Side fingerprint input method, electronic device, and computer-readable storage medium

Abstract

The application provides a side fingerprint input method, an electronic device and a computer readable storage medium. The side fingerprint input method comprises the following steps: initializing an n*n input matrix, the element of the a-th row and the b-th column of the input matrix being a matrix H ab ; sequentially extracting feature template information of first to n-th fingerprint images, and calculating a transformation matrix between feature template information of any two fingerprint images; for the a-th fingerprint image and the b-th fingerprint image, when the transformation matrix can be calculated, filling the transformation matrix into the matrix H ab , and when the transformation matrix cannot be calculated, filling a zero matrix into the matrix H ab ; traversing the zero matrix in the input matrix, for the zero matrix H jk , finding transformation matrices H jl and H lk , and replacing the zero matrix H jk with the product of the two. The feature information templates of mutually isolated fingerprint images can be reduced, the correlation between the feature information templates of the fingerprint images is improved, and the accuracy of fingerprint input and subsequent unlocking of a user is improved.

Description

Technical Field

[0001] This invention relates to the field of fingerprint recognition, and more specifically to a side fingerprint enrollment method, an electronic device, and a computer-readable storage medium. Background Technology

[0002] With the popularity of full-screen smartphones, fingerprint recognition sensors have gradually shifted from capacitive to under-display optical sensors. However, due to the high cost of optical sensors, side-mounted fingerprint sensors have become another option for manufacturers to reduce costs while maintaining aesthetics. Side-mounted fingerprint sensors refer to capacitive fingerprint sensors that are the same size as the phone's power button and are placed on one side of the phone. They are characterized by their particularly narrow width and relatively long length.

[0003] This type of narrow-edge sensor collects less information in the width direction. Existing fingerprint feature registration methods are relatively simple: with a single finger press, the fingerprint sensor captures a fingerprint image and extracts feature information. If the finger is pressed again, the fingerprint sensor captures a fingerprint image, extracts feature information, and compares it with the previous information. If the overlapping area is large, registration fails; otherwise, registration succeeds. Registration ends when the maximum number of feature templates is reached.

[0004] This simple registration method, when using a side-mounted fingerprint sensor, often results in isolated templates that fail to form a cohesive whole. Consequently, the entered feature information is unrelated, and authentication relies solely on individual features against the currently acquired template. Due to the limited information on the narrow edge, authentication frequently fails. Furthermore, when authenticating by rotating the finger, the direction is nearly perpendicular to the input direction. Many templates in the template library are nearly perpendicular to the currently authenticated template. However, with a narrow fingerprint sensor, the overlap area is small, leading to poor unlocking performance regardless of the number of template attempts. Summary of the Invention

[0005] The purpose of this invention is to provide a side fingerprint enrollment method, an electronic device, and a computer-readable storage medium.

[0006] This invention provides a side fingerprint enrollment method, characterized by comprising the following steps:

[0007] Initialize an n×n input matrix, where the element in the a-th row and b-th column of the input matrix is ​​matrix H. ab ;

[0008] Extract the feature template information of the first to nth fingerprint images sequentially, and calculate the transformation matrix between any two fingerprint image feature template information;

[0009] For fingerprint images a and b, when the transformation matrix can be calculated, the transformation matrix is ​​filled into matrix H. abWhen the transformation matrix cannot be calculated, the zero matrix is ​​filled into the matrix H. ab Inside;

[0010] Iterate through the zero matrix within the input matrix, for the zero matrix H jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix H with the product of the two. jk .

[0011] As a further improvement of the present invention, "an n×n input matrix is ​​initialized, wherein the elements of the a-th row and b-th column of the input matrix are matrix H". ab Specifically, it includes:

[0012] Initialize an n×n input matrix, where the element in the a-th row and b-th column of the input matrix is ​​matrix H. ab And fill the identity matrix into the matrix located on the main diagonal of the input matrix.

[0013] As a further improvement of the present invention, "sequentially extracting the feature template information of the first to nth fingerprint images and calculating the transformation matrix between any two fingerprint image feature template information" specifically includes:

[0014] Feature template information is extracted sequentially from the first to the nth fingerprint images. After each extraction of feature template information, the transformation matrix of the feature template information between the xth (2≤x≤n) fingerprint image and the first to the (x-1)th fingerprint images is calculated sequentially.

[0015] As a further improvement of the present invention, "when the transformation matrix can be calculated, the transformation matrix is ​​filled into the corresponding elements of the input matrix; when the transformation matrix cannot be calculated, the zero matrix is ​​filled into the corresponding elements of the input matrix" specifically includes:

[0016] For the x-th fingerprint image and the xi (1≤i<x)-th fingerprint image, when the transformation matrix M of both can be calculated... x,x-i When, the transformation matrix M x,x-i and its inverse matrix M x-i,x Fill the initial matrix H with the following values ​​respectively. x,x-i sum matrix H x-i,x Within a given location, when the transformation matrix for both cannot be calculated, the matrix H is filled with zeros. x,x-i sum matrix H x-i,x Within the location.

[0017] As a further improvement to the present invention, "traversing the zero matrix within the input matrix, for the zero matrix H..." jk Find the transformation matrix H jl and transformation matrix Hlk And replace the zero matrix H with the product of the two. jk Specifically, it includes:

[0018] After extracting the feature template information of the first to nth fingerprint images and completely filling the input matrix, the zero matrix within the input matrix is ​​traversed. For the zero matrix H... jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix H with the product of the two. jk .

[0019] As a further improvement to the present invention, "traversing the zero matrix within the input matrix, for the zero matrix H..." jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix H with the product of the two. jk Specifically, it includes:

[0020] After extracting the feature template information from the first to the n / 2th fingerprint images and filling the corresponding input matrix, the zero matrix within the input matrix is ​​traversed. For the zero matrix H... jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix H with the product of the two. jk .

[0021] As a further improvement to the present invention, the following steps are also included:

[0022] For the zero matrix H jk When the transformation matrix H does not exist simultaneously jl and transformation matrix H lk When the fingerprint image j and the fingerprint image k are used as isolated feature information templates, the feature information templates are used.

[0023] As a further improvement to the present invention, the following steps are also included:

[0024] After each extraction of the feature template information of a fingerprint image, it is compared with the feature template information of other fingerprint images. If the repetition of the fingerprint image with the feature template information of any of the fingerprint images is greater than the repetition threshold, the fingerprint image is required to be re-entered and the feature template information is re-extracted.

[0025] As a further improvement of the present invention, the feature template information includes key point information within the fingerprint image and descriptors corresponding to the key points.

[0026] As a further improvement of the present invention, the transformation matrix is ​​an affine transformation matrix.

[0027] As a further improvement of the present invention, the transformation matrix is ​​a rigid body transformation matrix.

[0028] As a further improvement of the present invention, each of the independent parameters of the rigid body transformation matrix is ​​provided with a deviation threshold.

[0029] The present invention also provides an electronic device, including a fingerprint sensor, a memory, and a processor. The fingerprint sensor is disposed on the side of the electronic device, and the memory stores a computer program that can run on the processor. When the processor executes the program, it implements the steps of the above-described side fingerprint enrollment method.

[0030] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed, causes the device containing the computer storage medium to perform the steps of the side fingerprint enrollment method described above.

[0031] The beneficial effects of this invention are as follows: By utilizing the matrix transformation characteristics, this invention integrates and calculates the transformation matrix between the feature information templates of multiple fingerprint images, which can reduce the number of mutually isolated fingerprint image feature information templates, improve the correlation between the feature information templates of each fingerprint image, and form a whole, thereby improving the accuracy of fingerprint enrollment and subsequent unlocking by the user, and enhancing the user experience. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the steps of a side fingerprint enrollment method according to an embodiment of the present invention.

[0033] Figures 2a to 2c These are schematic diagrams of the first fingerprint image, the second fingerprint image, and the third fingerprint image, respectively, in a specific embodiment of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0035] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0036] For ease of explanation, this document uses terms indicating relative spatial position, such as "above," "below," "behind," and "front," to describe the relationship of one unit or feature shown in the accompanying drawings relative to another unit or feature. Terms indicating relative spatial position can include different orientations of the device during use or operation besides those shown in the figures. For example, if the device in the figures is flipped, a unit described as being "below" or "above" other units or features will be located "below" or "above" other units or features. Therefore, the exemplary term "below" can encompass both "below" and "above" spatial orientations.

[0037] like Figure 1 As shown, the present invention provides a side fingerprint enrollment method, applicable to enrolling a user's fingerprint using a side fingerprint sensor in electronic devices such as mobile phones and tablets, comprising the following steps:

[0038] S1: Initialize the input matrix A, and input the elements of the a-th row and b-th column of the matrix as matrix H. ab .

[0039] Specifically, as shown below, an n×n input matrix A is initialized, where the element in the a-th row and b-th column of input matrix A is matrix H. ab And fill the identity matrix I into the matrix located on the main diagonal of the input matrix A, where matrix H ab It is a 3×3 matrix, corresponding to the subsequent transformation matrix.

[0040]

[0041]

[0042] For the input matrix, it has transition properties, matrix H ab It can be represented as matrix H ac sum matrix H cb The product of, such as H 22 =H 21 ×H 12 This property is strictly satisfied when the transformation relationship between matrices is a rigid body transformation. However, the fingerprint images collected by the fingerprint sensor often have certain errors due to factors such as finger bending and image deformation. Therefore, by setting a deviation threshold, it can be determined to be approximately satisfied to a certain extent.

[0043] Here, n is the maximum number of times the user can perform fingerprint image enrollment. Preferably, the value of n is in the range of 20 to 25, so as to ensure that the user's fingerprint can be completely enrolled using the side fingerprint sensor, while avoiding excessive fingerprint enrollment and long time consumption, which would lead to a decline in user experience.

[0044] S2: Extract feature template information from multiple fingerprint images sequentially, and calculate the transformation matrix between any two fingerprint image feature templates.

[0045] Specifically, feature template information is extracted sequentially from the first to the nth fingerprint images, and after each extraction of feature template information, the transformation matrix of feature template information between the xth (2≤x≤n) fingerprint image and the first to the (x-1)th fingerprint images is calculated sequentially.

[0046] Furthermore, after extracting the feature template information of each fingerprint image, the following steps are also included:

[0047] The feature information template of the currently extracted fingerprint image is compared with the feature information templates of other fingerprint images. If the repetition of the fingerprint image with the feature information template of any fingerprint image is greater than the repetition threshold, the fingerprint image is required to be re-entered and the feature information template is re-extracted.

[0048] For example, the repetition threshold can be set to 80%, and the fingerprint image feature information template extracted in the current time can be compared with the previous fingerprint image feature information template in the order of entry, thereby avoiding the entry of duplicate fingerprint images and improving the accuracy of fingerprint information collection.

[0049] Information feature templates are information templates generated from fingerprint images captured by a user through a fingerprint sensor located on the side of an electronic device. Due to the width limitation of the side of the electronic device, a single fingerprint image usually corresponds to the shape of the side fingerprint sensor, that is, it is usually long and narrow. After processing the fingerprint image, the key point information and the descriptors corresponding to the key points are extracted as the information feature template of the fingerprint image.

[0050] S3: When the transformation matrix can be calculated, fill the corresponding element of the input matrix with the transformation matrix. When the transformation matrix cannot be calculated, fill the corresponding element of the input matrix with the zero matrix.

[0051] Specifically, for the x-th fingerprint image and the xi (1≤i<x)-th fingerprint image, when the transformation matrix M of both can be calculated... x,x-i When, the transformation matrix M x,x-i and its inverse matrix M x-i,x Fill the initial matrix H with the following values ​​respectively. x,x-i And matrix H x-i,x Within a given location, if the transformation matrix between the two cannot be calculated, fill matrix H with zeros. x,x-i And matrix H x-i,x Within the location.

[0052] When there is an overlap between the feature information templates of two fingerprint images or the degree of overlap between them is greater than a judgment threshold, the transformation matrix between them can be calculated. When the two fingerprint images are isolated from each other, the transformation matrix between them cannot be calculated. In this case, the zero matrix is ​​temporarily used to fill the matrix corresponding to it.

[0053] Specifically, in some embodiments of this invention, the transformation matrix is ​​an affine transformation matrix M, meaning that the transformation between the feature template information of two fingerprint images is measured using an affine transformation matrix. An affine transformation is the process of converting one two-dimensional coordinate system to another. During the transformation, the relative positions and attributes of the coordinate points do not change; it is a linear transformation, involving only rotation and translation. The affine transformation matrix M is a 3×3 non-zero matrix containing six independent parameters.

[0054]

[0055] Where a1, a2, a3, and a4 represent rotation amounts, and ox and oy represent translation amounts.

[0056] In other embodiments of the present invention, the transformation matrix can also be a rigid body transformation matrix, that is, the transformation between the feature template information of two fingerprint images is measured by a rigid body transformation matrix. Since the rigid body transformation matrix only includes three independent parameters—rotation angle, horizontal translation component, and vertical translation component—the computational load can be reduced.

[0057] Furthermore, due to factors such as finger bending, the transfer characteristics under rigid body transformation often cannot be strictly satisfied. Therefore, each of the three parameters of the rigid body transformation matrix is ​​given a deviation threshold: the allowable deviation of the rotation angle is within ±5°, and the horizontal translation component and the vertical translation component are within ±5 pixels.

[0058] It should be noted that steps S2 and S3 are actually performed alternately and continuously, rather than executing step S2 first and then step S3.

[0059] S4: Traverse and input the zero matrix within the matrix, for the zero matrix H jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix H with the product of the two. jk .

[0060] Furthermore, for the zero matrix H jk When the transformation matrix H does not exist simultaneously jl and transformation matrix H lk When the fingerprint image j and the fingerprint image k are used as isolated feature information templates, the feature information templates are used.

[0061] Specifically, in some embodiments of the present invention, after the feature template information of the first to nth fingerprint images is completely extracted and the recording matrix A is completely filled, the zero matrix in the recording matrix is ​​traversed, and for the zero matrix H... jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix H with the product of the two. jk .

[0062] By utilizing the transfer properties to attempt to integrate zero elements within the input matrix A into non-zero elements, the number of isolated fingerprint image feature information templates can be reduced, allowing the fingerprint images to form a unified whole and improving the accuracy of fingerprint enrollment and subsequent unlocking. Furthermore, after fully filling the input matrix A and then integrating it, the data volume within matrix A is larger, and each zero element has a higher probability of integrating into a non-zero element. This results in stronger fingerprint coupling and a higher unlocking rate during subsequent matching.

[0063] In other embodiments of the present invention, after extracting the feature template information of the first to n / 2 fingerprint images and correspondingly filling the recording matrix, the zero matrix within the recording matrix is ​​traversed. For the zero matrix H... jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix H with the product of the two. jk .

[0064] At this point, the input matrix A is integrated. Compared to integrating after complete input, the integration time is only one-quarter due to the smaller data volume. Furthermore, the integration of input matrix A can be performed while the user is inputting the last n / 2 fingerprint images, thus completing the integration process without the user's awareness. Although the input matrix has fewer elements, potentially containing more zero elements and isolated fingerprint templates, the integration time is shorter, reducing user waiting time.

[0065] Alternatively, the time for integrating the input matrix A can be adjusted according to the value of n to achieve a better balance between input accuracy and user waiting time.

[0066] For example, taking the creation of a 3×3 input matrix A1 as an example, the specific steps of the present invention are explained as follows:

[0067] Sa: Initialize a 3×3 input matrix A1, and fill the matrix located on the main diagonal of the input matrix A1 with the identity matrix I.

[0068] Sb: After the user enters the first fingerprint image 1, such as Figure 2aThe diagram shown is a schematic of the first fingerprint image 1 in this embodiment. The feature template information of the first fingerprint image is extracted, and the identity matrix I is filled into matrix H. 11 Since the identity matrix I has already been filled into the elements on the main diagonal of the input matrix A1, the step of filling the identity matrix can be omitted.

[0069] At this point, the input matrix A1 can be represented as:

[0070]

[0071] Sc: After the user enters the second fingerprint image 2, as follows Figure 2b The diagram shows the first fingerprint image and the second fingerprint image 2 in this embodiment. The overlap between the first and second fingerprint images is greater than a judgment threshold. Feature template information of the second fingerprint image is extracted, and the affine transformation matrix M between it and the feature template information of the first fingerprint image is calculated. 21 Fill it into matrix H 21 Within the position, and transform matrix M 21 The inverse matrix M 12 Fill to matrix H 12 Within the location.

[0072] At this point, the input matrix A1 can be represented as:

[0073]

[0074] Where the brackets represent matrix H 12 sum matrix H 21 The actual content to be filled in is the transformation matrix within the parentheses.

[0075] Sd: After the user enters the third fingerprint image, such as Figure 2c The diagram shows the first, second, and third fingerprint images in this embodiment. The overlap between the first and second fingerprint images exceeds a threshold, the overlap between the second and third fingerprint images exceeds the threshold, and there is no overlap between the first and third fingerprint images. The feature template information of the third fingerprint image is extracted, but since the affine transformation matrix M cannot be calculated between it and the feature template information of the first fingerprint image... 31 Therefore, the zero matrix is ​​filled into matrix H. 31 and matrix H 13 Within the location. Calculate the affine transformation matrix M between the feature template information of the third fingerprint image and the feature template information of the second fingerprint image. 32 Fill it into matrix H 32 Within the position, and transform matrix M 32 The inverse matrix M23 Fill to matrix H 23 Within the location.

[0076] At this point, the input matrix A1 can be represented as:

[0077]

[0078] Se: Integrate the data within the input matrix, traverse its elements, and find the element located in the original matrix H. 31 sum matrix H 13 The element at position M is a zero matrix, and the affine transformation matrix M is... 32 With the affine transformation matrix M 21 The affine transformation matrix M is obtained through matrix multiplication. 31 And fill it into matrix H 31 At the location, the affine transformation matrix M 32 With the affine transformation matrix M 21 The affine transformation matrix M is obtained through matrix multiplication. 31 And fill it into matrix H 31 At the designated location, the integration of the input matrix is ​​completed.

[0079] At this point, the input matrix A1 can be represented as:

[0080]

[0081] The above steps are illustrated using a 3×3 entry matrix. When the maximum number of entries is greater, the above method is repeated until the integrated entry matrix is ​​obtained. Furthermore, after the fingerprint image is entered, a step of detecting its repeatability may also be included.

[0082] The present invention also provides an electronic device, including a fingerprint sensor, a memory, and a processor. The fingerprint sensor is disposed on the side of the electronic device, and the memory stores a computer program that can run on the processor. When the processor executes the program, it implements the steps of the above-described side fingerprint enrollment method.

[0083] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed, causes the device in which the computer storage medium is located to perform the steps according to the side fingerprint enrollment method described above.

[0084] In summary, by utilizing the matrix transformation properties to integrate and calculate the transformation matrix between feature information templates of multiple fingerprint images, this invention can reduce the number of isolated fingerprint image feature information templates, improve the correlation between feature information templates of each fingerprint image, and form a whole, thereby improving the accuracy of fingerprint enrollment and subsequent unlocking, and enhancing the user experience.

[0085] It should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0086] The detailed descriptions listed above are merely specific descriptions of feasible implementations of the present invention and are not intended to limit the scope of protection of the present invention. All equivalent implementations or modifications made without departing from the spirit of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for side fingerprint enrollment, characterized in that, Including the following steps: Initialize a n×n Input matrix, the input matrix of the first a line, number b The elements of the column are matrices H ab ; Extract the first to the second in sequence n The feature template information of fingerprint images is used to calculate the transformation matrix between any two fingerprint image feature templates. For the a fingerprint images and the b For fingerprint images, when the transformation matrix can be calculated, the transformation matrix is ​​used to fill the matrix. H ab When the transformation matrix cannot be calculated, the zero matrix is ​​filled into the matrix. H ab Inside; Iterate through the zero matrix within the input matrix, for the zero matrix H jk Find the transformation matrix H jl and transformation matrix H lk and using the transformation matrix H jl and transformation matrix H lk The product replaces the zero matrix H jk .

2. The side fingerprint enrollment method according to claim 1, characterized in that, Initialize a n×n Input matrix, the input matrix of the first a line, number b The elements of the column are matrices H ab Specifically, it includes: Initialize a n×n The input matrix, the input matrix of the first... a line, number b The elements of the column are matrices H ab And fill the identity matrix into the matrix located on the main diagonal of the input matrix.

3. The side fingerprint enrollment method according to claim 2, characterized in that, Extract the first to the last in sequence n The process of "using fingerprint image feature template information and calculating the transformation matrix between any two fingerprint image feature templates" specifically includes: Extract the first to the second in sequence n The fingerprint image's feature template information is used, and after each extraction of feature template information, the first step is calculated sequentially. x fingerprint image and the first fingerprint image to the second x-1 The transformation matrix of feature template information between fingerprint images, where, 2 ≤x≤n .

4. The side fingerprint enrollment method according to claim 3, characterized in that, "When the transformation matrix can be calculated, the transformation matrix is ​​filled into the corresponding elements of the input matrix; when the transformation matrix cannot be calculated, the zero matrix is ​​filled into the corresponding elements of the input matrix." Specifically, this includes: For the x fingerprint images and the xi Fingerprint images, when the transformation matrix between the two can be calculated. M x,x-i When, the transformation matrix is... M x,x-i and its inverse matrix M x-i,x Fill the initial matrix with the matrix. H x,x-i sum matrix H x-i,x Within a given location, when the transformation matrix for both cannot be calculated, fill the matrix with zeros. H x,x-i sum matrix H x-i,x Within the location, 1≤i<x .

5. The side fingerprint enrollment method according to claim 3, characterized in that, "Traverse the zero matrix within the input matrix, for the zero matrix..." H jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix with the product of the two. H jk Specifically, it includes: When the first to the last are completely extracted n After obtaining the feature template information of the fingerprint image and completely filling the input matrix, the zero matrix within the input matrix is ​​traversed. For the zero matrix... H jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix with the product of the two. H jk .

6. The side fingerprint enrollment method according to claim 3, characterized in that, "Traverse the zero matrix within the input matrix, for the zero matrix..." H jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix with the product of the two. H jk Specifically, it includes: When extracting the first to the last n / 2 The feature template information of the fingerprint image is used to fill the corresponding input matrix. Then, the zero matrix within the input matrix is ​​traversed. For the zero matrix... H jk Find the transformation matrix H jl and transformation matrix H lk And replace the zero matrix with the product of the two. H jk .

7. The side fingerprint enrollment method according to claim 5 or 6, characterized in that, It also includes the following steps: For the zero matrix H jk When transformation matrices do not exist simultaneously H jl and transformation matrix H lk At that time, the first j fingerprint images and the k The feature information template of the fingerprint image is used as an isolated feature information template.

8. The side fingerprint enrollment method according to claim 1, characterized in that, It also includes the following steps: After each extraction of the feature template information of a fingerprint image, it is compared with the feature template information of other fingerprint images. If the repetition of the fingerprint image with the feature template information of any of the fingerprint images is greater than the repetition threshold, the fingerprint image is required to be re-entered and the feature template information is re-extracted.

9. The side fingerprint enrollment method according to claim 5, characterized in that, The feature template information includes key point information within the fingerprint image and descriptors corresponding to the key points.

10. The side fingerprint enrollment method according to claim 1, characterized in that, The transformation matrix is ​​an affine transformation matrix.

11. The side fingerprint enrollment method according to claim 1, characterized in that, The transformation matrix is ​​a rigid body transformation matrix.

12. The side fingerprint enrollment method according to claim 11, characterized in that, Each of the independent parameters of the rigid body transformation matrix is ​​assigned a deviation threshold.

13. An electronic device, characterized in that, The device includes a fingerprint sensor, a memory, and a processor. The fingerprint sensor is disposed on the side of the electronic device. The memory stores a computer program that can run on the processor. When the processor executes the program, it implements the steps of the side fingerprint enrollment method according to any one of claims 1-12.

14. A computer-readable storage medium, characterized in that, The computer program is stored therein, and when the computer program is executed, it causes the device containing the computer storage medium to perform the steps of the side fingerprint enrollment method according to any one of claims 1-12.

Images