Method and system for removing cross grains of fingerprint image

Abstract

The invention discloses a fingerprint image cross grain removal method and system, and the method comprises the steps: S1, receiving a collected fingerprint image, and carrying out the preliminary cross grain removal of the collected fingerprint image, and obtaining a processed fingerprint image; s2, Gaussian filtering processing in the row direction and the column direction is carried out on the processed fingerprint image through a Gaussian filtering algorithm; s3, dividing the fingerprint image after Gaussian filtering into a plurality of image blocks according to a set size, calculating a gray average value and a gray variance of each image block, and dividing all the image blocks into foreground image blocks and background image blocks based on the gray variance; s4, calculating a pixel gray level distribution histogram in the foreground image block, obtaining a maximum value and a minimum value of the pixel gray level according to the pixel gray level distribution histogram to obtain a gray level mapping coefficient, and performing gray level mapping on the fingerprint image according to the gray level mapping coefficient; s5, adaptive contrast adjustment in the row direction and the column direction is carried out on the fingerprint image after gray mapping, and a final fingerprint image after cross grain removal is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a fingerprint image horizontal line removal method and system. BACKGROUND

[0002] With the development of fingerprint recognition technology, the application scenarios of fingerprint recognition technology are also increasing. Nowadays, in some scenarios, single fingerprint recognition function cannot meet the needs of people's work and life, and fingerprint recognition is often combined with traditional card swiping verification and face recognition, so that a machine can have multiple identification functions. Although it is relatively convenient to use, there are some new problems when multiple devices are combined together, one of which is the horizontal line of the fingerprint image. The main reason for the generation of the horizontal line of the fingerprint image is that the fingerprint collection device is affected by the card swiping coil during the collection of the fingerprint image. The horizontal line of the fingerprint image will have a significant impact on the fingerprint recognition algorithm based on minutiae, mainly in the aspects of not finding minutiae and introducing multiple false points, which will ultimately affect the fingerprint recognition experience.

[0003] Therefore, in view of the above technical problems, the present application provides a fingerprint image horizontal line removal method and system. SUMMARY

[0004] The purpose of the present application is to overcome the defects of the prior art and provide a fingerprint image horizontal line removal method and system. The method mainly fuses the image information of the local area with the image information of the surrounding area to preliminarily eliminate the conspicuous horizontal line information, then uses a horizontal direction and a vertical direction Gaussian filter operator to perform Gaussian filter processing on the image to smooth the fingerprint image, then performs overall gray mapping on the image to make the details of the image clearer and enhance the visual effect, and finally performs adaptive contrast adjustment on the image in the horizontal direction and the vertical direction to adjust the dynamic range of the image and improve the contrast of the image, so as to ultimately remove the horizontal lines in the fingerprint image and obtain a high-quality fingerprint image for fingerprint recognition.

[0005] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0006] A fingerprint image horizontal line removal method, comprising:

[0007] S1. receiving a collected fingerprint image, performing preliminary horizontal line removal processing on the collected fingerprint image to obtain a processed fingerprint image;

[0008] S2. using a Gaussian filter algorithm to perform Gaussian filter processing on the processed fingerprint image in the row direction and the column direction, respectively;

[0009] S3. Dividing the fingerprint image processed by Gaussian filtering into a plurality of image blocks according to a set size, calculating the gray mean value and the gray variance of each image block, and dividing all the image blocks into foreground image blocks and background image blocks based on the gray variance;

[0010] S4. Calculating the pixel gray distribution histogram in the foreground image block, obtaining the maximum value and the minimum value of the pixel gray according to the pixel gray histogram, obtaining the gray mapping coefficient, and performing gray mapping on the fingerprint image according to the gray mapping coefficient;

[0011] S5. Performing adaptive contrast adjustment on the fingerprint image after gray mapping in the row direction and the column direction respectively, and obtaining the final fingerprint image after removing the cross lines.

[0012] Further, the step S1 of performing preliminary cross line removal processing on the collected fingerprint image is specifically:

[0013] Calculating the gray mean value of each row of the fingerprint image to obtain an initial row gray mean value array;

[0014] Obtaining each data in the row gray mean value array and the corresponding adjacent data above and below, and calculating the average value of the obtained data to obtain a row gray mean value array after smoothing processing;

[0015] Calculating the difference between the row gray mean value array after smoothing processing of each pixel and the initial row gray mean value array, and fusing the calculated difference with the original gray value of the current pixel to obtain the processed fingerprint image.

[0016] Further, the step S1 of performing preliminary cross line removal processing on the collected fingerprint image is specifically:

[0017] For the top two rows and the bottom two rows of the fingerprint image, the missing data is supplemented by the image background gray value 255.

[0018] Further, the Gaussian filtering algorithm used in the step S2 is a one-dimensional Gaussian filtering algorithm.

[0019] Further, the step S3 of dividing all the image blocks into foreground image blocks and background image blocks based on the gray variance is specifically:

[0020] Judging whether the gray variance of the image block is greater than a first preset variance threshold, if yes, the current image block is a foreground image block; if not, the current image block is a background image block.

[0021] Further, the gray mapping coefficient in the step S4 is represented as:

[0022] α=(high-low) / max

[0023] Wherein, a represents the gray mapping coefficient; high represents the maximum value of pixel gray scale; low represents the minimum value of pixel gray scale; max represents the maximum range of image gray scale distribution.

[0024] Further, the step S5 is to perform adaptive contrast adjustment on the gray mapped fingerprint image in the column direction, specifically as follows.

[0025] Calculate the average of all pixel gray scales in each column of the fingerprint image.

[0026] Calculate the gray scale variance of all pixels in each column according to the corresponding average of gray scales.

[0027] Determine whether the gray scale variance is less than the second preset variance threshold, if yes, it means that the current column does not need to be adjusted; if not, calculate the image adjustment coefficient of the current column, and adjust the current column using the image adjustment coefficient.

[0028] Further, the step S5 is to perform adaptive contrast adjustment on the gray mapped fingerprint image in the row direction, specifically as follows.

[0029] Calculate the average of all pixel gray scales in each row of the fingerprint image.

[0030] Calculate the gray scale variance of all pixels in each row according to the corresponding average of gray scales.

[0031] Determine whether the gray scale variance is less than the second preset variance threshold, if yes, it means that the current row does not need to be adjusted; if not, calculate the image adjustment coefficient of the current column, and adjust the current row using the image adjustment coefficient.

[0032] Further, the image adjustment coefficient is represented as follows.

[0033] seed=sqrt(VAR0 / var)

[0034] Wherein, seed represents the image adjustment coefficient; VAR0 represents the preset reference variance; var represents the gray scale variance.

[0035] Correspondingly, a fingerprint image cross-line removal system is also provided, comprising:

[0036] A preprocessing module is configured to receive a collected fingerprint image, and perform preliminary cross-line removal processing on the collected fingerprint image to obtain a processed fingerprint image.

[0037] A Gaussian filter processing module is configured to perform Gaussian filter processing on the processed fingerprint image in the row direction and the column direction respectively using a Gaussian filter algorithm.

[0038] The first calculation module is used for dividing the fingerprint image after the Gaussian filtering into a plurality of image blocks according to a set size, calculating the gray mean value and the gray variance of each image block, and dividing all the image blocks into foreground image blocks and background image blocks based on the gray variance.

[0039] The second calculation module is used for calculating a pixel gray distribution histogram in the foreground image block, obtaining the maximum value and the minimum value of the pixel gray according to the pixel gray histogram, obtaining a gray mapping coefficient, and performing gray mapping on the fingerprint image according to the gray mapping coefficient.

[0040] The adjusting module is used for respectively performing adaptive contrast adjustment on the fingerprint image after the gray mapping in the row direction and the column direction, so as to obtain a final fingerprint image after the cross lines are removed.

[0041] Compared with the prior art, the fingerprint image removing cross lines method is used for removing the cross lines caused by other interference in the fingerprint image acquisition process, and is beneficial to improving the fingerprint recognition use experience. The method mainly uses the image information of the surrounding area to preliminarily eliminate the cross lines in the fingerprint image, uses the Gaussian filtering to smooth the fingerprint image after the cross lines are preliminarily eliminated, and improves the overall quality of the fingerprint image and enhances the fingerprint recognition experience through the image gray mapping and the image adaptive contrast adjustment. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 is a fingerprint image removing cross lines method flow chart provided by example one;

[0043] Figure 2 is a schematic diagram of a fingerprint image with cross lines provided by example one;

[0044] Figure 3 is a schematic diagram of a fingerprint image after the cross lines are preliminarily eliminated provided by example one;

[0045] Figure 4 is a schematic diagram of an image after the X direction Gaussian filtering provided by example one;

[0046] Figure 5 is a schematic diagram of an image after the Y direction Gaussian filtering provided by example one;

[0047] Figure 6 is a schematic diagram of an image after the image gray mapping provided by example one;

[0048] Figure 7 is a schematic diagram of an image after the Y direction adaptive contrast adjustment provided by example one;

[0049] Figure 8 is a schematic diagram of an image after the X direction adaptive contrast adjustment provided by example one. DETAILED DESCRIPTION

[0050] Following, the embodiments of the present application are illustrated by specific examples, and other advantages and effects of the present application can be easily understood by those skilled in the art from the disclosure of the present specification. The present application can also be implemented or applied by other different specific embodiments, and various modifications or changes can be made to the details in the present specification based on different views and applications without departing from the spirit of the present application. It should be noted that the following examples and features in the examples can be combined with each other without conflict.

[0051] The purpose of the present application is to provide a fingerprint image cross-line removal method and system to overcome the defects of the prior art.

[0052] Embodiment one

[0053] The present embodiment provides a fingerprint image cross-line removal method, as shown in the following formula (1), which comprises the following steps: Figure 1

[0054] S1. Receiving the collected fingerprint image, performing preliminary cross-line removal processing on the collected fingerprint image to obtain a processed fingerprint image;

[0055] S2. Using a Gaussian filter algorithm to perform Gaussian filter processing on the processed fingerprint image in the row direction and the column direction, respectively;

[0056] S3. Dividing the Gaussian filter processed fingerprint image into a plurality of image blocks according to a set size, calculating the gray mean value and gray variance of each image block, and dividing all image blocks into foreground image blocks and background image blocks based on the gray variance;

[0057] S4. Calculating the pixel gray distribution histogram in the foreground image block, obtaining the maximum value and minimum value of the pixel gray histogram, obtaining the gray mapping coefficient, and performing gray mapping on the fingerprint image according to the gray mapping coefficient;

[0058] S5. Performing adaptive contrast adjustment on the gray mapped fingerprint image in the row direction and the column direction, respectively, to obtain a final cross-line removed fingerprint image.

[0059] In step S1, the collected fingerprint image is received, and preliminary cross-line removal processing is performed on the collected fingerprint image to obtain a processed fingerprint image.

[0060] The present embodiment takes the size of the obtained fingerprint image as an example, which is W*H, and the image gray value range is 0-255. If the image gray value is less than 0 during calculation, the gray value is set to 0, and if it exceeds 255, the gray value is set to 255.

[0061] As shown in the following formula (2), it is a fingerprint image with cross-lines, and there are many cross-lines in the horizontal direction of the image. Figure 2

[0062] ​​Assuming the image size is W*H, like a pixel grid composed of W columns and H rows, the embodiment performs preliminary anti-crossing processing on the fingerprint image, specifically:

[0063] Calculate the average gray value row by row: first calculate the average gray value of each row of the image, that is, add up the gray values of all pixels in each row and then divide by the number of pixels in the row (that is, the width W of the image), so that an average value reflecting the overall light and dark degree of the row is obtained. Save all these average values to form an array of length H (image height), which is the row average gray value array AverageRow[H].

[0064] Smooth the row average gray value array: for each value in the row average gray value array AverageRow[H], take each value's adjacent upper and lower values (a total of five values), calculate the average of the five values, and thus obtain a new average row gray value array AverageRow_1[H]. The purpose of this processing is to smooth the current row's gray value by the surrounding row's gray value, making the transition between rows more natural. However, for the top two rows and the bottom two rows of the image, there are not enough adjacent rows above and below, so the image background gray value 255 is used to supplement the missing data, and then the average is taken.

[0065] Adjust the pixel value to eliminate the cross: according to the row average gray value array AverageRow[H] and the average row gray value array AverageRow_1[H], adjust each pixel in the fingerprint image. For example: for each pixel in the mth row of the image, first calculate the difference between the smoothed row average gray value (set as AverageRow_1[m]) and the original row average gray value (set as AverageRow[m]), that is, AverageRow_1[m] - AverageRow[m]. Then add the calculated difference to the original gray value of the current pixel to obtain a new pixel value. In this way, the cross information in the image is preliminarily eliminated by combining the pixel information in the surrounding area. The fingerprint image after preliminary elimination of the cross is shown in Figure 3 .

[0066] In step S2, a Gaussian filter algorithm is used to perform Gaussian filter processing in the row direction and the column direction on the processed fingerprint image.

[0067] Gaussian filter is a commonly used image smoothing technique, mainly used to reduce noise in the image and reduce image detail levels, making the image look smoother. The Gaussian filter algorithm used in this embodiment is a one-dimensional Gaussian filter operator of [1, 4, 6, 4, 1]. Using a one-dimensional Gaussian filter operator can reduce the amount of calculation and speed up the image processing.

[0068] As shown in Figure 4 , the Gaussian filtering process in the X direction is as follows:

[0069] For each pixel in each row of the image, the gray scale values of the pixel and its two neighboring pixels on the left and right (a total of five pixels) are multiplied by the weights in [1, 4, 6, 4, 1] respectively, then the results are added, and finally the sum is divided by the sum of the weights (1+4+6+4+1=16) to obtain a new pixel value.

[0070] As shown in Figure 5 , the Gaussian filtering process in the Y direction is as follows:

[0071] For each pixel in each column of the image, the gray scale values of the pixel and its two neighboring pixels above and below (a total of five pixels) are multiplied by the weights in [1, 4, 6, 4, 1] respectively, then the results are added, and finally the sum is divided by the sum of the weights (1+4+6+4+1=16) to obtain a new pixel value.

[0072] The Gaussian filtering adopted in this embodiment can smooth the image, reduce noise and reduce the level of detail, making the image look smoother. After removing the horizontal lines in the fingerprint image, this can further improve the image quality and make it more suitable for fingerprint recognition.

[0073] In step S3, the fingerprint image after Gaussian filtering is divided into a plurality of image blocks according to a set size, the gray scale mean and gray scale variance of each image block are calculated, and all image blocks are divided into foreground image blocks and background image blocks based on the gray scale variance.

[0074] First, the entire fingerprint image after Gaussian filtering is divided into image blocks with a size of 8x8 pixels,

[0075] The average value of the gray scale values (i.e. the brightness of the pixels, the larger the value, the brighter) of all pixels in each image block is calculated; the gray scale mean and gray scale variance of each image block are calculated, wherein the calculation of the gray scale mean and gray scale variance of each image block can be realized by using the existing technology, and this embodiment will not be described in detail.

[0076] It is determined whether the gray scale variance of each image block is greater than a first preset variance threshold threshold1, if yes, it indicates that the current image block is a foreground image block; if not, it indicates that the current image block is a background image block.

[0077] In step S4, a pixel gray scale distribution histogram in the foreground image block is calculated, the maximum and minimum values of the pixel gray scale are obtained according to the pixel gray scale histogram, a gray scale mapping coefficient is obtained, and the fingerprint image is subjected to gray scale mapping according to the gray scale mapping coefficient.

[0078] The distribution of gray values ​​of all pixels in the foreground image patch is statistically analyzed to form a gray-level distribution histogram. From this histogram, the maximum (brightest) and minimum (darkest) gray values ​​of each pixel are identified.

[0079] A stretching coefficient (i.e., gray-level mapping coefficient) is calculated using the maximum and minimum gray values ​​in the foreground image, as well as the maximum range of gray-level distribution in the stretched image as preset.

[0080] like Figure 6 As shown, the grayscale value of each pixel in the entire fingerprint image is adjusted according to the stretching coefficient. This makes the bright areas brighter and the dark areas darker, thereby making the image details clearer and enhancing the visual effect.

[0081] In this embodiment, the grayscale mapping coefficients are counted as follows:

[0082] α = (high - low) / max

[0083] Where α represents the gray-level mapping coefficient; high represents the maximum value of pixel gray level; low represents the minimum value of pixel gray level; and max represents the maximum range of image gray-level distribution.

[0084] In step S5, adaptive contrast adjustment is performed on the grayscale mapped fingerprint image in both row and column directions to obtain the final fingerprint image after removing horizontal lines.

[0085] The purpose of adaptive contrast adjustment is to make the contrast of an image more appropriate and to make the details in the image clearer. If the image contrast is too low, the image will appear blurry and the details will not be obvious; if the contrast is too high, the image will appear too glaring and some details will also be lost.

[0086] In this embodiment, the adjustment of the fingerprint image obtained in step S4 in the Y and X directions is specifically as follows:

[0087] The Y-axis of the fingerprint image is adjusted as follows:

[0088] Calculate the average gray value of a column: For each column of the image, calculate the average gray value (ave) of all pixels in that column.

[0089] Calculate the grayscale variance of a column: Calculate the grayscale variance var of all pixels in the column, which reflects the degree of brightness variation of the pixels in the column.

[0090] Determine if adjustment is needed: Check if the grayscale variance var is less than the second preset variance threshold threshold2. If it is, it means that the pixel brightness of the current column does not change much and no adjustment is needed. Jump directly to the next column. If not, it means that there is a lot of grayscale variation in the current column and the contrast needs to be enhanced.

[0091] Adjusting pixel value: for each pixel of the current column that needs to be adjusted, a new gray value is calculated according to the adjustment coefficient. The adjustment coefficient (seed) is a factor for controlling the adjustment intensity of the contrast, which is expressed as:

[0092] seed = sqrt(VAR0 / var)

[0093] wherein seed represents the image adjustment coefficient; VAR0 represents a preset reference variance for ensuring that the adjusted contrast is appropriate; and var represents the gray variance.

[0094] Suppose the current pixel is pixel, the base coefficient value M0, and the base coefficient M0 is the image gray value, which takes the value of the image gray range. The adjusted pixel gray value is (pixel-ave)*seed+M0. By adjusting each column of pixels in this way, the effect of adjusting the image contrast is achieved.

[0095] Adjusting the X direction of the fingerprint image, specifically:

[0096] Calculating the gray mean of the row: for each row of the image, the gray mean ave (average brightness) of all pixels in the row is calculated.

[0097] Calculating the gray variance of the row: the gray variance var of all pixels in the row is calculated, which reflects the degree of light and dark change of the column of pixels.

[0098] Judging whether adjustment is needed: judging whether the gray variance var is less than a third preset variance threshold threshold3. If yes, it means that the light and dark change of the current row of pixels is not large, and no adjustment is needed, and it directly jumps to the next row. If not, it means that there are more details in the current row, and the contrast needs to be enhanced.

[0099] Adjusting pixel value: for each pixel of the current row that needs to be adjusted, a new gray value is calculated according to the adjustment coefficient. The adjustment coefficient (seed) is a factor for controlling the adjustment intensity of the contrast, which is expressed as:

[0100] seed = sqrt(VAR0 / var)

[0101] wherein seed represents the image adjustment coefficient; VAR0 represents a preset reference variance for ensuring that the adjusted contrast is appropriate; and var represents the gray variance.

[0102] Assuming that the current pixel is pixel, the base coefficient value M0, the base coefficient M0 is the image gray value, the value is the image gray range, then the adjusted pixel gray value is (pixel-ave)*seed+M0. By adjusting each row of pixels in this way, the effect of adjusting the image contrast is achieved.

[0103] Therefore, the adjusted fingerprint is obtained according to the above manner, as shown in Figure 7 It is shown as the image after Y direction adaptive contrast adjustment, as shown in Figure 8 It is shown as the image after X direction adaptive contrast adjustment.

[0104] Compared with the prior art, the embodiment is used for removing horizontal lines caused by other interference in the process of collecting the fingerprint image, and is beneficial to improving the use experience of fingerprint identification. The method mainly uses the image information of the surrounding area to preliminarily eliminate the horizontal lines in the fingerprint image, uses Gaussian filtering to smooth the fingerprint image after the horizontal lines are preliminarily eliminated, and improves the overall quality of the fingerprint image and enhances the fingerprint identification experience through image gray mapping and image adaptive contrast adjustment.

[0105] Embodiment two

[0106] The embodiment also provides a fingerprint image horizontal line removal system, comprising:

[0107] A preprocessing module is configured to receive a collected fingerprint image, and perform preliminary horizontal line removal processing on the collected fingerprint image to obtain a processed fingerprint image.

[0108] A Gaussian filtering processing module is configured to perform Gaussian filtering processing on the processed fingerprint image in a row direction and a column direction respectively by using a Gaussian filtering algorithm.

[0109] A first calculation module is configured to divide the fingerprint image after the Gaussian filtering processing into a plurality of image blocks according to a set size, calculate a gray mean value and a gray variance of each image block, and divide all the image blocks into foreground image blocks and background image blocks based on the gray variance.

[0110] A second calculation module is configured to calculate a pixel gray distribution histogram in the foreground image blocks, obtain a maximum value and a minimum value of pixel gray according to the pixel gray histogram, obtain a gray mapping coefficient, and perform gray mapping on the fingerprint image according to the gray mapping coefficient.

[0111] An adjustment module is configured to perform adaptive contrast adjustment on the fingerprint image after the gray mapping in the row direction and the column direction respectively to obtain a final fingerprint image after the horizontal lines are removed.

[0112] It should be noted that the fingerprint image horizontal line removal system provided in the embodiment is similar to the embodiment one, and thus will not be described in detail here.

[0113] Note that the above merely describes preferred embodiments of the present application and the principles of the technology applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, modifications and substitutions can be made without departing from the scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and can include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. A method of cross-hatching removal of a fingerprint image, characterized by, The method comprises the following steps: S1. Receiving a collected fingerprint image, and performing preliminary cross-line removal processing on the collected fingerprint image to obtain a processed fingerprint image; S2. Performing Gaussian filtering processing on the processed fingerprint image in the row direction and the column direction respectively by using a Gaussian filtering algorithm; S3. Dividing the fingerprint image after the Gaussian filtering processing into a plurality of image blocks according to a set size, calculating the gray mean value and the gray variance of each image block, and dividing all the image blocks into foreground image blocks and background image blocks based on the gray variance; S4. Calculating a pixel gray distribution histogram in the foreground image blocks, obtaining the maximum value and the minimum value of the pixel gray based on the pixel gray histogram, obtaining a gray mapping coefficient, and performing gray mapping on the fingerprint image according to the gray mapping coefficient; S5. Performing adaptive contrast adjustment on the fingerprint image after the gray mapping in the row direction and the column direction respectively to obtain a final fingerprint image after the cross-line removal.

2. The method of claim 1, wherein, The step S1 of performing preliminary cross-line removal processing on the collected fingerprint image specifically comprises the following steps: calculating the gray mean value of each row of the fingerprint image to obtain an initial row gray mean value array; obtaining each data in the row gray mean value array and the corresponding adjacent data above and below, and calculating the average value of the obtained data to obtain a row gray mean value array after the smoothing processing; calculating the difference between the row gray mean value array after the smoothing processing of each pixel and the initial row gray mean value array, fusing the calculated difference with the original gray value of the current pixel to obtain the processed fingerprint image.

3. The method of claim 2, wherein, The step of obtaining each data in the row gray mean value array and the corresponding adjacent data above and below further comprises the following steps: for the top two rows and the bottom two rows of the fingerprint image, the missing data is supplemented by using the image background gray value 255.

4. The method of claim 1, wherein, The Gaussian filtering algorithm used in the step S2 is a one-dimensional Gaussian filtering algorithm.

5. The method of claim 1, wherein, The step S3 of dividing all the image blocks into foreground image blocks and background image blocks based on the gray variance specifically comprises the following steps: judging whether the gray variance of the image block is greater than a first preset variance threshold, if yes, the current image block is a foreground image block; if no, the current image block is a background image block.

6. The method of claim 1, wherein, The gray mapping coefficient in the step S4 is represented as: α=(high-low) / max wherein, α represents the gray mapping coefficient; high represents the maximum value of the pixel gray; low represents the minimum value of the pixel gray; and max represents the maximum range of the image gray distribution.

7. The method of claim 1, wherein, The step S5 of performing adaptive contrast adjustment on the fingerprint image after the gray mapping in the column direction specifically comprises the following steps: calculating the gray mean value of all the pixels in each column of the fingerprint image; calculating the gray variance of all the pixels in each column according to the corresponding gray mean value; judging whether the gray variance is less than a second preset variance threshold, if yes, the current column does not need to be adjusted; if no, calculating an image adjustment coefficient of the current column, and adjusting the current column by using the image adjustment coefficient.

8. The method of claim 7, wherein, The step S5 of performing adaptive contrast adjustment on the fingerprint image after the gray mapping in the row direction specifically comprises the following steps: calculating the gray mean value of all the pixels in each row of the fingerprint image; calculating the gray variance of all the pixels in each row according to the corresponding gray mean value; It is judged whether the gray scale variance is less than a second preset variance threshold, and if yes, it is indicated that the current row does not need to be adjusted. If no, an image adjustment coefficient of the current column is calculated, and the current row is adjusted by using the image adjustment coefficient.

9. The method of claim 8, wherein, The image adjustment coefficient is expressed as: seed=sqrt(VAR0 / var) Wherein, seed represents the image adjustment coefficient; VAR0 represents a preset reference variance; and var represents the gray scale variance.

10. A system for delaminating a fingerprint image according to any one of claims 1-9, characterized in that It comprises: A preprocessing module, configured to receive a collected fingerprint image, and perform preliminary cross-line processing on the collected fingerprint image to obtain a processed fingerprint image; A Gaussian filter processing module, configured to perform Gaussian filter processing on the processed fingerprint image in a row direction and a column direction respectively by using a Gaussian filter algorithm; A first calculation module, configured to divide the Gaussian filter processed fingerprint image into a plurality of image blocks according to a set size, calculate a gray scale mean value and a gray scale variance of each image block, and divide all the image blocks into foreground image blocks and background image blocks based on the gray scale variance; A second calculation module, configured to calculate a pixel gray scale distribution histogram in the foreground image blocks, obtain a maximum value and a minimum value of pixel gray scale according to the pixel gray scale histogram, obtain a gray scale mapping coefficient, and perform gray scale mapping on the fingerprint image according to the gray scale mapping coefficient; An adjustment module, configured to perform adaptive contrast adjustment on the gray scale mapped fingerprint image in the row direction and the column direction respectively to obtain a final fingerprint image with cross lines removed.

Images