based on the curvature value of each pixel to obtain a corrected image. In the specific curvature calculation process, the curvature is calculated based on the preset direction, so that there is a corresponding benchmark in the calculation process of the curvature value, and then it is convenient to screen pixels based on the benchmark to obtain pixels with suitable pixels, or Pixels of other pixels in the palmprint image to be corrected are corrected based on the curvature values of some pixels to obtain a corrected image that is convenient for palmprint extraction. Furthermore, in order to improve the success rate of palmprint extraction, the present application provides In the embodiment of the invention, the corrected image also needs to be binarized. After this process, the palm pattern in the image can be clearly distinguished from other parts of the palm to obtain the palm pattern.

The server provided by the embodiment of the present invention can implement the embodiment of the palmprint extraction method provided above. For specific function implementation, please refer to the description in the method embodiment, which will not be repeated here.

The above are only some embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made. It should be regarded as the protection scope of the present invention.

Claims (10)

1. a palmprint extraction method, is characterized in that, comprises: Obtain the palm image, preprocess the palm image, and obtain the palm print image to be corrected; Calculate the curvature value of each characteristic pixel point in the palmprint image to be corrected, and correct the palmprint image to be corrected based on the curvature value of the characteristic pixel point to obtain a corrected image; Binarization is performed on the corrected image to obtain a palmprint image. 2. The palmprint extraction method according to claim 1, wherein the calculation of the curvature value of each characteristic pixel point in the palmprint image to be corrected is performed based on the curvature value of the characteristic pixel point. The palmprint image to be corrected is corrected to obtain a corrected image, including: Cut the palmprint image to be corrected according to the preset cutting direction and the preset pixel interval to obtain n cutting lines, where n is a positive integer; Calculate the curvature value of each pixel point on each of the cutting lines, and correct the palmprint image to be corrected based on the curvature value to obtain a corrected image, and each pixel point on the cutting line is the characteristic pixel point. 3. The palmprint extraction method according to claim 2, wherein the correction of the palmprint image to be corrected based on the curvature value to obtain a corrected image comprises: Determining the pixel point with the curvature value greater than zero as the evaluation pixel point, and determining the area where the continuous evaluation pixel point is located as the local palmprint area; Obtain the width of the local palmprint area, multiply the width by the curvature value of each of the evaluation pixels in the local palmprint area, and obtain the width of each evaluation pixel in the local palmprint area. assessment score; Adjust the pixel value of the evaluation pixel point according to the evaluation score, obtain the modified pixel value of each evaluation pixel point, modify the palmprint image to be modified based on the modified pixel value, and obtain the modified image . 4. palmprint extraction method according to claim 1, is characterized in that, after described obtaining correction image, also comprises: For each pixel in the corrected image, obtain the pixel value of the first adjacent pixel on one side of the pixel and adjacent to the pixel, and the first interval pixel that is separated from the pixel by one pixel The pixel value of the point; wherein, the first interval pixel point and the first adjacent pixel point are located on the same side of the pixel point; Obtain the pixel value of the second adjacent pixel point on the other side of the pixel point and adjacent to the pixel point, and the pixel value of the second interval pixel point separated from the pixel point by one pixel point; The spaced pixel points and the second adjacent pixel points are located on the same side of the pixel point, and the one side is opposite to the other side; The pixel value of the pixel point is modified according to the pixel value of the one side and the other side. 5. The palmprint extraction method according to any one of claims 1 to 4, wherein the preset cutting directions include at least two directions; the modified image is subjected to binarization processing to obtain Palm print images, including: Taking the corrected image obtained according to each of the cutting directions as the image to be synthesized; Comparing the pixel values of the same pixel in each of the images to be synthesized, the maximum pixel value of the pixel is determined as the synthesized pixel value of the pixel in the synthesized image; obtaining a composite image according to the composite pixel value of each pixel; Binarization processing is performed on the composite image to obtain the palmprint image. 6. palmprint extraction method according to any one of claim 1 to 4, is characterized in that, described palm image is preprocessed, obtains palmprint image to be corrected, comprising: Traverse the palm image, obtain the RGB component value of each pixel, and perform grayscale processing on the RGB component value of each pixel through a grayscale processing rule to obtain a grayscale image; The grayscaled image is inverted, and the inverted grayscaled image is processed by filtering transformation to obtain the palmprint image to be corrected. 7. A palmprint extraction device, characterized in that, comprising: The palmprint image acquisition module to be corrected is used to acquire the palm image, and preprocess the palm image to obtain the palmprint image to be corrected; A corrected image obtaining module is used to calculate the curvature value of each characteristic pixel point in the palmprint image to be corrected, and correct the palmprint image to be corrected based on the curvature value of the characteristic pixel point to obtain a corrected image ; The binarization processing module is configured to perform binarization processing on the corrected image to obtain a palmprint image. 8. The palmprint extraction device according to claim 7, wherein the corrected image obtaining module comprises: a cutting unit, configured to cut the palmprint image to be corrected according to a preset cutting direction and a preset pixel interval to obtain n cutting lines, where n is a positive integer; The curvature calculation unit is used to calculate the curvature value of each pixel point on each of the cutting lines, and based on the curvature value, the palmprint image to be corrected is corrected to obtain a corrected image, and each pixel point on the cutting line is the feature pixels. 9. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the program is executed by a processor, the palmprint extraction method according to any one of claims 1 to 6 is realized . 10. A server, characterized in that, comprising: one or more processors; memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, the one or more application programs are configured to execute The steps of the palmprint extraction method according to any one of claims 1 to 6.