Fingerprint recognition method, fingerprint recognition sensor and electronic equipment

Abstract

The present invention provides a fingerprint recognition method, which includes: calculating the actual rotation angle and actual translation amount between a sample fingerprint image and a fingerprint image to be verified; rotating one of the sample fingerprint image and the fingerprint image to be verified by the actual rotation angle and translating the actual translation amount so that the sample fingerprint image and the fingerprint image to be verified form a first fingerprint image to be paired and a second fingerprint image to be paired respectively; calculating the overlapping area of ​​the first fingerprint image to be paired and the second fingerprint image to be paired; matching the first overlapping fingerprint image and the second overlapping fingerprint image corresponding to the overlapping area to obtain a verification result. The above fingerprint recognition method can accurately identify a narrow fingerprint image by matching and identifying the image of the overlapping area of ​​the sample fingerprint image and the fingerprint image to be verified. In addition, the present invention also provides a fingerprint recognition sensor and an electronic device.

Description

Technical Field

[0001] The present invention relates to the field of fingerprint recognition technology, and in particular to a fingerprint recognition method, a fingerprint recognition sensor and an electronic device. Background Art

[0002] In recent years, the technology of using fingerprints for identity authentication has been widely used in electronic devices such as mobile phones and door locks. Taking smartphones as an example, with the popularity of full-screen technology, the space left for embedded sensors of fingerprint chips on mobile phones is getting smaller and smaller. For this reason, many manufacturers have adopted side fingerprint technology, that is, placing the fingerprint module on the side of the phone. Side fingerprint technology is convenient and easy to use, and it can also take into account the user's full-screen needs, but at the same time, the size of the side fingerprint sensing area is very limited, and only a small amount of fingerprint information can be obtained at a time.

[0003] Therefore, how to accurately identify fingerprints under low information entropy has become an urgent problem to be solved in side fingerprint recognition technology or narrow fingerprint recognition. Summary of the invention

[0004] In view of this, it is necessary to provide a fingerprint recognition method applicable to small fingerprint images, and a fingerprint recognition sensor and electronic device using the fingerprint recognition method.

[0005] In one aspect, an embodiment of the present invention provides a fingerprint recognition method, the fingerprint recognition method comprising:

[0006] Calculate the actual rotation angle and actual translation amount between the sample fingerprint image and the fingerprint image to be verified;

[0007] Rotate one of the sample fingerprint image and the fingerprint image to be verified by the actual rotation angle and translate by the actual translation amount so that the sample fingerprint image and the fingerprint image to be verified form a first fingerprint image to be paired and a second fingerprint image to be paired respectively;

[0008] Calculating an overlapping area between the first fingerprint image to be paired and the second fingerprint image to be paired;

[0009] A matching operation is performed on the first overlapping fingerprint image and the second overlapping fingerprint image corresponding to the overlapping area to obtain a verification result, wherein: the first overlapping fingerprint image is an image corresponding to the first fingerprint image to be paired and the overlapping area, and the second overlapping fingerprint image is an image corresponding to the second fingerprint image to be paired and the overlapping area.

[0010] On the other hand, an embodiment of the present invention further provides a fingerprint recognition sensor, the fingerprint recognition sensor comprising a storage medium and a processor. The storage medium is used to store fingerprint recognition program instructions. The processor is used to execute the fingerprint recognition program instructions to implement the above fingerprint recognition method.

[0011] On the other hand, an embodiment of the present invention further provides an electronic device, which includes a main body and the above-mentioned fingerprint recognition sensor arranged on the main body.

[0012] The fingerprint recognition method, fingerprint recognition sensor and electronic device can accurately recognize a small fingerprint image by operating the sample fingerprint image to overlap with the fingerprint image to be verified, and matching and identifying the image of the overlapping area between the sample fingerprint image and the fingerprint image to be verified. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, 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 invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0014] Figure 1 A schematic diagram of a fingerprint recognition method flow chart provided by an embodiment.

[0015] Figure 2 A schematic diagram of an electronic device provided by an embodiment, wherein the electronic device has an application Figure 1 A fingerprint recognition sensor for the fingerprint recognition method shown.

[0016] Figure 3 A schematic diagram of changes in a fingerprint image during processing provided by an embodiment.

[0017] Figure 4 for Figure 1 A schematic diagram of a sub-process of the fingerprint recognition method is shown.

[0018] Figure 5 A partial flow chart of a fingerprint recognition method provided for another embodiment.

[0019] Figure 6 A schematic diagram of another change of a fingerprint image during processing provided by an embodiment.

[0020] Figure 7 for Figure 1 Another sub-process diagram of the fingerprint recognition method is shown.

[0021] Figure 8 for Figure 7 A schematic diagram of a sub-process of the fingerprint recognition method is shown.

[0022] Fig. 9 A partial flow chart of another fingerprint recognition method.

[0023] Fig.10 A schematic diagram of a program module of a fingerprint recognition sensor provided by an embodiment.

[0024] Fig.11 A schematic diagram of a program module of a fingerprint recognition sensor provided in another embodiment

[0025] Fig.12 A schematic diagram of another change of a fingerprint image during processing provided by an embodiment

[0026] Fig.13 A schematic diagram of a program module of a fingerprint recognition sensor provided by another embodiment.

[0027] Fig.14 for Figure 7 Another sub-process diagram of the fingerprint recognition method is shown.

[0028] Fig.15 for Figure 7 Another sub-process diagram of the fingerprint recognition method is shown.

[0029] Fig.16 A schematic diagram of the internal structure of a fingerprint recognition sensor provided by an embodiment.

[0030] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. The drawings in the specification illustrate examples of embodiments of the present invention. It is understood that the proportions shown in the drawings in the specification are not the proportions of the actual implementation of the present invention. They are only for the purpose of illustrative explanation and are not drawn according to the original size. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0032] The present invention provides a fingerprint recognition method that can recognize a narrow fingerprint image (such as a narrow strip or a narrow arc strip). The fingerprint recognition method can be used to set a narrow fingerprint recognition sensor on an electronic device, for example, a fingerprint recognition sensor on the side of a mobile phone, a tablet computer, etc.

[0033] Please see Figure 1 and Figure 2, which is a flow chart of a fingerprint recognition method provided by an embodiment. The fingerprint recognition method is applied to a fingerprint recognition sensor 1 provided in an electronic device 100. The fingerprint recognition sensor 1 includes a sensing area 10, and the fingerprint recognition sensor 1 is used to sense a fingerprint image when a user's finger is placed in the sensing area 10. The fingerprint recognition method includes the following steps.

[0034] Step S200, calculating the actual rotation angle and actual translation amount between the sample fingerprint image and the fingerprint image to be verified. The actual rotation angle and actual translation amount respectively represent the rotation angle and translation distance between the sample fingerprint image and the fingerprint image to be verified. The actual rotation angle and actual translation amount between the sample fingerprint image and the fingerprint image to be verified can be calculated by using existing deep learning algorithms, homography matrix algorithms, etc. to achieve the rotation angle and translation distance between the two images. In this embodiment, the calculation of the actual rotation angle and actual translation amount between the sample fingerprint image and the fingerprint image to be verified is achieved by using a phase correlation algorithm, and the specific content will be described below.

[0035] In this embodiment, the sample fingerprint image and the fingerprint image to be verified are fingerprint images obtained after the original sample fingerprint image and the original fingerprint image to be verified are enhanced. Among them, the fingerprint recognition sensor 1 includes a setting mode and a verification mode. The original sample fingerprint image is input by the user when setting the fingerprint password, that is, the fingerprint recognition sensor 1 is in the setting mode, and the user places the finger in the sensing area 10 for sensing. The original fingerprint image to be verified is input when the user needs to decrypt the fingerprint password, that is, the fingerprint recognition sensor 1 is in the verification mode, and the user places the finger in the sensing area 10 for sensing. The fingerprint recognition sensor 1 can be an optical fingerprint sensor, a capacitive fingerprint sensor, etc., which is not limited here.

[0036] Taking the original sample fingerprint image M and the original fingerprint image to be verified R as examples, since there are more noises in the original sample fingerprint image M and the original fingerprint image to be verified R, part of the noise may be caused by the fingerprint recognition sensor 1 itself; part of the noise may be caused by the external environment, for example, the fingerprint may be affected by stains, water stains, weather temperature and humidity, etc. Further, in order to make fingerprint recognition more accurate, it is necessary to enhance the original sample fingerprint image M and the original fingerprint image to be verified R. That is, before step S200, the fingerprint recognition method further includes: step S100 enhances the original sample fingerprint image M and the original fingerprint image to be verified R to obtain the sample fingerprint image M0 and the fingerprint image to be verified R0. Among them, the enhancement processing of the original sample fingerprint image M and the original fingerprint image to be verified R can be processed by the existing image enhancement method. In this embodiment, the enhancement processing of the original sample fingerprint image M and the original fingerprint image to be verified R is completed by Gaussian difference processing, windowing processing, Fourier transform processing, and frequency domain filtering in sequence, and the specific content will be described below.

[0037] It can be understood that in some feasible embodiments, if the noise of the fingerprint recognition sensor 1 is small and has little impact on the external environment, the sample fingerprint image M0 and the fingerprint image to be verified R0 can also use the original sample fingerprint image M and the original fingerprint image to be verified R.

[0038] Step S400: rotate one of the sample fingerprint image and the fingerprint image to be verified by an actual rotation angle and translate the actual translation amount so that the sample fingerprint image and the fingerprint image to be verified form a first fingerprint image to be paired and a second fingerprint image to be paired. Figure 3 As shown, the actual rotation angle is θ and the actual translation amount is D. If the sample fingerprint image M0 is rotated and translated, the first fingerprint image to be matched is the sample fingerprint image MT after being rotated by θ and translated by D; the second fingerprint image to be matched is the fingerprint image to be verified R. It can be understood that since the actual rotation angle θ and the actual translation amount D are the rotation angle and translation amount between the sample fingerprint image M0 and the fingerprint image to be verified R0, the sample fingerprint image M0 is rotated by an angle θ and translated by a distance D, so that the processed sample fingerprint image MT and the fingerprint image to be verified R0 can overlap.

[0039] It can be understood that in some feasible embodiments, if the fingerprint image to be verified R0 is rotated and translated, the first fingerprint image to be paired is the sample fingerprint image M0; the second fingerprint image to be paired is the fingerprint image after the fingerprint image to be verified R is rotated θ and translated D.

[0040] Step S500, calculating the overlapping area of ​​the first fingerprint image to be paired and the second fingerprint image to be paired. For example, the first fingerprint image to be paired is the fingerprint image MT obtained by rotating the sample fingerprint image M0 by θ and translating by D. Since the first fingerprint image to be paired is obtained by rotating and translating the sample fingerprint image M0 according to the rotation angle θ and the translation amount D between the sample fingerprint image M0 and the fingerprint image to be verified R0, the first fingerprint image to be paired and the second fingerprint image to be paired, that is, the fingerprint image to be verified R0, at least partially overlap, and the overlapping area is O (such as Figure 3 As shown). It can be understood that, in some feasible embodiments, the overlapping area O can also be the overlapping area between the fingerprint image to be verified and the sample fingerprint image M0 after rotation and translation.

[0041] Step S600, the first overlapping fingerprint image corresponding to the overlapping area is matched with the second overlapping fingerprint image to obtain a verification result. The first overlapping fingerprint image is an image corresponding to the first fingerprint image to be matched and the overlapping area, and the second overlapping fingerprint image is an image corresponding to the second fingerprint image to be matched and the overlapping area. The matching operation of the first overlapping fingerprint image and the second overlapping fingerprint image can be calculated using an existing image matching algorithm.

[0042] In this embodiment, the matching operation of the first overlapping fingerprint image and the second overlapping fingerprint image is implemented by using a scoring function. That is, step S600 is to score the first overlapping fingerprint image and the second overlapping fingerprint image using a scoring function to obtain a verification result. The scoring function is an NCC function. The scoring function is: ; Wherein, g(x,y) is used to represent the pixel points of the first overlapping fingerprint image; h(x,y) is used to represent the pixel points of the second overlapping fingerprint image; Used to represent the sum of products between the one-to-one corresponding pixel points in the first overlapping fingerprint image and the second overlapping image; Used to represent the square root of the product of the sum of the squares of the pixels in the first overlapping fingerprint image and the sum of the squares of the pixels in the second overlapping image.

[0043] Please refer to Figure 4 In some feasible embodiments, before using the scoring function to score, the first overlapping fingerprint image and the second overlapping fingerprint image are weighted according to the detail features. That is, in some feasible embodiments, step S600 includes the following steps.

[0044] Step S601, identifying the fingerprint details of the first overlapping fingerprint image and the second overlapping fingerprint image. The fingerprint details include details such as ridge end points and bifurcation points. The fingerprint detail feature recognition method can be recognized by using the existing fingerprint feature point recognition method, which will not be described in detail here.

[0045] Step S603, weighting the pixels in the first overlapping fingerprint image and the second overlapping fingerprint image according to the identified detail features to obtain the pixels of the first overlapping fingerprint image and the pixels of the second overlapping fingerprint image, wherein the weight of the pixels with detail features is greater than the weight of the pixels without detail features. For example, if a pixel G1 of the first overlapping fingerprint image G has a detail feature of the fingerprint, a weight of 0.8 is added to the pixel G1 of the first overlapping fingerprint image. When another pixel G2 of the first overlapping fingerprint image G does not have a detail feature of the fingerprint, a weight of 0.2 is added to the other pixel G2 of the first overlapping fingerprint image. The two fingerprint images are generally distinguished by the difference in detail features. Therefore, the weight of the pixel with detail features is increased according to the detail features to further improve the recognition accuracy.

[0046] Step S605, using a scoring function to score the pixels of the weighted first overlapping fingerprint image and the weighted second overlapping fingerprint image to obtain a verification result. The scoring function is an NCC function. The scoring function is ;

[0047] Wherein, g(x,y) is used to represent the pixel point of the weighted first overlapping fingerprint image; h(x,y) is used to represent the pixel point of the weighted second overlapping fingerprint image; Used to represent the sum of products of corresponding pixels in the weighted first overlapping fingerprint image and the weighted second overlapping image; It is used to represent the square root of the product of the weighted sum of the squares of the pixels in the first overlapping fingerprint image and the weighted sum of the squares of the pixels in the second overlapping image.

[0048] The fingerprint recognition method operates the sample fingerprint image to overlap with the fingerprint image to be verified, and matches and recognizes the overlapped portion of the sample fingerprint image after the operation with the image of the fingerprint image to be verified, so that even a small fingerprint image can be accurately recognized. In addition, since the fingerprint recognition method uses NCC scoring and weighting to score and recognize during matching and recognition, the accuracy of fingerprint recognition is further improved.

[0049] Please see Figure 5 , which is a schematic diagram of the sub-step flow chart of step S100 in an embodiment. Step S100 includes the following steps.

[0050] Step S101, Gaussian difference processing is performed on the original sample fingerprint image and the original fingerprint image to be verified to obtain a first processed sample fingerprint image and a first processed fingerprint image to be verified. Specifically, the first processed sample fingerprint image and the first processed fingerprint image to be verified obtained after the Gaussian difference processing of the original sample fingerprint image and the original fingerprint image to be verified can eliminate the background noise interference in the original sample fingerprint image and the original fingerprint image to be verified. Figure 6 As shown, it shows the effect of a fingerprint image S1 obtained after a fingerprint image is processed by Gaussian difference.

[0051] Step S103, windowing the first processed sample fingerprint image and the first processed fingerprint image to be verified to obtain a second processed sample fingerprint image and a second processed fingerprint image to be verified. In this embodiment, the Hanning window function is used for windowing. In some feasible embodiments, rectangular window, Kaiser window, Hamming window and other window functions may also be used, which are not limited here. Figure 6 As shown, it also shows the effect diagram of the fingerprint image S2 obtained after the fingerprint image S1 is processed by windowing. It can be understood that after the windowing process, the fingerprint image is truncated and the frequency energy leakage is reduced as much as possible.

[0052] Step S105, Fourier transform the second processed sample fingerprint image and the second processed fingerprint image to be verified to obtain a third processed sample fingerprint image and a third processed fingerprint image to be verified. It can be understood that since the phase correlation algorithm is used to perform correlation operation on the two fingerprint images to obtain the rotation angle, it is necessary to transform the two fingerprint images from time domain images to frequency domain images. Figure 6 As shown, it also shows the effect diagram of the fingerprint image S3 obtained by Fourier transforming the fingerprint image S2.

[0053] Step S107, frequency domain filtering is performed on the third processed sample fingerprint image and the third processed fingerprint image to be verified to obtain a sample fingerprint image and a fingerprint image to be verified. Specifically, in this embodiment, the frequency domain images of the low-frequency parts of the third processed sample fingerprint image and the third processed fingerprint image to be verified are first obtained; then the frequency domain images of the low-frequency parts are subjected to bandpass filtering to obtain the sample fingerprint image M0 and the fingerprint image to be verified R0.

[0054] Please refer to Fig.10 In order to further clearly show the process of obtaining the sample fingerprint image M0 and the fingerprint image to be verified R0 after the original sample fingerprint image M and the original fingerprint image to be verified R are enhanced, Fig.10The program modules of the enhancement processing module 400 corresponding to the implementation step S100 are shown in FIG. The enhancement processing module 400 includes a Gaussian difference filter module P1, a windowing processing module P2, a Fourier transform module P3, and a frequency domain filter module P4, which are arranged in sequence. The original sample fingerprint image M and the original fingerprint image to be verified R are respectively input into the Gaussian difference filter module P1 for calculation to obtain the first processed sample fingerprint image and the first processed fingerprint image to be verified. Then, the first processed sample fingerprint image and the first processed fingerprint image to be verified are input into the windowing processing module P2 to obtain the second processed sample fingerprint image and the second processed fingerprint image to be verified. Then, the second processed sample fingerprint image and the second processed fingerprint image to be verified are input into the Fourier transform module P3 to obtain the third processed sample fingerprint image and the third processed fingerprint image to be verified; then, the third processed sample fingerprint image and the third processed fingerprint image to be verified are input into the frequency domain filter module P4 to obtain the sample fingerprint image M0 and the fingerprint image to be verified R0.

[0055] Please see Figure 7 , which is a schematic diagram of the sub-step flow chart of step S200 provided in an embodiment, that is, step S200 includes the following steps.

[0056] Step S201, using a phase correlation algorithm to calculate the correlation rotation angle between the sample fingerprint image and the fingerprint image to be verified, the correlation rotation angle includes a first correlation rotation angle and a second correlation rotation angle, the difference between the first correlation rotation angle and the second correlation rotation angle is 180 0 In this embodiment, the phase correlation algorithm adopts a pure phase correlation (Phase-only correlation, POC) algorithm.

[0057] Specifically, in this embodiment, the sample fingerprint image M0 and the fingerprint image to be verified R0 are both rectangular coordinate images. When the POC algorithm is used to perform phase correlation operation on the two images to obtain the relevant rotation angle, the two images need to meet the condition that the two images are polar coordinate images. Therefore, before the POC algorithm performs phase correlation on the two images, the sample fingerprint image M0 and the fingerprint image to be verified R0 are transformed from the rectangular coordinate system to the polar coordinate system. Fig.14 As shown, step S201 further includes: step S2010, transforming the sample fingerprint image M0 and the fingerprint image to be verified R0 from the rectangular coordinate system to the polar coordinate system to obtain the polar coordinate system sample fingerprint image and the polar coordinate system to be verified fingerprint image; step S2012, using the POC algorithm to perform phase correlation operation on the polar coordinate system sample fingerprint image and the polar coordinate system to be verified fingerprint image to obtain the relevant rotation angle. Please refer to Figure 6 , taking fingerprint image S4 as an example, Figure 6The effect diagram of fingerprint image S4 obtained after fingerprint image S3 is subjected to frequency domain filtering and polar coordinate transformation is also shown.

[0058] Please refer to Fig.10 In order to further clearly show the process of obtaining the relevant rotation angle through phase correlation operation between the sample fingerprint image and the fingerprint image to be verified, Fig.10 The polar coordinate transformation module P5 and phase correlation module P6 corresponding to step S201 are shown in FIG. Specifically, the sample fingerprint image M0 and the fingerprint image to be verified R0 output by the frequency domain filtering module P4 are respectively input into the polar coordinate transformation module P5 to be converted into the polar coordinate system sample fingerprint image and the polar coordinate system fingerprint image to be verified.

[0059] Since the relevant rotation angles are calculated using the POC algorithm, the relevant rotation angles include a first relevant rotation angle θ1 and a second relevant rotation angle θ2. For example, the first relevant rotation angle is 30 0 , the second related rotation angle is 210 0 . The actual rotation angle is one of the first relevant rotation angle θ1 or the second relevant rotation angle θ2. How to determine the actual rotation angle θ, that is, whether θ is the first relevant rotation angle θ1 or the second relevant rotation angle θ2, will be described in detail below. It can be understood that when the sample fingerprint image M0 and the fingerprint image to be verified R0 are polar coordinate system images, no coordinate system transformation is required.

[0060] Step S202, using the phase correlation algorithm, respectively calculates the first correlation translation amount and the second correlation translation amount between the sample fingerprint image and the fingerprint image to be verified at the first correlation rotation angle and the second correlation rotation angle. That is, the correlation translation amount between the sample fingerprint image M and the fingerprint image to be verified R at the first correlation rotation angle θ1 is the first correlation translation amount D1, and the correlation translation amount between the sample fingerprint image M and the fingerprint image to be verified R at the first correlation rotation angle θ2 is the second correlation translation amount D2.

[0061] like Figure 8 As shown, in this embodiment, step S202 specifically includes the following steps.

[0062] Step S2020: Expand the sample fingerprint image and the fingerprint image to be verified to obtain an expanded sample fingerprint image and an expanded fingerprint image to be verified. Fig.12As shown, the sample fingerprint image M0 is expanded, that is, the background of the sample fingerprint image M0 is expanded to obtain the background image A. The sample fingerprint image M0 is located at the center of the background image A. Specifically, the width w of the background image A is the diagonal length C1 of the sample fingerprint image M0. Similarly, the fingerprint image R0 to be verified is expanded, that is, the background of the fingerprint image R0 to be verified is expanded to obtain its background image (not shown). The fingerprint image R0 to be verified is located at the center of its background image. Specifically, the width of the background image of the fingerprint image R0 to be verified is the diagonal length of the fingerprint image R0 to be verified.

[0063] Step S2021, the expanded sample fingerprint image and the expanded fingerprint image to be verified are respectively rotated by a first related rotation angle to obtain a first rotated sample fingerprint image and a first rotated fingerprint image to be verified. Since the sample fingerprint image M0 and the fingerprint image to be verified R0 are expanded before this step, when the sample fingerprint image M0 and the fingerprint image to be verified R0 are rotated, the sample fingerprint image M0 and the fingerprint image to be verified R0 are in their respective background images, thereby ensuring that the data of the sample fingerprint image M0 and the fingerprint image to be verified R0 are not lost.

[0064] Step S2022: Fourier transform the first rotated sample fingerprint image and the first rotated fingerprint image to be verified to obtain a first transformed sample fingerprint image and a first transformed fingerprint image to be verified. It can be understood that Fourier transforming the first rotated sample fingerprint image and the first rotated fingerprint image to be verified can remove noise.

[0065] Step S2023: performing phase correlation operation on the first transformed sample fingerprint image and the first transformed fingerprint image to be verified to obtain a first correlation shift amount.

[0066] Step S2024: Rotate the expanded sample fingerprint image and the expanded fingerprint image to be verified by a second related rotation angle to obtain a second rotated sample fingerprint image and a second rotated fingerprint image to be verified.

[0067] Step S2025, Fourier transform the second rotated sample fingerprint image and the second rotated fingerprint image to be verified to obtain a second transformed sample fingerprint image and a second transformed fingerprint image to be verified. The Fourier transform of the second rotated sample fingerprint image and the second rotated fingerprint image to be verified can also remove noise from the second rotated sample fingerprint image and the second rotated fingerprint image to be verified.

[0068] Step S2026: performing phase correlation operation on the second transformed sample fingerprint image and the second transformed fingerprint image to be verified to obtain a second correlation shift amount.

[0069] Please see Fig.13In order to further clearly show the process of obtaining the relevant translation amount through phase correlation operation of the original sample fingerprint image M and the original fingerprint image to be verified R, Fig.13 , the program modules of the translation amount calculation module 500 for implementing step S200 are shown in FIG. The translation amount calculation module 500 includes a background expansion module 501, a rotation processing module 502, a Fourier transform module 503, and a phase correlation module 505, which are arranged in sequence. The sample fingerprint image M0 and the fingerprint image to be verified R0 are respectively input into the background expansion module 501 for background expansion to obtain an expanded sample fingerprint image and an expanded fingerprint image to be verified. Then, the expanded sample fingerprint image and the expanded fingerprint image to be verified are respectively input into the rotation processing module 502 to rotate the first relevant rotation angle to obtain a first rotated sample fingerprint image and a first rotated fingerprint image to be verified; then, the first rotated sample fingerprint image and the first rotated fingerprint image to be verified are input into the Fourier transform module 503 for transformation to obtain a first transformed sample fingerprint image and a first transformed fingerprint image to be verified; then, the first transformed sample fingerprint image and the first transformed fingerprint image to be verified are input into the phase correlation calculation module 505 for correlation operation to obtain a first relevant translation amount. Next, the expanded sample fingerprint image and the expanded fingerprint image to be verified are respectively input into the rotation processing module 502 to be rotated by a second relevant rotation angle to obtain a second rotated sample fingerprint image and a second rotated fingerprint image to be verified; next, the second rotated sample fingerprint image and the second rotated fingerprint image to be verified are input into the Fourier transform module 503 for transformation to obtain a second transformed sample fingerprint image and a second transformed fingerprint image to be verified; then, the second transformed sample fingerprint image and the second transformed fingerprint image to be verified are input into the phase correlation operation module 505 to obtain a second relevant translation amount.

[0070] Please see again Figure 7, step S203, determining the actual rotation amount and the actual translation amount according to the peak values ​​of the sample fingerprint image or the image to be verified on the first relevant translation amount and the second relevant translation amount. Specifically, the first relevant translation amount and the second relevant translation amount corresponding to the maximum peak values ​​of the sample fingerprint image on the first relevant translation amount and the second relevant translation amount are selected as the actual translation amount, and the first rotation angle or the second rotation angle corresponding to the actual translation amount is used as the actual rotation angle. Among them, when it is necessary to rotate the sample fingerprint image M0, the actual rotation amount and the actual translation amount are determined according to the peak values ​​of the sample fingerprint image M0 on the first relevant translation amount and the second relevant translation amount. When it is necessary to rotate the fingerprint image R0 to be verified, the actual rotation amount and the actual translation amount are determined according to the peak values ​​of the fingerprint image R0 to be verified on the first relevant translation amount and the second relevant translation amount. Taking the sample fingerprint image M0 that needs to be rotated as an example, when the sample fingerprint image has the largest peak value on the first relevant translation amount D1 and the second relevant translation amount D2, the corresponding relevant translation amount is the first relevant translation amount D1, and the first relevant translation amount D1 and the first relevant rotation angle θ1 are used as the actual translation amount D and the actual rotation angle θ. Conversely, the second relevant translation amount D2 and the second relevant rotation angle θ2 are used as the actual translation amount D and the actual rotation angle θ.

[0071] In the above embodiment, the phase correlation algorithm is used to determine the actual rotation amount and the actual translation amount, and when calculating the actual translation amount, the sample fingerprint image and the fingerprint image to be verified are expanded to ensure that the data of the sample fingerprint image and the fingerprint image to be verified are not lost.

[0072] In some feasible embodiments, the sample fingerprint image M0 includes a plurality of sample fingerprint image slices M01~M0n. The fingerprint image R0 to be verified includes a plurality of fingerprint image slices R01~R0n to be verified. The plurality of sample fingerprint image slices M01~M0n and the plurality of fingerprint image slices R01~R0n to be verified correspond to each other one by one. In order to further improve the accuracy of calculating the relevant rotation angle between the sample fingerprint image M0 and the fingerprint image R0 to be verified, firstly, the relevant rotation angle of each pair of sample fingerprint image slices and fingerprint image slices to be verified is obtained for each pair of sample fingerprint image slices and fingerprint image slices to be verified, thereby obtaining a plurality of relevant rotation angles; then, the relevant rotation angle between the sample fingerprint image M0 and the fingerprint image R0 to be verified is determined according to the minimum value of the plurality of relevant rotation angles. Among them, the steps of forming the plurality of sample fingerprint image slices M01~M0n from the sample fingerprint image M0 and forming the plurality of fingerprint image slices R01~R0n to be verified from the fingerprint image R01 to be verified, and the steps of calculating the relevant rotation angle between the sample fingerprint image M0 and the fingerprint image R0 to be verified will be described below.

[0073] Please see Fig. 9 and Fig.11, which is a sub-step flow chart of step S100 provided in another embodiment. Step S100 provided in another embodiment includes the following steps.

[0074] Step S101', cutting the original sample fingerprint image and the original fingerprint image to be verified to form multiple pairs of original sample fingerprint image slices and original fingerprint image slices to be verified. Specifically, the original sample fingerprint image M and the original fingerprint image to be verified R are cut to form multiple pairs of original sample fingerprint image slices M1~Mn and original fingerprint image slices R1~Rn. Wherein, n is an integer greater than 1. For example, the original sample fingerprint image M and the original fingerprint image to be verified R are images of size 160*80 respectively, and 8*8 original sample fingerprint image slices and 8*8 original fingerprint image slices to be verified need to be formed, and each original sample fingerprint image slice and each original fingerprint image slice to be verified can be set to an image of size 20*10.

[0075] Step S102', performing Gaussian difference processing on each pair of original sample fingerprint image slices and original fingerprint image slices to be verified in the plurality of pairs to obtain a plurality of pairs of first processed sample fingerprint image slices and first processed fingerprint image slices to be verified.

[0076] Step S103', performing windowing processing on each pair of the plurality of first processed sample fingerprint image slices and the first processed fingerprint image slices to be verified to obtain a plurality of pairs of second processed sample fingerprint image slices and second processed fingerprint image slices to be verified.

[0077] Step S104', Fourier transform each pair of the plurality of pairs of second processed sample fingerprint image slices and second processed fingerprint image slices to be verified to obtain a plurality of pairs of third processed sample fingerprint image slices and third processed fingerprint image slices to be verified.

[0078] Step S105', frequency domain filtering each pair of the third processed sample fingerprint image slices and the third processed fingerprint image slices to be verified to obtain multiple pairs of sample fingerprint image slices and the fingerprint image slices to be verified.

[0079] Please refer to Fig.11 , in order to further clearly demonstrate the process of obtaining sample fingerprint image slices M01~M0n and fingerprint image slices R01~R0n after the original sample fingerprint image M and the original fingerprint image R to be verified are enhanced. Fig.11, the program modules of the n enhancement processing modules 400 corresponding to the implementation of step S100 are shown. Where n is an integer greater than 1. Each enhancement processing module 400 processes a pair of original sample fingerprint image slices and a fingerprint image slice to be verified respectively. That is, the first enhancement processing module 400 performs enhancement processing on the original sample fingerprint image slice M1 and the original fingerprint image slice to be verified R1, and the second enhancement processing module performs enhancement processing on the sample fingerprint image slice. Wherein, the enhancement processing module 400 processes a pair of sample fingerprint image slices and a fingerprint image slice to be verified respectively, which is the same as the enhancement processing module 400 processes the sample fingerprint image M and the fingerprint image R to be verified, and will not be repeated here.

[0080] Please see Fig.15 , which is a sub-step flow chart of step S201 provided in another embodiment. Step S201 provided in another embodiment includes the following steps.

[0081] Step S2010', transform each pair of sample fingerprint image slices and fingerprint image slices to be verified from the rectangular coordinate system to the polar coordinate system to obtain multiple pairs of polar coordinate system sample fingerprint image slices and polar coordinate system fingerprint image slices to be verified.

[0082] Step S2012', performing phase correlation operation on each pair of polar coordinate system sample fingerprint image slices and polar coordinate system to-be-verified fingerprint image slices in the plurality of polar coordinate system sample fingerprint image slices and polar coordinate system to-be-verified fingerprint image slices to obtain a plurality of pending rotation angles.

[0083] Step S2014': taking the smallest rotation angle among the multiple rotation angles to be determined as the first relevant rotation angle.

[0084] Please refer to Fig.11 In order to further clearly show the process of obtaining the relevant rotation angle through phase correlation operation between the sample fingerprint image and the fingerprint image to be verified, Fig.11 The polar coordinate transformation module P5, phase correlation module P6 and comparator P7 for implementing step S201 are also shown. Specifically, each pair of sample fingerprint image slices and fingerprint image slices to be verified output by the frequency domain filtering module P4 are respectively input into the polar coordinate transformation module P5 to be converted into multiple pairs of polar coordinate system sample fingerprint image slices and polar coordinate system fingerprint image slices to be verified. Each pair of polar coordinate system sample fingerprint image and polar coordinate system fingerprint image to be verified is subjected to correlation operation by the phase correlation module P6 to obtain multiple rotation angles to be determined. Multiple rotation angles to be rotated are input into the comparator P7 for comparison, and the smallest rotation angle is selected as the correlation rotation angle between the sample fingerprint image and the fingerprint image to be verified.

[0085] Please see Fig.16 , Fig.16Schematic diagram of the internal structure of the fingerprint recognition sensor 1. The fingerprint recognition sensor 1 includes a storage medium 15 and a processor 16. The storage medium 15 is used to store fingerprint recognition program instructions. The processor 16 is used to execute the fingerprint recognition program instructions to implement the above-mentioned fingerprint recognition method, which is the same as the above-mentioned fingerprint recognition methods and will not be repeated here.

[0086] Fig.16 Only the electronic device 100 having the components 15 and 16 is shown, and it can be understood by those skilled in the art that Fig.15 The structure shown does not constitute a limitation on the fingerprint recognition sensor 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0087] Please refer to Figure 2 and Fig.16 , the fingerprint recognition sensor 1 is applied to the electronic device 100. The electronic device 100 may be a mobile phone, a tablet computer, a laptop computer, an access control system, etc. In this embodiment, the specific application of the fingerprint recognition sensor 1 is described by taking the electronic device 100 as a mobile phone as an example. The electronic device 100 includes a main body 1001. The main body 1001 includes a front side 1002, a back side 1003, and a side 1004 located between the front side 1002 and the back side 1003. The sensing area 10 of the fingerprint recognition sensor 1 is arranged on the side 1004 of the main body 1001. In some feasible embodiments, the sensing area 10 can also be arranged at other positions of the electronic device 100, which is not limited here. It can be understood that when the electronic device is other electronic products, the position of the sensing area 10 can be changed according to the actual design. For example, when the electronic device 100 is an access control system, the sensing area 10 can be arranged on the front of the access control system.

[0088] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A fingerprint recognition method, characterized in that: The fingerprint recognition method comprises: Calculating the actual rotation angle and the actual translation amount between the sample fingerprint image and the fingerprint image to be verified, wherein calculating the actual rotation angle between the sample fingerprint image and the fingerprint image to be verified specifically comprises: calculating multiple pending rotation angles between multiple pairs of sample fingerprint image slices and fingerprint image slices to be verified, and taking the smallest pending rotation angle among the multiple pending rotation angles as the actual rotation angle, wherein the sample fingerprint image comprises multiple sample fingerprint image slices, the fingerprint image to be verified comprises multiple fingerprint image slices to be verified, and the multiple sample fingerprint image slices correspond to the multiple fingerprint image slices to be verified one by one; Rotate one of the sample fingerprint image and the fingerprint image to be verified by the actual rotation angle and translate by the actual translation amount so that the sample fingerprint image and the fingerprint image to be verified form a first fingerprint image to be paired and a second fingerprint image to be paired respectively; Calculating an overlapping area between the first fingerprint image to be paired and the second fingerprint image to be paired; The first overlapping fingerprint image and the second overlapping fingerprint image corresponding to the overlapping area are matched to obtain a verification result, specifically comprising: using a scoring function to score the pixel points of the first overlapping fingerprint image and the pixel points of the second overlapping fingerprint image weighted according to the detail features to obtain a verification result; wherein: the first overlapping fingerprint image is an image corresponding to the first fingerprint image to be paired and the overlapping area, and the second overlapping fingerprint image is an image corresponding to the second fingerprint image to be paired and the overlapping area.

2. The fingerprint recognition method according to claim 1, characterized in that: The fingerprint recognition method further comprises: The original sample fingerprint image and the original fingerprint image to be verified are enhanced to obtain the sample fingerprint image and the fingerprint image to be verified.

3. The fingerprint recognition method according to claim 2, characterized in that: Performing enhancement processing on the original sample fingerprint image and the original fingerprint image to be verified to obtain the sample fingerprint image and the fingerprint image to be verified specifically includes: Performing Gaussian difference processing on the original sample fingerprint image and the original fingerprint image to be verified to obtain a first processed sample fingerprint image and a first processed fingerprint image to be verified; Performing windowing processing on the first processed sample fingerprint image and the first processed fingerprint image to be verified to obtain a second processed sample fingerprint image and a second processed fingerprint image to be verified; Performing Fourier transformation on the second processed sample fingerprint image and the second processed fingerprint image to be verified to obtain a third processed sample fingerprint image and a third processed fingerprint image to be verified; and The third processed sample fingerprint image and the third processed fingerprint image to be verified are subjected to frequency domain filtering to obtain the sample fingerprint image and the fingerprint image to be verified.

4. The fingerprint recognition method according to claim 1, characterized in that: Calculate the actual rotation angle and actual translation between the sample fingerprint image and the fingerprint image to be verified, including: Calculating the correlation rotation angle between the sample fingerprint image and the fingerprint image to be verified by using a phase correlation algorithm, wherein the correlation rotation angle includes a first correlation rotation angle and a second correlation rotation angle, and the difference between the first correlation rotation angle and the second correlation rotation angle is 180°; Calculating a first correlation translation amount and a second correlation translation amount between the sample fingerprint image and the fingerprint image to be verified at the first correlation rotation angle and the second correlation rotation angle respectively by using a phase correlation algorithm; and The actual rotation angle and the actual translation amount are determined according to peak values ​​of the sample fingerprint image or the image to be verified on the first related translation amount and the second related translation amount.

5. The fingerprint recognition method according to claim 4, characterized in that: The sample fingerprint image and the fingerprint image to be verified are both rectangular coordinate system images, and the phase correlation algorithm is used to calculate the correlation rotation angle between the sample fingerprint image and the fingerprint image to be verified, specifically including: Transforming the sample fingerprint image and the fingerprint image to be verified from the rectangular coordinate system to the polar coordinate system to obtain a polar coordinate system sample fingerprint image and a polar coordinate system fingerprint image to be verified; and A phase correlation operation is performed on the polar coordinate system sample fingerprint image and the polar coordinate system fingerprint image to be verified to obtain the correlation rotation angle.

6. The fingerprint recognition method according to claim 5, characterized in that: Calculating the first correlation translation and the second correlation translation between the sample fingerprint image and the fingerprint image to be verified at the first correlation rotation angle and the second correlation rotation angle respectively by using a phase correlation algorithm, specifically comprising: Expanding the sample fingerprint image and the fingerprint image to be verified to obtain an expanded sample fingerprint image and an expanded fingerprint image to be verified; Rotate the expanded sample fingerprint image and the expanded fingerprint image to be verified by the first relevant rotation angle to obtain a first rotated sample fingerprint image and a first rotated fingerprint image to be verified; Performing Fourier transformation on the first rotated sample fingerprint image and the first rotated fingerprint image to be verified to obtain a first transformed sample fingerprint image and a first transformed fingerprint image to be verified; Performing a phase correlation operation on the first transformed sample fingerprint image and the first transformed fingerprint image to be verified to obtain a first correlation translation amount; Rotate the expanded sample fingerprint image and the expanded fingerprint image to be verified by the second relevant rotation angle to obtain a second rotated sample fingerprint image and a second rotated fingerprint image to be verified; Performing Fourier transformation on the second rotated sample fingerprint image and the second rotated fingerprint image to be verified to obtain a second transformed sample fingerprint image and a second transformed fingerprint image to be verified; and A phase correlation operation is performed on the second transformed sample fingerprint image and the second transformed fingerprint image to be verified to obtain the second correlation shift amount.

7. The fingerprint recognition method according to claim 2, characterized in that: The original sample fingerprint image and the original fingerprint image to be verified are enhanced to obtain the sample fingerprint image and the fingerprint image to be verified, specifically including: Cutting the original sample fingerprint image and the original fingerprint image to be verified to form multiple pairs of original sample fingerprint image slices and original fingerprint image slices to be verified; Performing Gaussian difference processing on each pair of original sample fingerprint image slices and original fingerprint image slices to be verified among the plurality of pairs of original sample fingerprint image slices and original fingerprint image slices to be verified to obtain a plurality of pairs of first processed sample fingerprint image slices and first processed fingerprint image slices to be verified; Performing windowing processing on each pair of the plurality of pairs of first processed sample fingerprint image slices and first processed fingerprint image slices to be verified to obtain a plurality of pairs of second processed sample fingerprint image slices and second processed fingerprint image slices to be verified; Performing Fourier transformation on each pair of the plurality of pairs of second processed sample fingerprint image slices and second processed fingerprint image slices to be verified to obtain a plurality of pairs of third processed sample fingerprint image slices and third processed fingerprint image slices to be verified; and Each pair of the plurality of pairs of third processed sample fingerprint image slices and third processed fingerprint image slices to be verified is subjected to frequency domain filtering to obtain the plurality of pairs of sample fingerprint image slices and the fingerprint image slices to be verified.

8. The fingerprint recognition method according to claim 7, characterized in that: Calculate the actual rotation angle and actual translation between the sample fingerprint image and the fingerprint image to be verified, including: Calculating the correlation rotation angle between the sample fingerprint image and the fingerprint image to be verified by using a phase correlation algorithm, wherein the correlation rotation angle includes a first correlation rotation angle and a second correlation rotation angle, and the difference between the first correlation rotation angle and the second correlation rotation angle is 180°; Calculating a first correlation translation amount and a second correlation translation amount between the sample fingerprint image and the fingerprint image to be verified at the first correlation rotation angle and the second correlation rotation angle respectively by using a phase correlation algorithm; and The actual rotation angle and the actual translation amount are determined according to peak values ​​of the sample fingerprint image or the image to be verified on the first related translation amount and the second related translation amount.

9. The fingerprint recognition method according to claim 8, characterized in that: Calculating the correlation rotation angle between the sample fingerprint image and the fingerprint image to be verified by using a phase correlation algorithm specifically includes: Transforming each pair of sample fingerprint image slices and the fingerprint image slices to be verified in the plurality of pairs of sample fingerprint image slices and the fingerprint image slices to be verified from a rectangular coordinate system to a polar coordinate system to obtain a plurality of pairs of polar coordinate system sample fingerprint image slices and polar coordinate system fingerprint image slices to be verified; Performing phase correlation operations on each pair of polar coordinate system sample fingerprint image slices and polar coordinate system to-be-verified fingerprint image slices in the plurality of pairs of polar coordinate system sample fingerprint image slices and polar coordinate system to-be-verified fingerprint image slices to obtain a plurality of undetermined rotation angles; and The smallest of the multiple undetermined rotation angles is used as the first related rotation angle.

10. The fingerprint recognition method according to claim 1, characterized in that: Performing a matching operation on the first overlapping fingerprint image and the second overlapping fingerprint image corresponding to the overlapping area to obtain a verification result specifically includes: The first overlapping fingerprint image and the second overlapping fingerprint image are scored using a scoring function to obtain a verification result, and the scoring function is: Wherein, g(x,y) is used to represent the pixel points of the first overlapping fingerprint image; h(x,y) is used to represent the pixel points of the second overlapping fingerprint image; ∑ x,y g(x,y).h(x,y) is used to represent the sum of the products of the one-to-one corresponding pixels in the first overlapping fingerprint image and the second overlapping image; Used to represent the square root of the product of the sum of the squares of the pixels in the first overlapping fingerprint image and the sum of the squares of the pixels in the second overlapping image.

11. The fingerprint recognition method according to claim 1, characterized in that: Performing a matching operation on the first overlapping fingerprint image and the second overlapping fingerprint image corresponding to the overlapping area to obtain a verification result specifically includes: identifying fingerprint minutiae features of the first overlapping fingerprint image and the second overlapping fingerprint image; Weighting the pixels in the first overlapping fingerprint image and the second overlapping fingerprint image according to the identified detail features to obtain weighted pixel points of the first overlapping fingerprint image and weighted pixel points of the second overlapping fingerprint image, wherein the weight of the pixel points with detail features is greater than the weight of the pixel points without detail features; The pixel points of the weighted first overlapping fingerprint image and the pixel points of the weighted second overlapping fingerprint image are scored using a scoring function to obtain a verification result, and the scoring function is: Wherein, g(x, y) is used to represent the pixel points of the weighted first overlapping fingerprint image; h(x, y) is used to represent the pixel points of the weighted second overlapping fingerprint image; Σ x,y g(x,y).h(x,y) is used to represent the sum of the products of the one-to-one corresponding pixels in the weighted first overlapping fingerprint image and the weighted second overlapping image; Used to represent the square root of the product of the sum of the squares of the pixels in the weighted first overlapping fingerprint image and the sum of the squares of the pixels in the weighted second overlapping image.

12. A fingerprint recognition sensor, characterized in that: The fingerprint recognition sensor comprises: A storage medium for storing fingerprint recognition program instructions; and A processor, configured to execute the fingerprint recognition program instructions to implement the fingerprint recognition method according to any one of claims 1 to 11.

13. An electronic device, characterized in that: The electronic device includes a main body and the fingerprint recognition sensor as claimed in claim 12 arranged on the main body.

Images