A method for removing contamination from multi-fingerprint images

Abstract

The present application provides a method for removing contamination from a multi-finger fingerprint image, comprising the steps of: acquiring an initial multi-finger fingerprint image; extracting all fingerprint targets and their contours; selecting a preset number of contours that meet contour attributes and filtering unselected contours; obtaining a fingerprint mask image mask of the selected contours, superimposing the fingerprint mask image mask with the initial multi-finger fingerprint image, and obtaining a multi-finger fingerprint image from which contamination has been removed; the method for removing contamination from a multi-finger fingerprint image of the present application extracts all fingerprint targets and their contours from the initial multi-finger fingerprint image, selects a preset number of contours that meet contour attributes, and filters the contours of contaminated interference targets, and then superimposes the fingerprint mask image mask with the initial multi-finger fingerprint image, and obtains a multi-finger fingerprint image from which contamination has been removed, so as to improve the success rate of multi-finger fingerprint recognition.

Description

Technical Field

[0001] The present application belongs to the field of image processing technology, and more specifically, relates to a method for removing contamination from a multi-finger fingerprint image. Background Art

[0002] The application of automatic fingerprint recognition technology is becoming more and more widespread. The multi-finger fingerprint collector can collect multiple fingerprints at one time. Because it can effectively fix the finger position of the collector and avoid various fingerprint deformations and other unqualified fingerprints, it improves the accuracy and efficiency of collection.

[0003] The Chinese invention patent with publication number CN 110110697A discloses a multi-fingerprint segmentation and extraction method, system, device and medium based on direction correction. The patent first calculates the inclination angle between the lowest boundary of the minimum circumscribed rectangle and the horizontal line, then extracts the independent fingerprint area, and finally performs direction correction according to the inclination angle to obtain multiple independent fingerprint images.

[0004] From the above patents, we can see that the prior art generally directly gives the final fingerprint image during multi-finger fingerprint extraction. However, in the actual use of the multi-finger fingerprint device, because some people sweat easily, the weather is humid, and multiple people press fingerprints, it is easy to leave stains on the pressing panel of the multi-finger fingerprint device, which will interfere with each subsequent fingerprint recognition and seriously affect the success rate of multi-finger fingerprint recognition. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a method for removing contamination from a multi-finger fingerprint image, so as to solve the technical problem of contamination interference existing in the prior art in the multi-finger fingerprint recognition process.

[0006] To achieve the above purpose, the technical solution adopted in this application is: to provide a method for removing contamination from a multi-finger fingerprint image, comprising the following steps:

[0007] Collecting initial multi-fingerprint images;

[0008] Extract all fingerprint targets and their contours;

[0009] Select a preset number of contours that match the contour attributes and filter out unselected contours;

[0010] The fingerprint mask image mask of the selected contour is obtained, and the fingerprint mask image mask is superimposed on the initial multi-finger fingerprint image to obtain the multi-finger fingerprint image with contamination removed.

[0011] Preferably, after collecting the initial multi-fingerprint image, image preprocessing is also included, and the image preprocessing method includes the following steps:

[0012] For the current pixel (x, y) to be processed, select a template consisting of m pixels of its nearest neighbors;

[0013] Find the mean of all pixels in the template, and then assign the mean to the current pixel (x, y) as the grayscale src(x, y) of the processed image at that point. The formula of src(x, y) is:

[0014]

[0015] Preferably, the method for extracting all fingerprint targets and their contours comprises the following steps:

[0016] Select the initial threshold T1;

[0017] The multi-fingerprint image after image preprocessing is selected by threshold T1 to obtain a binary image that can reflect the overall and local features of the image;

[0018] The binary image is corroded and expanded in sequence to obtain the fingerprint target;

[0019] The binary image after erosion and dilation is subjected to edge detection to extract the contour of the fingerprint target.

[0020] Preferably, the method for selecting a preset number of contours that meet contour attributes comprises the following steps:

[0021] Calculate the perimeter or area of ​​all contours separately;

[0022] Sort all contours by perimeter or area size;

[0023] Select a preset number of contours with the largest perimeter or area.

[0024] Preferably, the method for selecting a preset number of contours that meet contour attributes further comprises the following steps:

[0025] Get the centroid coordinates of each selected contour;

[0026] Determine whether the contour corresponding to the centroid with the largest or smallest horizontal coordinate has the smallest perimeter or area among all selected contours;

[0027] If yes, filter out the unselected contours, if no, readjust the threshold T1 in the method of extracting all fingerprint objects and their contours.

[0028] Preferably, the method of selecting a preset number of contours that meet the contour attributes is applicable to the collection of four-finger fingerprints.

[0029] The following steps are also included:

[0030] Get the centroid coordinates of each selected contour;

[0031] Fit all centroid curves in turn according to the size of the horizontal coordinate of each centroid, output the slope K(x) of the curve, and judge whether |K(x)|≤T2 is satisfied, where T2 is the set threshold;

[0032] If yes, filter out the unselected contours, if no, readjust the threshold T1 in the method of extracting all fingerprint objects and their contours.

[0033] Preferably, T2∈[0,5].

[0034] Preferably, after obtaining the multi-fingerprint image with contamination removed, the method further comprises the following steps:

[0035] The centroid coordinates corresponding to the unselected contours are stored in a contamination memory library. The contamination memory library is used to filter the interfering contours in advance according to the records in the contamination memory library when selecting a preset number of contours that meet the contour attributes again.

[0036] Preferably, the method for obtaining a fingerprint mask image of a selected contour comprises the following steps:

[0037] Fill the selected outline;

[0038] Set the grayscale value of the filled area to 255 and the grayscale value of the non-filled area to 0;

[0039] Generate fingerprint mask image mask.

[0040] Preferably, the method for superimposing the fingerprint mask image mask with the initial multi-finger fingerprint image comprises the following steps:

[0041] Create a new dst image with the same size as the initial multi-fingerprint image;

[0042] The dst image is assigned a value of 255;

[0043] Starting from the fingerprint mask image mask, traverse the entire image;

[0044] If the value of the current pixel is 255, the corresponding pixel on the dst image remains unchanged. If the value of the current pixel is 0, the corresponding pixel on the dst image is assigned the grayscale value of the corresponding position of the initial multi-fingerprint image.

[0045] The beneficial effect of the multi-fingerprint image contamination removal method provided in the present application is that, compared with the prior art, by extracting all fingerprint targets and their contours of the initial multi-finger fingerprint image, selecting a preset number of contours that meet the contour attributes to filter out the contours of the contaminated interference targets, and then superimposing the fingerprint mask image mask with the initial multi-finger fingerprint image to obtain a multi-finger fingerprint image with contamination removed, so as to improve the success rate of multi-finger fingerprint recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0047] Figure 1 A schematic diagram of a flow chart of a method for removing contamination from a multi-fingerprint image provided in an embodiment of the present application;

[0048] Figure 2 The initial multi-fingerprint image with contamination in the method provided in the embodiment of the present application;

[0049] Figure 3 The binarized image in the method provided in the embodiment of the present application;

[0050] Figure 4 The image after the fingerprint target and contour are extracted in the method provided in the embodiment of the present application;

[0051] Figure 5 A curve fitting diagram when the slope K(x) of the output curve in the method provided in the embodiment of the present application satisfies |K(x)|≤T2;

[0052] Figure 6 A curve fitting diagram when the slope K(x) of the output curve in the method provided in the embodiment of the present application does not satisfy K(x)≤T2;

[0053] Figure 7 The fingerprint mask image mask in the method provided in the embodiment of the present application;

[0054] Figure 8 A multi-fingerprint image for removing contamination in the method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element.

[0057] It should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0058] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0059] Please also read Figure 1 The method for removing contamination from a multi-fingerprint image provided by an embodiment of the present application is now described. The method for removing contamination from a multi-finger fingerprint image comprises the following steps:

[0060] Step S01, collecting an initial multi-finger fingerprint image;

[0061] Step S02, extracting all fingerprint targets and their contours;

[0062] Step S03, selecting a preset number of contours that meet the contour attributes, and filtering unselected contours;

[0063] Step S04, obtaining a fingerprint mask image mask of the selected contour, and superimposing the fingerprint mask image mask with the initial multi-finger fingerprint image to obtain a multi-finger fingerprint image with contamination removed.

[0064] It is understandable that in step S01, the initial multi-finger fingerprint image should be understood to contain at least two fingerprint targets, preferably four fingers. For example, using the MV1088 fingerprint scanner produced by Shengshi Technology, four fingers can be naturally placed in the center of the fingerprint collection panel of the MV1088 fingerprint scanner to collect a clear initial multi-finger fingerprint image.

[0065] In step S02, when all fingerprint objects and their contours are extracted, the contamination is also extracted as a fingerprint object and the contour is extracted together.

[0066] In step S03, a preset number of contours that meet the contour attributes are selected, wherein the preset number is set according to the actual application scenario, for example, when the multi-finger fingerprint image contamination removal method is applied to a four-finger fingerprint collector, the preset number is set to 4, and when the multi-finger fingerprint image contamination removal method is applied to a five-finger fingerprint collector, the preset number is set to 5. Based on the difference between the real fingerprint target and the interference target in the image, the fingerprint target is screened by contour attributes, and the unselected fingerprint targets are filtered, that is, the real multi-finger fingerprint is retained, and the interference targets caused by contamination are filtered, wherein the contour attributes include but are not limited to contour area, contour perimeter, and contour shape.

[0067] In step S04, the fingerprint mask image mask, that is, the mask of multiple real fingerprints, is superimposed on the initial multi-finger fingerprint image to obtain a decontaminated multi-finger fingerprint image, thereby achieving noise elimination of the contaminated fingerprint image.

[0068] The algorithm of the present invention can solve the problem of leaving stains on the pressing panel easily after sweating, humid weather, and multiple people pressing fingerprints during the actual use of the fingerprint device by optimizing the algorithm. The calculation intensity is low, so the hardware computing power requirement is low, and it can achieve real-time performance. Finally, a neat and beautiful multi-finger fingerprint image is output, which conforms to human aesthetics and increases user experience.

[0069] The beneficial effect of the multi-fingerprint image contamination removal method provided in the present application is that, compared with the prior art, by extracting all fingerprint targets and their contours of the initial multi-finger fingerprint image, selecting a preset number of contours that meet the contour attributes to filter out the contours of the contaminated interference targets, and then superimposing the fingerprint mask image mask with the initial multi-finger fingerprint image to obtain a multi-finger fingerprint image with contamination removed, so as to improve the success rate of multi-finger fingerprint recognition.

[0070] In another embodiment of the present application, please refer to Figure 2 After collecting the initial multi-fingerprint image, the method further includes image preprocessing, wherein the image preprocessing method includes the following steps:

[0071] For the current pixel (x, y) to be processed, select a template consisting of m pixels of its nearest neighbors;

[0072] Find the mean of all pixels in the template, and then assign the mean to the current pixel (x, y) as the grayscale src(x, y) of the processed image at that point. The formula of src(x, y) is:

[0073]

[0074] It can be understood that by finding the mean of all pixels in the template and then assigning the mean to the current pixel point (x, y) as the grayscale src(x, y) of the processed image at this point, a mean filtering effect can be achieved. It is worth adding that m is the total number of pixels in the template including the current pixel, and the window size of the template is adjusted according to the distance between the camera and the human body. The smaller the distance between the camera and the human body, the smaller the window of the template, and the larger the distance between the camera and the human body, the larger the window of the template.

[0075] Please also refer to Figure 3 and Figure 4 The method for extracting all fingerprint targets and their contours comprises the following steps:

[0076] Select the initial threshold T1;

[0077] The multi-fingerprint image after image preprocessing is selected by threshold T1 to obtain a binary image that can reflect the overall and local features of the image;

[0078] The binary image is corroded and expanded in sequence to obtain the fingerprint target;

[0079] The binary image after erosion and dilation is subjected to edge detection to extract the contour of the fingerprint target.

[0080] It can be understood that, after selecting the initial threshold T1, threshold can be set as a fixed threshold, and all pixels of the image can be traversed; let bin(x,y) be the pixel after binarization, if src(x,y)>threshold, then bin(x,y)=0, otherwise bin(x,y)=255, set the image bin to the same specifications as src (including size and depth), and select the grayscale image of 256 brightness levels through the appropriate threshold T1 to obtain a binary image that can still reflect the overall and local features of the image. The binarization of the image is conducive to further processing of the image, making the image simpler, and reducing the amount of data, which can highlight the contour of the target of interest. The initial multi-finger fingerprint image has been processed into a black and white image, the black is the background, and the white fingerprint area includes the index finger, middle finger, ring finger, and little finger. The other parts need to be whitened, and other interference can be eliminated by corrosion and expansion. The Canny operator is used to perform edge detection on the binary closed operation image and extract all contours.

[0081] Furthermore, the method for selecting a preset number of contours that meet contour attributes comprises the following steps:

[0082] Calculate the perimeter or area of ​​all contours separately;

[0083] Sort all contours by perimeter or area size;

[0084] Select a preset number of contours with the largest perimeter or area.

[0085] It is understandable that, based on the size difference between the real fingerprint target and the interference target in the image, the contour of the interference target is usually smaller than the contour of the real fingerprint target. Therefore, by selecting a preset number of contours with the largest perimeter or area, the contour of the interference target can be filtered.

[0086] Furthermore, the method for selecting a preset number of contours that meet contour attributes further includes the following steps:

[0087] Get the centroid coordinates of each selected contour;

[0088] Determine whether the contour corresponding to the centroid with the largest or smallest horizontal coordinate has the smallest perimeter or area among all selected contours;

[0089] If yes, filter out the unselected contours, if no, readjust the threshold T1 in the method of extracting all fingerprint objects and their contours.

[0090] It is understandable that, since the contour of the fingerprint target is an irregular geometric shape, the coordinate position of each contour can be accurately represented by calculating the centroid, which is the effect of simplifying the complex. Since the volume of the little finger is usually smaller than that of the thumb, index finger, middle finger and ring finger, the contour area or perimeter of the little finger is usually smaller than that of the thumb, index finger, middle finger and ring finger. When the acquisition object is the right hand, the little finger is located on the right side of the image, that is, the horizontal coordinate of the centroid of the little finger is the largest. When the acquisition object is the left hand, the little finger is located on the left side of the image, that is, the horizontal coordinate of the centroid of the little finger is the smallest. Therefore, by using the obvious distinguishing features of the little finger, we can judge whether the contour corresponding to the center of mass with the largest or smallest horizontal coordinate has the smallest perimeter or area among all selected contours. If so, it means that the selected contour is a true multi-finger fingerprint, and the unselected contours are filtered. If not, it means that the contour selection is incorrect and the contour of the contaminated target is mistakenly selected. In this case, the threshold T1 in the method of extracting all fingerprint targets and their contours is readjusted. By adjusting the degree of prominence of the contaminated target in the binary image, the binary image is corroded and expanded in turn, and the contour of the contaminated target can be filtered or reduced to further achieve the accuracy of contamination removal of the multi-finger fingerprint image.

[0091] It is worth noting that the coordinates in this application should be understood to conform to the common practice in visual image processing technology, that is, the upper left corner of the image is the coordinate origin, x represents the horizontal axis coordinate, and y represents the vertical axis coordinate. In addition, if the image is mirrored, the above "left side" and "right side" can be interchanged.

[0092] In another embodiment, see Figure 5 and Figure 6, which is applicable to four-finger fingerprint collection, that is, collecting the index finger, middle finger, ring finger and little finger. The above "determine whether the contour corresponding to the centroid with the largest or smallest horizontal coordinate has the smallest perimeter or area among all selected contours" can also be replaced by "fit all centroid curves in turn according to the horizontal coordinate size of each centroid, output the slope K(x) of the curve, and determine whether |K(x)|≤T2 is satisfied, where T2 is the set threshold value".

[0093] It is understandable that, since the contour circumference or area corresponding to the pinky fingerprint is smaller, it often happens that the contour circumference or area of ​​the contaminated target is larger than the contour circumference or area corresponding to the pinky fingerprint. Similar problems can be optimized through curve fitting. Based on the fact that the length ratios of the index finger, middle finger, ring finger and pinky of ordinary people are within a certain range, the slope |K(x)| of the curve fitted by the centroid corresponding to the normally collected four-finger fingerprint is within a certain range, and when collecting multi-finger fingerprints, in order to ensure the integrity of fingerprint collection, the fingers are usually required to be spread out, thus further narrowing the range of the slope |K(x)|. Based on the characteristics of the four-finger fingerprints, by judging whether |K(x)|≤T2 is satisfied, it can be judged whether the contour selection is incorrect. Among them, all the centroid curves are fitted in sequence according to the size of the horizontal coordinate of each centroid, in order to ensure that the curve fitting is performed in the order of the corresponding centroids of the index finger, middle finger, ring finger and little finger or the little finger, ring finger, middle finger and index finger, that is, when the four-finger fingerprint collected is for the left hand, the curve fitting is performed in the order of the corresponding centroids of the little finger, ring finger, middle finger and index finger, and when the four-finger fingerprint collected is for the right hand, the curve fitting is performed in the order of the corresponding centroids of the index finger, middle finger, ring finger and little finger. If |K(x)|≤T2 is not satisfied, it means that the contour selection is incorrect and the contour of the contaminated target is mistakenly selected, then the threshold T1 in the method of extracting all fingerprint targets and their contours is readjusted, and by adjusting the degree of contamination of the contaminated target in the binary image, the contour of the contaminated target can be filtered or reduced after the binary image is corroded and expanded in sequence, so as to further achieve the accuracy of contamination removal of multi-finger fingerprint images. Preferably, T2∈[0,5].

[0094] Furthermore, after obtaining the multi-finger fingerprint image with contamination removed, the method further comprises the following steps:

[0095] The centroid coordinates corresponding to the unselected contours are stored in a contamination memory library. The contamination memory library is used to filter the interfering contours in advance according to the records in the contamination memory library when selecting a preset number of contours that meet the contour attributes again.

[0096] It is understandable that when there is contamination on the panel of the multi-finger fingerprint instrument, it will cause interference every time the fingerprint is collected, and the position of the multi-finger fingerprint after imaging remains unchanged. Therefore, based on this feature, this application stores the centroid coordinates corresponding to the unselected contours in the contamination memory library. The contamination memory library is used to filter the interference contours in advance according to the records in the contamination memory library when selecting a preset number of contours that meet the contour attributes again. For example, calculate the distance D between the centroid of the contour in the current image and the centroid in the contamination memory library. The smaller D is, the higher the probability that the contour is an interference target, and vice versa.

[0097] In another embodiment of the present application, the method for obtaining a fingerprint mask image of a selected contour comprises the following steps:

[0098] Fill the selected outline;

[0099] Set the grayscale value of the filled area to 255 and the grayscale value of the non-filled area to 0;

[0100] Generate fingerprint mask image mask.

[0101] It can be understood that by filling the contour image, the fingerprint is perfectly enclosed, thereby obtaining the fingerprint mask image mask.

[0102] Furthermore, the method of superimposing the fingerprint mask image mask with the initial multi-finger fingerprint image comprises the following steps:

[0103] Create a new dst image with the same size as the initial multi-fingerprint image;

[0104] The dst image is assigned a value of 255;

[0105] Starting from the fingerprint mask image mask, traverse the entire image;

[0106] If the value of the current pixel is 255, the corresponding pixel on the dst image remains unchanged. If the value of the current pixel is 0, the corresponding pixel on the dst image is assigned the grayscale value of the corresponding position of the initial multi-fingerprint image.

[0107] It can be understood that, starting from the fingerprint mask image mask, let y = 0, and traverse horizontally from the top of the mask image along the x direction. If a white point is encountered, the point at the corresponding position of the dst image is assigned the grayscale value of the corresponding position of the initial multi-finger fingerprint image. If a black point is encountered, the grayscale value of dst remains unchanged. Then let y = y + 1, and continue to traverse horizontally from top to bottom. Until the entire dst image is traversed, the obtained dst image is the multi-finger fingerprint image with the contamination removed.

[0108] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for removing contamination from a multi-finger fingerprint image, characterized in that: The following steps are involved: Collecting initial multi-fingerprint images; Extract all fingerprint targets and their contours; Select a preset number of contours that match the contour attributes and filter out unselected contours; The method for selecting a preset number of contours that meet contour attributes comprises the following steps: Calculate the perimeter or area of ​​all contours separately; Sort all contours by perimeter or area size; Select a preset number of contours with the largest perimeter or area; The method for selecting a preset number of contours that meet the contour attributes is applicable to four-finger fingerprint collection, and further includes the following steps: Get the centroid coordinates of each selected contour; Fit all centroid curves in turn according to the size of the horizontal coordinate of each centroid, output the slope K(x) of the curve, and judge whether |K(x)|≤T2 is satisfied, where T2 is the set threshold; If yes, filter out the unselected contours; if no, readjust the threshold T1 in the method for extracting all fingerprint targets and their contours; obtain the fingerprint mask image mask of the selected contour, superimpose the fingerprint mask image mask with the initial multi-finger fingerprint image, and obtain the multi-finger fingerprint image with the contamination removed.

2. The method for removing contamination from a multi-fingerprint image as claimed in claim 1, characterized in that: After collecting the initial multi-fingerprint image, the method further includes image preprocessing, wherein the image preprocessing method includes the following steps: For the current pixel (x, y) to be processed, select a template consisting of m pixels of its nearest neighbors; Find the mean of all pixels in the template, and then assign the mean to the current pixel (x, y) as the grayscale src(x, y) of the processed image at the corresponding position.

3. The method for removing contamination from a multi-fingerprint image as claimed in claim 2, characterized in that: The method for extracting all fingerprint targets and their contours comprises the following steps: Select the initial threshold T1; The multi-fingerprint image after image preprocessing is selected through the threshold T1 to obtain a binary image reflecting the overall and local features of the image; The binary image is corroded and expanded in sequence to obtain the fingerprint target; The binary image after erosion and dilation is subjected to edge detection to extract the contour of the fingerprint target.

4. The method for removing contamination from a multi-fingerprint image as claimed in claim 1, characterized in that: The method for selecting a preset number of contours that meet contour attributes also includes the following steps: Get the centroid coordinates of each selected contour; Determine whether the contour corresponding to the centroid with the largest or smallest horizontal coordinate has the smallest perimeter or area among all selected contours; If yes, filter out the unselected contours, if no, readjust the threshold T1 in the method of extracting all fingerprint objects and their contours.

5. The method for removing contamination from a multi-fingerprint image as claimed in claim 1, characterized in that: T2∈[0，5]。 6. The method for removing contamination from a multi-fingerprint image according to any one of claims 1 to 5, characterized in that: After obtaining the decontaminated multi-fingerprint image, the following steps are also included: The centroid coordinates corresponding to the unselected contours are stored in a contamination memory library. The contamination memory library is used to filter the interfering contours in advance according to the records in the contamination memory library when selecting a preset number of contours that meet the contour attributes again.

7. The method for removing contamination from a multi-fingerprint image according to any one of claims 1 to 5, characterized in that: The method for obtaining a fingerprint mask image of a selected contour comprises the following steps: Fill the selected outline; Set the grayscale value of the filled area to 255 and the grayscale value of the non-filled area to 0; Generate fingerprint mask image mask.

8. The method for removing contamination from a multi-fingerprint image as claimed in claim 7, characterized in that: The method for superimposing the fingerprint mask image mask with the initial multi-finger fingerprint image comprises the following steps: Create a new dst image with the same size as the initial multi-fingerprint image; The dst image is assigned a value of 255; Starting from the fingerprint mask image mask, traverse the entire image; If the value of the current pixel is 255, the corresponding pixel on the dst image remains unchanged. If the value of the current pixel is 0, the corresponding pixel on the dst image is assigned the grayscale value of the corresponding position of the initial multi-fingerprint image.

Images