Signature quality inspection method and device

Abstract

The present disclosure relates to the technical field of artificial intelligence, and particularly relates to a signature quality inspection method and device. The signature quality inspection method comprises the following steps: correcting a first image by using an affine transformation matrix to obtain a second image; comparing a template image corresponding to a form type of the first image with the second image to obtain a first signature area and a second signature area in the second image; and performing quality inspection on the first signature area and the second signature area by using structural similarity SSIN to determine the authenticity of the signature corresponding to the first image. The present disclosure can provide a signature quality inspection mechanism and improve the accuracy of signature quality inspection.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and in particular to a signature quality inspection method and apparatus. Background Technology

[0002] With the continuous progress of society, installation and maintenance services have been applied to various businesses. After the installation and maintenance service is completed, it can be evaluated. For example, users can confirm the completion of the service. The user's signature can be verified after completion. For example, it can be compared with the signature on the service order; however, there are cases where the user's signature cannot be obtained, making verification impossible. Summary of the Invention

[0003] This disclosure provides a signature quality inspection method and apparatus to provide a signature quality inspection mechanism and improve the accuracy of signature quality inspection. The technical solution of this disclosure is as follows:

[0004] According to a first aspect of the present disclosure, a signature quality inspection method is provided, comprising:

[0005] The first image is corrected using an affine transformation matrix to obtain the second image;

[0006] The template image corresponding to the form type of the first image and the second image are compared to obtain the first signature area and the second signature area in the second image;

[0007] The structural similarity index (SSIN) is used to perform quality checks on the first signature region and the second signature region to determine the authenticity of the signature corresponding to the first image.

[0008] According to some embodiments, before comparing the template image corresponding to the form type of the first image with the second image to obtain the first signature area and the second signature area in the second image, the method further includes:

[0009] Obtain the first set of key points of the first image and the second set of key points of any template image in the template image set;

[0010] Obtain the first key point descriptor vector corresponding to any first key point in the first key point set, and obtain the second key point descriptor vector corresponding to any second key point in the second key point set;

[0011] Normalize the first key point descriptor vector corresponding to any first key point to obtain the first SIFT feature vector corresponding to the first image.

[0012] Normalize the second key point descriptor vector corresponding to any second key point to obtain the second SIFT feature vector corresponding to any template image;

[0013] The similarity between the first image and any template image is obtained based on the Euclidean distance between the first SIFT feature vector and the second SIFT feature vector.

[0014] The form type of the template image with the highest similarity is used as the form type of the first image.

[0015] According to some embodiments, obtaining the first key point set of the first image and the second key point set of any template image in the template image set includes:

[0016] Obtain the first set of extreme points of the first image, and obtain the second set of extreme points of any template image in the template image set;

[0017] Remove the extreme points in the first extreme point set whose contrast is less than the contrast threshold, and remove the extreme points in the first extreme point set whose curvature is less than the curvature threshold, to obtain the first key point set;

[0018] Remove the extreme points in the second set of extreme points whose contrast is less than the contrast threshold and whose curvature is less than the curvature threshold to obtain the second set of key points.

[0019] According to some embodiments, removing extreme points in the first set of extreme points whose contrast is less than a contrast threshold includes:

[0020] Obtain the absolute value of the scale space function of the Taylor expansion of any first extreme point in the first extreme point set;

[0021] The absolute value of the scale space function of the Taylor expansion is used as the contrast of any first extreme point;

[0022] If the contrast is less than the contrast threshold, remove any of the first extreme points.

[0023] According to some embodiments, removing extreme points in the first set of extreme points whose curvature is less than a curvature threshold includes:

[0024] Obtain the Hessian matrix corresponding to any first extreme point in the first set of extreme points;

[0025] Obtain the set of eigenvalues ​​of the Hessian matrix;

[0026] Based on the maximum and minimum eigenvalues ​​in the eigenvalue set, obtain the sum of the diagonal elements of the matrix corresponding to the Hessian matrix, and obtain the determinant value of the Hessian matrix.

[0027] The principal curvature of any first extreme point is obtained based on the sum of the diagonal elements of the matrix and the determinant value.

[0028] If the principal curvature is less than the principal curvature threshold, remove any of the first extreme points.

[0029] According to some embodiments, obtaining the first keypoint descriptor vector corresponding to any first keypoint in the first keypoint set includes:

[0030] A first image region with a first preset range is selected, centered on any first key point in the first set of key points.

[0031] Obtain the first gradient of any first pixel point in the first image region;

[0032] Centered on any of the first key points, a second image region within a second preset range is selected, wherein the second image region is larger than the first image region;

[0033] Obtain the second gradient of any second pixel point in the second image region;

[0034] Based on the first gradient and the second gradient, obtain the first keypoint descriptor vector corresponding to any first keypoint.

[0035] According to some embodiments, the step of using structural similarity indexing (SSIN) to perform quality checks on the first signature region and the second signature region to determine the authenticity of the signature corresponding to the first image includes:

[0036] Obtain the brightness comparison function value, contrast comparison function value, and structure comparison function value of the first signature area and the second signature area;

[0037] Based on the brightness contrast function value, the contrast contrast function value, and the structure contrast function value, obtain the SSIM function values ​​of the first signature region and the second signature region;

[0038] If the SSIM function value is less than the function value threshold, the signature authenticity of the first image is determined to be genuine.

[0039] According to a second aspect of the present disclosure, a signature quality inspection device is provided, comprising:

[0040] The image correction unit is used to correct the first image using an affine transformation matrix to obtain the second image;

[0041] The region determination unit is used to compare the template image corresponding to the form type of the first image with the second image to obtain the first signature region and the second signature region in the second image;

[0042] The signature quality inspection unit is used to perform quality inspection on the first signature region and the second signature region using structural similarity index (SSIN) to determine the authenticity of the signature corresponding to the first image.

[0043] According to a third aspect of the present disclosure, a network-side device is provided, comprising:

[0044] processor;

[0045] Memory used to store the processor's executable instructions;

[0046] The processor is configured to execute the instructions to implement the signature quality inspection method described in any one of the preceding aspects.

[0047] According to a fourth aspect of this application, a storage medium is provided that, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform the signature quality inspection method described in any of the preceding aspects.

[0048] According to a fifth aspect of this application, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in any one of the preceding aspects.

[0049] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:

[0050] In some or related embodiments, a second image is obtained by correcting the first image using an affine transformation matrix; the template image corresponding to the form type of the first image is compared with the second image to obtain the first signature region and the second signature region in the second image; structural similarity index (SSIN) is used to perform quality checks on the first signature region and the second signature region to determine the authenticity of the signature corresponding to the first image. Therefore, by judging the similarity between the first signature region and the second signature region, the cases where the first signature region and the second signature region are identical can be reduced, the cases of substitute signatures can be reduced, and the cases where the reserved signature cannot be obtained, making signature quality checks impossible to perform can also be reduced. This provides a signature quality check mechanism and improves the accuracy of signature quality checks.

[0051] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0052] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0053] Figure 1 This is a flowchart illustrating a signature quality inspection method according to an exemplary embodiment;

[0054] Figure 2 This is an example schematic diagram of an image according to an exemplary embodiment;

[0055] Figure 3 This is a flowchart illustrating a signature quality inspection method according to an exemplary embodiment;

[0056] Figure 4 This is an example schematic diagram illustrating a signature area according to an exemplary embodiment;

[0057] Figure 5 This is an example schematic diagram illustrating a signature area according to an exemplary embodiment;

[0058] Figure 6 This is an example schematic diagram illustrating a signature area according to an exemplary embodiment;

[0059] Figure 7 This is an example schematic diagram illustrating a signature area according to an exemplary embodiment;

[0060] Figure 8 This is a block diagram illustrating a signature quality inspection device according to an exemplary embodiment;

[0061] Figure 9 This is a block diagram illustrating a signature quality inspection device according to an exemplary embodiment;

[0062] Figure 10 This is a block diagram illustrating a network-side device according to an exemplary embodiment. Detailed Implementation

[0063] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0064] This disclosure provides a signature quality inspection method and apparatus. In some embodiments, the terms "signature quality inspection method" and "information processing method" and "communication method" can be used interchangeably; the terms "signature quality inspection apparatus" and "information processing apparatus" and "communication apparatus" can be used interchangeably; and the terms "information processing system" and "communication system" can be used interchangeably.

[0065] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0066] In each of the disclosed embodiments, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0067] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0068] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0069] In the embodiments disclosed herein, "multiple" refers to two or more.

[0070] In some embodiments, the terms “at least one of”, “one or more”, “a plurality of”, “multiple”, etc., may be used interchangeably.

[0071] In some embodiments, the notation "at least one of A and B", "A and / or B", "A in one case, B in another", "in response to one case A, in response to another case B", etc., may include the following technical solutions depending on the situation: in some embodiments, A (execute A regardless of B); in some embodiments, B (execute B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed); in some embodiments, A and B (both A and B are executed). The same applies when there are more branches such as A, B, C, etc.

[0072] In some embodiments, the notation "A or B" may include the following technical solutions, depending on the situation: in some embodiments, A (execution of A regardless of B); in some embodiments, B (execution of B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed). The same applies when there are more branches such as A, B, C, etc.

[0073] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0074] In some embodiments, “including A,” “containing A,” “for indicating A,” and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

[0075] In some embodiments, the terms “in response to…”, “in response to determining…”, “in the case of…”, “when…”, “if…”, “if…”, etc., can be used interchangeably.

[0076] In some embodiments, the terms “greater than,” “greater than or equal to,” “not less than,” “more than,” “more than or equal to,” “not less than,” “higher than,” “higher than or equal to,” “not lower than,” and “above” can be used interchangeably, as can the terms “less than,” “less than or equal to,” “not greater than,” “less than,” “less than or equal to,” “not more than,” “lower than,” “lower than or equal to,” “not higher than,” and “below”.

[0077] In some embodiments, the apparatus and device may be interpreted as physical or virtual, and their names are not limited to the names recorded in the embodiments. In some cases, they may also be understood as "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "body", etc.

[0078] In some embodiments, "network" can be interpreted as devices included in the network, such as access network devices, core network devices, etc.

[0079] In some embodiments, "terminal" or "terminal device" may be referred to as "user equipment (UE)," "user terminal," "mobile station (MS)," "mobile terminal (MT)," "subscriber station," "mobile unit," "subscriber unit," "wireless unit," "remote unit," "mobile device," "wireless device," "wireless communication device," "remote device," "mobile subscriber station," "access terminal," "mobile terminal," "wireless terminal," "remote terminal," "handset," "user agent," "mobile client," "client," etc.

[0080] In some embodiments, the acquisition of data, information, etc., may comply with the laws and regulations of the country where the location is situated.

[0081] In some embodiments, data, information, etc., may be obtained with the user's consent.

[0082] Furthermore, each element, each row, or each column in the table of this disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0083] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0084] Figure 1 This is a flowchart illustrating a signature quality inspection method according to an exemplary embodiment, such as... Figure 1 As shown, this signature quality inspection method can be used in communication scenarios and includes the following steps:

[0085] In step S11, the first image is corrected using an affine transformation matrix to obtain the second image;

[0086] According to some embodiments, the execution entity of this disclosure may be a network-side device, specifically a server. The server may be a single server, or it may be a server cluster consisting of multiple servers.

[0087] According to some embodiments, the affine transformation matrix can be, for example, a matrix used to perform spatial transformations, and this affine transformation matrix is ​​not specifically a fixed matrix. For example, when a certain number in the affine transformation matrix changes, the affine transformation matrix can also change accordingly.

[0088] In some embodiments, the first image refers to the image to be processed. This first image may be, for example, a photograph taken directly by a terminal device of a paper document and sent to the network-side device, or it may be pre-stored in the network-side device. The technical solutions of this disclosure do not limit the acquisition of the first image. The "first" in the first image is only used to distinguish it from the "second" and does not specifically refer to a fixed image. For example, when the acquisition time point corresponding to the first image changes, the first image may also change accordingly. For example, when the image content corresponding to the first image changes, the first image may also change accordingly.

[0089] According to some embodiments, the second image refers to the image obtained after correcting the first image. The second image does not specifically refer to a fixed image. For example, when the first image changes, the second image may also change accordingly. For example, when the affine transformation matrix changes, the second image may also change accordingly.

[0090] According to some embodiments, when a first image is acquired, an affine transformation matrix can be used to correct the first image to obtain a second image. For example, when an image to be inspected is acquired, an affine transformation matrix can be used to correct the image to be inspected to obtain a second image.

[0091] In step S12, the template image corresponding to the form type of the first image and the second image are compared to obtain the first signature area and the second signature area in the second image;

[0092] According to some embodiments, the form type is used to indicate the type of the form in the first image. This form type does not specifically refer to a fixed type. For example, when the content of the first image changes, the form type of the first image can also change accordingly. The first image, for example, could be an image obtained after photographing a paper-based installation and maintenance service confirmation form following the installation and debugging of various communication devices. Since the form content and style of the server confirmation form corresponding to each business type are different, the form type of the first image corresponding to each business type is also different.

[0093] In some embodiments, for example, the form type of the first image may also change accordingly when the business type changes.

[0094] In some embodiments, the template image may be, for example, a pre-stored installation and maintenance service confirmation image. This template image does not specifically refer to any particular image. For example, when the content of the image corresponding to the template image changes, the template image may also change accordingly.

[0095] In some embodiments, the first signature area refers to any area in the first image where the signature is located. The "first" in the first signature area is only used to distinguish it from the second signature area and does not specifically refer to a fixed area. The second signature area refers to another area in the first image besides the first signature area where the signature is located. In other words, the first signature area and the second signature area are only used to indicate different signature areas.

[0096] According to some embodiments, when the first image is obtained, an affine transformation matrix can be used to correct the first image to obtain a second image. The template image corresponding to the form type of the first image and the second image are compared to obtain the first signature region and the second signature region in the second image. For example, the first signature region and the second signature region in the template image can be obtained in advance, and the first signature region and the second signature region in the second image can be obtained based on image similarity or position comparison.

[0097] The second image includes areas not limited to the first and second signature areas; for example, it may also include a form name area, a service information area, a service duration area, etc. An example illustration of the second image could be as follows: Figure 2 As shown.

[0098] In step S13, structural similarity index (SSIN) is used to perform quality checks on the first signature region and the second signature region to determine the authenticity of the signature corresponding to the first image.

[0099] According to some embodiments, Structural Similarity (SSIM) is an indicator for measuring the similarity between two images. When the first signature region and the second signature region are obtained, SSIM can be used to perform quality checks on the first and second signature regions to determine the authenticity of the signature corresponding to the first image.

[0100] For example, when the similarity between the first signature region and the second signature region is greater than a similarity threshold, the authenticity of the signature corresponding to the first image can be determined.

[0101] In some or related embodiments, a second image is obtained by correcting the first image using an affine transformation matrix; the template image corresponding to the form type of the first image is compared with the second image to obtain the first signature region and the second signature region in the second image; structural similarity index (SSIN) is used to perform quality checks on the first signature region and the second signature region to determine the authenticity of the signature corresponding to the first image. Therefore, by judging the similarity between the first signature region and the second signature region, the cases where the first signature region and the second signature region are identical can be reduced, as can cases of substitute signatures. It can also reduce the situation where the pre-reserved signature cannot be obtained, making signature quality checks impossible. This provides a signature quality check mechanism and improves the accuracy of signature quality checks. Furthermore, this signature quality check mechanism requires no manual intervention, reducing labor costs and increasing the intelligence and information efficiency of signature quality checks.

[0102] Figure 3 This is a flowchart illustrating a signature quality inspection method according to an exemplary embodiment, such as... Figure 3 As shown, this signature quality inspection method can be used in communication scenarios and includes the following steps:

[0103] In step S21, the first image is corrected using an affine transformation matrix to obtain the second image;

[0104] The specific process is as described above and will not be repeated here.

[0105] According to some embodiments, the affine transformation matrix (the 3x3 matrix in Formula 1 can be, for example, an affine transformation matrix) is a commonly used transformation algorithm in image processing. In some cases, images may not be well aligned for various reasons. To ensure the authenticity verification and archiving of subsequent signatures, affine transformations are needed to correct the images. The correction formula is as follows:

[0106]

[0107] In the formula, (tx, ty) represents the translation amount, while the parameter ai reflects changes such as image rotation and scaling. Using the coordinate data of matching feature points on the first image and the second image, multiple sets of equations can be written. Calculating the parameters tx, ty, and ai (i = 1~4) yields the affine transformation matrix, thus obtaining the coordinate transformation relationship between the two images. The affine transformation matrix can, for example, be determined from pre-prepared sample images.

[0108] In some embodiments, for example, when multiple images are acquired, the multiple images can be identified to obtain a first image from the multiple images. For example, form recognition can be performed on multiple images, and if a form exists in an image, that image is selected as the first image.

[0109] According to some embodiments, the method further includes: obtaining a first set of key points for a first image and a second set of key points for any template image in a set of template images; obtaining a first key point descriptor vector corresponding to any first key point in the first set of key points, and obtaining a second key point descriptor vector corresponding to any second key point in the second set of key points; normalizing the first key point descriptor vector corresponding to any first key point to obtain a first SIFT feature vector corresponding to the first image; normalizing the second key point descriptor vector corresponding to any second key point to obtain a second SIFT feature vector corresponding to any template image; obtaining the similarity between the first image and any template image based on the Euclidean distance between the first SIFT feature vector and the second SIFT feature vector; and using the form type of the template image with the highest similarity as the form type of the first image. Therefore, determining the form type based on the SIFT feature vector can improve the accuracy of form type acquisition and the accuracy of signature quality inspection.

[0110] The form type of the first image is the same as that of the second image.

[0111] According to some embodiments, obtaining a first set of key points for a first image and a second set of key points for any template image in the template image set includes: obtaining a first set of extreme points for the first image, and obtaining a second set of extreme points for any template image in the template image set; removing extreme points in the first set of extreme points whose contrast is less than a contrast threshold, and removing extreme points in the first set of extreme points whose curvature is less than a curvature threshold, to obtain a first set of key points; and removing extreme points in the second set of extreme points whose contrast and curvature are both less than a contrast threshold, to obtain a second set of key points. Therefore, determining the second set of key points based on extreme points can improve the accuracy of key point set acquisition and improve the accuracy of signature quality inspection.

[0112] According to some embodiments, removing extreme points in the first set of extreme points whose contrast is less than a contrast threshold includes: obtaining the absolute value of the scale space function of the Taylor expansion of any first extreme point in the first set of extreme points; using the absolute value of the scale space function of the Taylor expansion as the contrast of the any first extreme point; and removing the any first extreme point if the contrast is less than the contrast threshold. Therefore, the absolute value of the scale space function of the Taylor expansion can be used to remove extreme points with low contrast, thus improving the accuracy of extreme point acquisition.

[0113] In some embodiments, a scale space can be constructed, for example, to detect the extreme points of the image to be detected (the first image) and the template image separately. The purpose of constructing the scale space is to simulate the multi-scale features of the image data. The scale space of a two-dimensional image is defined as follows:

[0114] L(x,y,σ)=G(x,y,σ)*I(x,y) (2)

[0115]

[0116] In the formula: (x, y) are spatial coordinates, which are scale coordinates; the magnitude of σ determines the smoothness of the image; I(x, y) are the position parameters of the original image.

[0117] To effectively detect stable keypoints in scale space, a Gaussian difference scale space is introduced. This scale space is generated by convolving the image with Gaussian difference kernels of different scales, and is defined as follows:

[0118] D(x,y,σ)=(G(x,y,kσ)-G(x,y,σ))*I(x,y)=L(x,y,kσ)-L(x,y,σ) (4)

[0119] In the formula: k is the scaling factor between two adjacent layers in scale space, L(x, y, σ) is the image after Gaussian blurring, G(x, y, σ) is the Gaussian kernel function, and D(x, y, σ) is the Different-of-Gaussian (DoG) image.

[0120] According to some embodiments, local extrema are the extreme points of the DoG image. Local extrema are determined by comparing each pixel in the image with all its neighbors to see if the pixel is larger or smaller than its neighbors in the image domain. If the above condition is met, it is considered a local extrema. A three-dimensional quadratic function is fitted to the local extrema to accurately determine the location and scale of the feature points. The absolute value of the scale space function obtained from the Taylor expansion is the contrast. For example, feature points with a contrast less than 0.03 are discarded. The Taylor expansion formula for the scale space function D(x, y, σ) is as follows:

[0121]

[0122] By setting the partial derivative of equation (5) with respect to x to 0, we can obtain the location of the limit point.

[0123]

[0124] Substituting formula (6) into formula (5), we get:

[0125]

[0126] If the absolute value function The feature point is retained if it is not, otherwise it is discarded.

[0127] According to some embodiments, removing extreme points in the first set of extreme points whose curvature is less than a curvature threshold includes: obtaining the Hessian matrix corresponding to any first extreme point in the first set of extreme points; obtaining the eigenvalue set of the Hessian matrix; obtaining the sum of the diagonal elements of the Hessian matrix based on the maximum and minimum eigenvalues ​​in the eigenvalue set, and obtaining the determinant value of the Hessian matrix; obtaining the principal curvature of the any first extreme point based on the sum of the diagonal elements and the determinant value; and removing the any first extreme point if the principal curvature is less than a principal curvature threshold. Therefore, removing edge points based on the principal curvature can improve the accuracy of extreme point acquisition.

[0128] In some implementations, the extrema of a poorly defined difference-of-Gaussian operator may have larger principal curvatures across edges and smaller principal curvatures at vertical edges. The principal curvatures are derived from the Hessian matrix.

[0129]

[0130] The principal curvature of D is proportional to the eigenvalues ​​of H. Let α be the largest eigenvalue and β be the smallest eigenvalue, then...

[0131] Tr(H)=D xx +D yy =α+β (9)

[0132] Det(H) = D xx D yy -(D xy ) 2 =αβ (10)

[0133] Let α = αβ, then:

[0134]

[0135] If the curvature is less than (r+1)2 / r, retain the feature point; otherwise, discard it.

[0136] According to some embodiments, obtaining the first keypoint descriptor vector corresponding to any first keypoint in the first keypoint set includes: selecting a first image region of a first preset range centered on any first keypoint in the first keypoint set; obtaining a first gradient of any first pixel in the first image region; selecting a second image region of a second preset range centered on the any first keypoint, wherein the second image region is larger than the first image region; obtaining a second gradient of any second pixel in the second image region; and obtaining the first keypoint descriptor vector corresponding to the any first keypoint based on the first gradient and the second gradient. Therefore, obtaining the keypoint descriptor vector based on the first gradient and the second gradient can improve the accuracy of vector acquisition and the accuracy of signature quality inspection.

[0137] In some embodiments, the first preset range may be, for example, an 8*8 field window, and the second preset range may be, for example, a 16*16 field window.

[0138] The gradient direction distribution characteristics of the neighboring pixels of a keypoint are used to specify the direction parameters of the keypoint. The formulas for the magnitude and direction of the gradient at any pixel (x, y) are as follows:

[0139]

[0140] θ(x,y)=αtan2((L(x,y+1)-L(x,y-1)) / (L(x+1,y)-L(x-1,y))) (13)

[0141] In the formula: L(x, y) represents the coordinates of the feature point, the scale used is the scale of the pixel, and a is the harmonic factor, set to 1. The coordinate axes are rotated to the direction of the keypoint to ensure rotation invariance. An 8x8 neighborhood window is selected around the keypoint, and the magnitude and direction of the gradient of all pixels within it are calculated. A Gaussian window is used to weight these gradients; the closer a pixel is to the keypoint, the greater its gradient contribution. This forms a gradient histogram. In the gradient histogram, there are 8 bars, each representing a 45-degree angle, representing 8 directions. By dividing the image region around the keypoint into blocks, the gradient histogram within each block is calculated, generating a unique vector. This vector is an abstraction of the image information of that region and is unique.

[0142] For example, centered on the keypoint, the gradient of each pixel within a 16x16 window around the keypoint can be calculated, and a Gaussian descent function can be used to reduce the weights far from the center. Within each 4x4 1 / 16-inch small window, a gradient direction histogram with 8 directions is formed using the same method as the previous step. This generates a 4*4*8 = 128-dimensional descriptor for each keypoint. Each keypoint descriptor contains three pieces of information: position, scale, and orientation. The second step generates the position and scale of the descriptor, and the third step provides the orientation.

[0143] According to some embodiments, the 128-dimensional keypoint descriptor vectors are normalized to obtain the SIFT feature vectors of the image. After the SIFT feature vectors of two images are generated, the Euclidean distance between the keypoint feature vectors is used as a similarity metric for the keypoints in the two images. A priority kd-tree search is used to find the approximate nearest neighbor feature point for each feature point. If the nearest neighbor distance divided by the second nearest neighbor distance is less than 0.6, the pair of matching points is accepted. The template image form type that matches the most SIFT feature points of the image to be detected is selected from the template images as the form type of the image to be detected.

[0144] In step S22, the template image corresponding to the form type of the first image and the second image are compared to obtain the first signature area and the second signature area in the second image;

[0145] The specific process is as described above and will not be repeated here.

[0146] According to some embodiments, the image to be detected may be, for example, a first image. While determining the form category of the image to be detected, it is also necessary to know the positions of the "Customer Contact" and "Customer Signature" information fields within that form. Simultaneously with form type matching, a correspondence between the image to be detected and the corresponding template image regarding SIFT feature points has been established: y = F(x), where x is the image to be detected, y is the corresponding template image, and F is the mapping function between x and y. Based on this correspondence, the information fields for "Customer Contact" and "Customer Signature" in the image to be detected, corresponding to those in the template image, can be calculated. The information fields for "Customer Contact" and "Customer Signature" in the second image are then calculated using the SIFT mapping function and cropped. An example diagram of the first signature area is shown below. Figure 4 As shown; an example diagram of the second signature area is shown below. Figure 5 As shown.

[0147] In step S23, the brightness comparison function value, contrast comparison function value, and structure comparison function value of the first signature area and the second signature area are obtained;

[0148] The specific process is as described above and will not be repeated here.

[0149] According to some embodiments, structural similarity (SSIM) is a technique that reflects the human visual system's judgment of the similarity between two images. It detects whether structural information has changed to perceive approximate information about image distortion and measures the similarity between two images. SSIM consists of three comparison modules: brightness, contrast, and structure. Therefore, the brightness comparison function value, contrast comparison function value, and structural comparison function value of the first signature region and the second signature region can be obtained.

[0150] According to some embodiments, the first signature area and the second signature area can be preprocessed, such as scaled and cropped, to obtain two images of the same size. An example illustration of the processed image of the first signature area is shown below. Figure 5 As shown; an example diagram of the image after processing the second signature region is shown below. Figure 6 As shown.

[0151] In some embodiments, the average gray level of the image is used as an estimate of the brightness measurement:

[0152]

[0153] The brightness comparison function between the first signature area and the second signature area is as follows:

[0154]

[0155] According to some embodiments, the standard deviation of an image can be used as an estimate of contrast measurement.

[0156]

[0157] Contrast comparison function between the first signature area and the second signature area

[0158]

[0159] In some embodiments, the structural contrast function value is calculated:

[0160]

[0161] In step S24, the SSIM function values ​​of the first signature region and the second signature region are obtained based on the brightness contrast function value, the contrast contrast function value, and the structure contrast function value.

[0162] The specific process is as described above and will not be repeated here.

[0163] According to some embodiments, the SSIM function values ​​of the first signature region and the second signature region are obtained based on the brightness contrast function value, the contrast contrast function value, and the structure contrast function value.

[0164] For example, three comparison functions can be combined to form an SSIM function:

[0165] SSIM(x,y)=f(l(x,y),c(x,y),s(x,y))=[l(x,y)] α [c(x,y)] β [s(x,y)] γ (19)

[0166] In the formula, α, β, γ > 0, which are used to adjust the importance of these three modules. Assuming α, β, and γ are all 1, C3 = C2 / 2, then...

[0167]

[0168] In step S25, if the SSIM function value is less than the function value threshold, the signature authenticity of the first image is determined to be genuine.

[0169] The specific process is as described above and will not be repeated here.

[0170] According to some embodiments, the function value threshold can be, for example, 0.65. When the value of the SSIM function is greater than or equal to 0.65, it is determined that the similarity between the first signature region and the second signature region is too high, and the quality inspection fails; when the value of the SSIM function is less than 0.65, it is determined that the similarity between the first signature region and the second signature region is too low, and the quality inspection passes.

[0171] According to some embodiments, after determining the authenticity of the signature in the first image, a prompt message can be issued. For example, when the value of the SSIM function is less than 0.65, it is determined that the similarity between the first signature region and the second signature region is low, the quality inspection passes, and the prompt message issued can be, for example, as follows: Figure 7 As shown.

[0172] In some or related embodiments, by obtaining the brightness contrast function value, contrast contrast function value, and structure contrast function value of the first signature region and the second signature region, and based on the brightness contrast function value, contrast contrast function value, and structure contrast function value, the SSIM function value of the first signature region and the second signature region is obtained; if the SSIM function value is less than a function value threshold, the signature authenticity of the first image is determined to be genuine. Therefore, the authenticity of the signature can be determined by brightness, contrast, and structure, reducing the number of situations where signature quality inspection cannot be performed and improving the accuracy of signature quality inspection.

[0173] Figure 8 This is a block diagram illustrating a signature quality inspection device according to an exemplary embodiment. (Refer to...) Figure 8 The signature quality inspection device 800 includes:

[0174] Image correction unit 801 is used to correct the first image using an affine transformation matrix to obtain the second image;

[0175] The region determination unit 802 is used to compare the template image corresponding to the form type of the first image with the second image to obtain the first signature region and the second signature region in the second image;

[0176] The signature quality inspection unit 803 is used to perform quality inspection on the first signature area and the second signature area using structural similarity (SSIN) to determine the authenticity of the signature corresponding to the first image.

[0177] According to some embodiments, Figure 9 This is a block diagram illustrating a signature quality inspection device according to an exemplary embodiment. (Refer to...) Figure 9 The signature quality inspection device 800 further includes a type acquisition unit 804, which is used to compare the template image corresponding to the form type of the first image with the second image, and to obtain the first signature area and the second signature area in the second image, and is also specifically used for:

[0178] Obtain the first set of key points of the first image and the second set of key points of any template image in the template image set;

[0179] Obtain the first key point descriptor vector corresponding to any first key point in the first key point set, and obtain the second key point descriptor vector corresponding to any second key point in the second key point set;

[0180] Normalize the first key point descriptor vector corresponding to any first key point to obtain the first SIFT feature vector corresponding to the first image.

[0181] Normalize the second key point descriptor vector corresponding to any second key point to obtain the second SIFT feature vector corresponding to any template image;

[0182] The similarity between the first image and any template image is obtained based on the Euclidean distance between the first SIFT feature vector and the second SIFT feature vector.

[0183] The form type of the template image with the highest similarity is used as the form type of the first image.

[0184] According to some embodiments, when the type acquisition unit 804 is used to acquire the first key point set of the first image and the second key point set of any template image in the template image set, it is specifically used for:

[0185] Obtain the first set of extreme points of the first image, and obtain the second set of extreme points of any template image in the template image set;

[0186] Remove the extreme points in the first extreme point set whose contrast is less than the contrast threshold, and remove the extreme points in the first extreme point set whose curvature is less than the curvature threshold, to obtain the first key point set;

[0187] Remove the extreme points in the second set of extreme points whose contrast is less than the contrast threshold and whose curvature is less than the curvature threshold to obtain the second set of key points.

[0188] According to some embodiments, when the type acquisition unit 804 is used to remove extreme points in the first extreme point set whose contrast is less than the contrast threshold, it is specifically used for:

[0189] Obtain the absolute value of the scale space function of the Taylor expansion of any first extreme point in the first extreme point set;

[0190] The absolute value of the scale space function of the Taylor expansion is used as the contrast of any first extreme point;

[0191] If the contrast is less than the contrast threshold, remove any of the first extreme points.

[0192] According to some embodiments, when the type acquisition unit 804 is used to remove extreme points in the first set of extreme points whose curvature is less than a curvature threshold, it is specifically used for:

[0193] Obtain the Hessian matrix corresponding to any first extreme point in the first set of extreme points;

[0194] Obtain the set of eigenvalues ​​of the Hessian matrix;

[0195] Based on the maximum and minimum eigenvalues ​​in the eigenvalue set, obtain the sum of the diagonal elements of the matrix corresponding to the Hessian matrix, and obtain the determinant value of the Hessian matrix.

[0196] The principal curvature of any first extreme point is obtained based on the sum of the diagonal elements of the matrix and the determinant value.

[0197] If the principal curvature is less than the principal curvature threshold, remove any of the first extreme points.

[0198] According to some embodiments, when the type acquisition unit 804 is used to acquire the first keypoint descriptor vector corresponding to any first keypoint in the first keypoint set, it is specifically used for:

[0199] A first image region with a first preset range is selected, centered on any first key point in the first set of key points.

[0200] Obtain the first gradient of any first pixel point in the first image region;

[0201] Centered on any of the first key points, a second image region within a second preset range is selected, wherein the second image region is larger than the first image region;

[0202] Obtain the second gradient of any second pixel point in the second image region;

[0203] Based on the first gradient and the second gradient, obtain the first keypoint descriptor vector corresponding to any first keypoint.

[0204] According to some embodiments, the signature quality inspection unit 803 is used to perform quality inspection on the first signature region and the second signature region using structural similarity index (SSIN) to determine the authenticity of the signature corresponding to the first image, specifically for:

[0205] Obtain the brightness comparison function value, contrast comparison function value, and structure comparison function value of the first signature area and the second signature area;

[0206] Based on the brightness contrast function value, the contrast contrast function value, and the structure contrast function value, obtain the SSIM function values ​​of the first signature region and the second signature region;

[0207] If the SSIM function value is less than the function value threshold, the signature authenticity of the first image is determined to be genuine.

[0208] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0209] In some or related embodiments, an image correction unit is used to correct the first image using an affine transformation matrix to obtain a second image; a region determination unit is used to compare the template image corresponding to the form type of the first image with the second image to obtain a first signature region and a second signature region in the second image; a signature quality inspection unit is used to perform quality inspection on the first signature region and the second signature region using structural similarity index (SSIN) to determine the authenticity of the signature corresponding to the first image. Therefore, by judging the similarity between the first signature region and the second signature region, the cases where the first signature region and the second signature region are identical can be reduced, the cases of substitute signatures can be reduced, a signature quality inspection mechanism can be provided, and the accuracy of signature quality inspection can be improved.

[0210] Figure 10 This is a block diagram of a network-side device 1000 provided in an embodiment of this disclosure. For example, the network-side device 1000 can be provided as a network-side device. (Refer to...) Figure 10 The network-side device 1000 includes a processing component 1022, which further includes at least one processor, and memory resources represented by memory 1032 for storing instructions, such as application programs, that can be executed by the processing component 1022. The application programs stored in memory 1032 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1022 is configured to execute instructions to perform any of the methods described above applied to the network-side device.

[0211] The network-side device 1000 may also include a power supply component 1027 configured to perform power management of the network-side device 1000, a wired or wireless network interface 1050 configured to connect the network-side device 1000 to a network, and an input / output (I / O) interface 1058. The network-side device 1000 may operate on an operating system stored in memory 1032, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0212] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0213] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0214] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0215] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0216] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0217] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0218] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0219] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A signature quality inspection method, characterized in that, include: The first image is corrected using an affine transformation matrix to obtain the second image; The template image corresponding to the form type of the first image and the second image are compared to obtain the first signature area and the second signature area in the second image; The Structural Similarity SSIM is used to perform quality checks on the first signature region and the second signature region to determine the authenticity of the signature corresponding to the first image. This includes determining that the signature of the first image is authentic if the SSIM function value of the first signature region and the second signature region is less than the function value threshold. The process of determining the template image with the highest similarity to the first image, using its form type as the form type of the first image, includes: A set of keypoints, keypoint descriptor vectors, and SIFT feature vectors are determined for a first image. The set of keypoints, keypoint descriptor vectors, and SIFT feature vectors for any template image in the template image set are also determined. Specifically, extreme points with contrast less than a contrast threshold and curvature less than a curvature threshold are removed from the set of extreme points in the image to obtain the set of keypoints for the image. For any keypoint, its keypoint descriptor vector is determined based on the first gradient of a first pixel within a first image region centered on it and the second gradient of a second pixel within a second image region centered on it, where the second image region is larger than the first image region. The keypoint descriptor vectors are then normalized to obtain the SIFT feature vector. The similarity between the first image and any template image is determined based on the SIFT feature vectors of the first image and any template image.

2. The method according to claim 1, characterized in that, The determination of the template image with the highest similarity to the first image specifically includes: Obtain the first set of key points of the first image and the second set of key points of any template image in the template image set; Obtain the first key point descriptor vector corresponding to any first key point in the first key point set, and obtain the second key point descriptor vector corresponding to any second key point in the second key point set; Normalize the first key point descriptor vector corresponding to any first key point to obtain the first SIFT feature vector corresponding to the first image. Normalize the second key point descriptor vector corresponding to any second key point to obtain the second SIFT feature vector corresponding to any template image; The similarity between the first image and any template image is obtained based on the Euclidean distance between the first SIFT feature vector and the second SIFT feature vector. The form type of the template image with the highest similarity is used as the form type of the first image.

3. The method according to claim 2, characterized in that, The acquisition of the first key point set of the first image and the second key point set of any template image in the template image set includes: Obtain the first set of extreme points of the first image, and obtain the second set of extreme points of any template image in the template image set; Remove the extreme points in the first extreme point set whose contrast is less than the contrast threshold, and remove the extreme points in the first extreme point set whose curvature is less than the curvature threshold, to obtain the first key point set; Remove the extreme points in the second set of extreme points whose contrast is less than the contrast threshold and whose curvature is less than the curvature threshold to obtain the second set of key points.

4. The method according to claim 3, characterized in that, Removing extreme points from the first set of extreme points whose contrast is less than a contrast threshold includes: Obtain the absolute value of the scale space function of the Taylor expansion of any first extreme point in the first extreme point set; The absolute value of the scale space function of the Taylor expansion is used as the contrast of any first extreme point; If the contrast is less than the contrast threshold, remove any of the first extreme points.

5. The method according to claim 3, characterized in that, The removal of extreme points in the first set of extreme points whose curvature is less than a curvature threshold includes: Obtain the Hessian matrix corresponding to any first extreme point in the first set of extreme points; Obtain the set of eigenvalues ​​of the Hessian matrix; Based on the maximum and minimum eigenvalues ​​in the eigenvalue set, obtain the sum of the diagonal elements of the matrix corresponding to the Hessian matrix, and obtain the determinant value of the Hessian matrix. The principal curvature of any first extreme point is obtained based on the sum of the diagonal elements of the matrix and the determinant value. If the principal curvature is less than the principal curvature threshold, remove any of the first extreme points.

6. The method according to claim 2, characterized in that, Obtain the first keypoint descriptor vector corresponding to any first keypoint in the first keypoint set, including: A first image region with a first preset range is selected, centered on any first key point in the first set of key points. Obtain the first gradient of any first pixel point in the first image region; Centered on any of the first key points, a second image region within a second preset range is selected, wherein the second image region is larger than the first image region; Obtain the second gradient of any second pixel point in the second image region; Based on the first gradient and the second gradient, obtain the first keypoint descriptor vector corresponding to any first keypoint.

7. The method according to claim 1, characterized in that, The step of using structural similarity indexing (SSIN) to perform quality checks on the first signature region and the second signature region to determine the authenticity of the signature corresponding to the first image includes: Obtain the brightness comparison function value, contrast comparison function value, and structure comparison function value of the first signature area and the second signature area; Based on the brightness contrast function value, the contrast contrast function value, and the structure contrast function value, obtain the SSIM function values ​​of the first signature region and the second signature region; If the SSIM function value is less than the function value threshold, the signature authenticity of the first image is determined to be genuine.

8. A signature quality inspection device, characterized in that, include: The image correction unit is used to correct the first image using an affine transformation matrix to obtain the second image; The region determination unit is used to compare the template image corresponding to the form type of the first image with the second image to obtain the first signature region and the second signature region in the second image; The signature quality inspection unit is used to perform quality inspection on the first signature region and the second signature region using structural similarity (SSIM) to determine the authenticity of the signature corresponding to the first image, including: if the SSIM function value of the first signature region and the second signature region is less than the function value threshold, the signature authenticity of the first image is determined to be authentic. The process of determining the template image with the highest similarity to the first image, using its form type as the form type of the first image, includes: A set of keypoints, keypoint descriptor vectors, and SIFT feature vectors are determined for a first image. The set of keypoints, keypoint descriptor vectors, and SIFT feature vectors for any template image in the template image set are also determined. Specifically, extreme points with contrast less than a contrast threshold and curvature less than a curvature threshold are removed from the set of extreme points in the image to obtain the set of keypoints for the image. For any keypoint, its keypoint descriptor vector is determined based on the first gradient of a first pixel within a first image region centered on it and the second gradient of a second pixel within a second image region centered on it, where the second image region is larger than the first image region. The keypoint descriptor vectors are then normalized to obtain the SIFT feature vector. The similarity between the first image and any template image is determined based on the SIFT feature vectors of the first image and any template image.

9. A network-side device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the signature quality inspection method as described in any one of claims 1 to 7.

10. A storage medium, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the signature quality inspection method as described in any one of claims 1 to 7.

Images