An electronic signature information collection device for electronic signature

Abstract

An embodiment of the present invention provides a signature information acquisition device for electronic signature, including: an electronic signature input module for receiving electronic signature data input by a first user, where the electronic signature data includes a handwritten signature and an image-based signature; a storage module for storing the handwritten feature parameters of a second user or the three-dimensional contour data and texture detail data of a physical signature, where the second user is the user represented by the electronic signature; an electronic signature consistency determination module for determining the consistency with the features represented by the electronic signature data according to the handwritten feature parameters of the second user or the three-dimensional contour data and texture detail data of the physical signature; and a communication module for transmitting the electronic signature data to a target device after the signature passes the consistency comparison. By implementing the present invention, the security of electronic signature can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of information security, and particularly relates to a signature information acquisition device for electronic signature. Background Art

[0002] In commercial activities and various document signing scenarios, signatures, as an important means to confirm the validity of documents and clarify the attribution of responsibilities, have always played a key role. Traditional signature methods mainly rely on the affixing of physical seals and handwritten signatures. With the rapid development of digital technology, this traditional method has been unable to keep up with the pace of development. Therefore, all industries are actively promoting digital transformation to improve work efficiency, reduce costs, and enhance competitiveness.

[0003] Against this backdrop, the electronic signature has become an inevitable trend. It can convert traditional physical signatures and handwritten signatures into digital forms, enabling online signing and management. However, the current process of electronic signature faces security challenges. Currently, the electronic signature also faces severe challenges at the security level. Insufficient information is provided for subsequent security verification during the process of electronic signature, increasing the risk of forging electronic signatures. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a signature information acquisition device for electronic signature to meet the need for improving the security of electronic signature.

[0005] To achieve the above objective, the present invention provides the following technical solutions:

[0006] The present invention provides a signature information acquisition device for electronic signature, including: an electronic signature input module for receiving electronic signature data input by a first user, where the electronic signature data includes a handwritten signature and an image-based signature; a storage module for storing the handwritten feature parameters of a second user or the three-dimensional contour data and texture detail data of a physical seal, where the second user is the user represented by the electronic signature; an electronic signature consistency determination module for determining the consistency with the features represented by the electronic signature data based on the handwritten feature parameters of the second user or the three-dimensional contour data and texture detail data of the physical seal; and a communication module for transmitting the electronic signature data to a target device after the signature passes the consistency comparison.

[0007] Optionally, a signature information acquisition device for electronic signature further includes: a biological information acquisition module for acquiring the biological information of the first user who inputs the electronic signature data; a first encryption module for encrypting and encoding the electronic signature data and the biological data of the first user to obtain target electronic signature data; where the communication module is used to transmit the target electronic signature data to the target device when the signature passes the consistency comparison.

[0008] Optionally, the electronic signature input module includes: an automatic light supplementing module for automatically supplementing light to the scanning area according to the current light; a scanning module for scanning the scanning area to obtain a signature image; and an automatic distortion correction module for automatically correcting the distortion of the signature image.

[0009] Optionally, a signature information collection device for electronic signature includes an electronic signature input module, which includes: an electronic signature module for receiving the user's handwritten signature data; a pressure sensor array for collecting the pressure of different areas when the user writes a signature; a timer connected to the pressure sensor array for recording the time when the pressure is generated in different areas when the user writes a signature; and a second encryption module for extracting handwritten feature parameters from the pressure sensor array and timer data when the user writes a signature data, preprocessing the handwritten feature parameters, and then asymmetrically encrypting the handwritten signature data according to the preprocessed handwritten feature parameters to obtain the encrypted user electronic signature.

[0010] Optionally, the electronic signature is obtained by converting a physical seal. To determine the consistency with the features represented by the electronic signature data according to the three-dimensional contour data and texture detail data of the physical signature, it includes: constructing a three-dimensional signature digital model according to the three-dimensional contour data and texture detail data of the physical signature; determining multiple target projection angles based on the geometric characteristics of the three-dimensional signature digital model; projecting the signature part of the three-dimensional signature digital model onto a two-dimensional plane at multiple target projection angles according to the projection algorithm of computer graphics to obtain multiple two-dimensional projections; performing feature pooling on the multiple two-dimensional projections and fusing the features of the multiple two-dimensional projections to obtain the fused features of the signature part in the three-dimensional signature digital model; extracting multiple types of feature data in the electronic signature according to the target feature extraction algorithm, including contour features and texture semantic features; determining the contour features in the fused features, and determining the consistency between the contour features in the electronic signature and the corresponding contour features in the fused features according to the first target algorithm; determining the texture semantic features in the fused features, and determining the consistency between the texture semantic features in the electronic signature and the corresponding texture semantic features in the fused features according to the second target algorithm; and determining the consistency of the physical signature and the electronic signature features according to the consistency between the contour features and the consistency between the texture semantic features.

[0011] Optionally, the electronic signature is an encrypted user electronic signature, and preprocessing is performed on the handwritten feature parameters, including: querying the set of numerical ranges where each handwritten feature parameter is located in the mapping table, and using the number of each set of numerical ranges as the preprocessed handwritten feature parameter; determining the consistency with the features represented by the electronic signature data according to the handwritten feature parameters of the second user, including: querying the set of numerical ranges where the handwritten feature parameters of the second user are located in the mapping table, using the number of each set of numerical ranges as the decryption key element, and performing decryption using the asymmetric decryption method. When the decryption is successful, it indicates that the features represented by the electronic signature data are consistent with the handwritten feature parameters of the second user; when the decryption is unsuccessful, it indicates that the features represented by the electronic signature data are inconsistent with the handwritten feature parameters of the second user.

[0012] Optionally, a signature information collection device for signature digitization further includes: when the features represented by the electronic signature data are consistent with the handwritten feature parameters of the second user, adding the second user number to the electronic signature data, and sending the electronic signature data with the second user number added to the target device.

[0013] Optionally, a signature information collection device for signature digitization further includes: a split signature recognition module for recognizing whether the electronic signature data entered by the first user is a split signature; a split signature optimization module for optimizing the split signature image when it is recognized as a split signature.

[0014] Optionally, a signature information collection device for signature digitization further includes: an electromagnetic shielding structure provided on the outer surface layer of the main body of the signature information collection device for signature digitization, including alternately arranged conductive layers and magnetic conductive layers, for blocking or attenuating external electromagnetic waves.

[0015] The embodiment of the present invention provides a signature information collection device for signature digitization. In terms of the data collection link, it ensures comprehensive and accurate collection of signature features, laying a solid foundation for subsequent security verification. In terms of data processing and verification, it can determine the authenticity of the entered signature by judging the consistency between the entered signature and the pre-stored valid signature, plugging loopholes and preventing criminals from forging signatures or electronically forging signatures. In terms of data storage and transmission, a secure and reliable encryption verification mechanism is adopted to protect the electronic signature data from being tampered with, and to prevent the data from being intercepted and damaged by a man-in-the-middle during transmission, improving the security of signature digitization.

[0016] Other advantages, objectives and features of the present invention will be described in the following specification, and to some extent, they are obvious to those skilled in the art, or those skilled in the art can obtain guidance from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To make the objectives, technical solutions, and beneficial effects of the present invention clearer, the following drawings are provided for illustration of the present invention:

[0018] Figure 1 It is a schematic diagram of the module composition of a signature information collection device for electronic signature in an embodiment of the present invention;

[0019] Figure 2 It is a schematic flowchart of a method for determining the consistency of features in an embodiment of the present invention. Specific Embodiments

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] In the description of the present invention, it should be noted that, unless otherwise clearly defined and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be a direct connection or an indirect connection through an intermediate medium, and it may also be the internal connection of two components. It may be a wireless connection or a wired connection. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0022] In addition, the technical features involved in different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0023] An embodiment of the present invention provides a signature information collection device for electronic signature, as Figure 1 shown, including:

[0024] An electronic signature input module 101, configured to receive electronic signature data input by a first user, where the electronic signature data includes a handwritten signature and an image-based signature;

[0025] A storage module 102, configured to store the handwritten feature parameters of a second user or the three-dimensional contour data and texture detail data of a physical signature, where the second user is the user represented by the electronic signature;

[0026] An electronic signature consistency determination module 103, configured to determine the consistency with the features represented by the electronic signature data according to the handwritten feature parameters of the second user or the three-dimensional contour data and texture detail data of the physical signature;

[0027] The communication module 104 is configured to transmit the electronic signature data to the target device after the signature passes the consistency comparison.

[0028] Exemplarily, the electronic signature input module 101 can be an electronic signature board, an image acquisition device, etc. Among them, the image acquisition device can be a camera or a scanning device, and this embodiment does not make any limitation thereto. The electronic signature data input by the first user can be a handwritten signature or an image-based signature. For example, it can be a stamped image on a paper document.

[0029] The storage module 102 can be a memory module or other devices with storage functions. The storage module 102 pre-receives the second user's handwritten feature parameters or the three-dimensional contour data and texture detail data of the physical seal imported from an external device. Among them, the handwritten feature parameters include stroke feature parameters, writing speed parameters, pressure feature parameters, etc. The stroke feature parameters can include stroke width and stroke inclination angle, etc. The three-dimensional contour data of the physical seal includes outer shape data, edge contour data, and concave-convex feature data, etc. Among them, the outer shape data includes shape parameters and size parameters. When the shape is a circle, the size parameter is the radius length. When the shape is a quadrilateral, the size parameter is the side length. When the shape is an ellipse, the size parameter is the length of the major and minor axes. When the shape is a polygon, the size parameter is the length of each side and the corresponding included angle. The edge contour data includes edge curve coordinates and / or edge roughness data, and the concave-convex feature data includes raised part data and sunken part data. The external device can be a computing device with a handwritten input function and handwritten feature parameter analysis, or a computing device with image scanning, image acquisition, and image data processing. It should be noted that the second user and the first user can be the same user or different users, and this embodiment does not make any limitation thereto.

[0030] The electronic signature consistency determination module 103 can be a computing module equipped with an image recognition algorithm and a consistency comparison algorithm. The image recognition algorithm can identify the user name represented by the signature text or signature, so as to retrieve the three-dimensional contour data and texture detail data of the corresponding physical signature or the handwritten feature parameters corresponding to the user name from the memory. The consistency comparison algorithm can compare the contour data of the electronic signature imageized signature extracted with the three-dimensional contour data of the stored physical signature. For example, a shape matching algorithm, such as a shape matching method based on Fourier descriptors, can be used to calculate the similarity of the contours of the two. For the texture detail data, a texture analysis algorithm, such as the gray-level co-occurrence matrix (GLCM) algorithm, is used to calculate the similarity of the texture features, and to compare whether features such as the direction and frequency of the texture are consistent. Combining the results of the contour similarity and texture similarity forms an overall consistency judgment index. A comprehensive threshold can be set. When both the contour similarity and texture similarity reach a certain degree and the comprehensive index exceeds the threshold, it is considered that the electronic signature data is consistent with the three-dimensional contour data and texture detail data of the physical signature. For the comparison of the handwritten feature parameters, the preprocessed handwritten feature parameters of the electronic signature are matched one by one with the handwritten feature parameters of the second user in the storage module. For example, it is compared whether the stroke lengths are equal within a certain error range and whether the deviation of the stroke angles is within an acceptable threshold. For dynamic features such as writing speed and pressure change, algorithms such as dynamic time warping are used to calculate the similarity between the two to adapt to the situation where the writing speeds are different but the features are similar. This embodiment does not limit the consistency comparison algorithm, and those skilled in the art can determine it according to needs.

[0031] The communication module 104 is used to transmit the electronic signature data to the target device after the signature passes the consistency comparison. Among them, the target device can be a computer, a storage device, etc. When the signature fails to pass the consistency comparison, an alarm can be issued and the electronic signature data of this time is not uploaded to the target device. It should be noted that in order to ensure that the signature information collection device is the only device for the target device to obtain the signature electronic information, a firewall white list can be set on the target device to only allow this device to upload data, so as to ensure that all the signatures in the target device are electronic, real and secure.

[0032] The embodiment of the present invention provides a signature information collection device for electronic signature. In terms of data collection, it ensures comprehensive and accurate collection of signature features, laying a solid foundation for subsequent security verification. In terms of data processing and verification, it can determine the authenticity of the entered signature by judging the consistency between the entered signature and the pre-stored valid signature, plugging loopholes and preventing lawbreakers from signing on behalf of others or forging signatures for electronicization. In terms of data storage and transmission, a safe and reliable encryption verification mechanism is used to protect electronic signature data from being tampered with, prevent data from being intercepted and destroyed by middlemen during transmission, and improve the security of electronic signatures.

[0033] As an optional implementation, a signature information collection device for electronic signature further includes:

[0034] The biometric information collection module is used to collect the biometric information of the first user for entering the electronic signature data; the first encryption module is used to encrypt and encode the electronic signature data and the first user's biometric data to obtain the target electronic signature data; wherein the communication module is used to transmit the target electronic signature data to the target device when the signature passes the consistency comparison.

[0035] Exemplarily, the biometric information collection module includes a fingerprint recognition device, a facial recognition camera, an iris recognition device, etc. It should be noted that in this embodiment, all biometric information collection processes are carried out after a clear prompt message has been sent to the user to inform the user that biometric collection is required and the consent is obtained. The first encryption module is a computing module that stores an encryption algorithm and a random number generator. The module associates and integrates the electronic signature data and the biometric information data to form a data set to be encrypted, ensuring that the corresponding relationship between the two is clear, so that subsequent encryption operations can simultaneously protect the security of the two parts of data. The encryption algorithm can be an advanced encryption standard (AES), elliptic curve encryption (ECC) or RSA algorithm, etc. A pair or a group of keys is generated using the key generation mechanism of the encryption algorithm. The key generation process usually relies on a secure random number generator to ensure the randomness and unpredictability of the key. For symmetric encryption algorithms (such as AES), a shared key is generated; for asymmetric encryption algorithms (such as RSA, ECC), a public key and a private key pair are generated. Based on the biometric information collection module and encryption module proposed in this embodiment, the electronic signature data transmitted to the target device in this embodiment is the target electronic signature data obtained by encrypting the electronic signature data and the first user's biometric data. The security of the data is ensured by encryption, and the traceability of the use of the electronic signature can be improved by collecting the first user's biometric data.

[0036] As an optional implementation, a signature information collection device for electronic signature, an electronic signature input module, includes:

[0037] An automatic light compensation module for automatically compensating the light of the scanning area according to the current light;

[0038] A scanning module for scanning the scanning area to obtain a signature image;

[0039] An automatic distortion correction module for automatically correcting the distortion of the signature image.

[0040] Exemplarily, the automatic light compensation module includes a light sensor and a control module. The light sensor is used to detect the intensity of the current ambient light in real time and transmit the detected light intensity value (usually in Lux) to the control module. The control module calculates whether the current ambient light meets the scanning requirements based on the data transmitted by the light sensor. If the light is insufficient, the control module calculates the light intensity that needs to be supplemented according to the preset light threshold. For example, if the preset minimum light threshold is 350 Lux and the currently detected light intensity is 200 Lux, then 150 Lux of light needs to be supplemented. The control module automatically adjusts the brightness of the fill light according to the calculation result. The fill light can be an LED light, and the brightness is controlled by adjusting the current or PWM (pulse width modulation) signal. The light direction and angle of the fill light can also be adjusted as needed to ensure that the light evenly irradiates the scanning area. After light compensation, the light sensor detects the light intensity of the scanning area again to ensure that the light is uniform and reaches the preset threshold. If the light still does not meet the requirements, the control module continues to adjust the brightness of the fill light until the light reaches the ideal state.

[0041] After the automatic light compensation module reaches the ideal state, the scanning module starts scanning to obtain a signature image. After scanning is completed, the acquired image can also be preliminarily preprocessed, such as denoising, contrast adjustment, etc., to improve the image quality. Then, the automatic distortion correction module obtains the scanned signature image data from the scanning module. Image processing algorithms (such as edge detection, corner detection, etc.) are used to detect the distortion situation in the image. For example, by detecting the edge lines of the image, it is judged whether there is perspective distortion or geometric distortion. According to the detected distortion type and degree, a suitable correction algorithm is selected, such as perspective correction, geometric distortion correction, etc. The correction algorithm is applied to process the image to obtain the corrected image. Finally, the quality of the corrected image is evaluated to check whether there is residual distortion or distortion. If the image quality is not ideal, the parameters of the correction algorithm can be adjusted and corrected again until the image quality meets the requirements. The embodiment of the present invention provides a signature information collection device for electronic signature, which can improve the effect of electronic signature input and the accuracy of subsequent comparison through the automatic light compensation module and the automatic distortion correction module.

[0042] As an optional implementation manner, the electronic signature input module includes:

[0043] An electronic signature module, configured to receive the user's handwritten signature data;

[0044] A pressure sensor array, configured to collect the pressure at different regions when the user writes a signature by hand;

[0045] A timer, connected to the pressure sensor array, configured to record the time when pressure is generated at different regions during the user's handwritten signature;

[0046] A second encryption module, configured to extract handwritten feature parameters from the pressure sensor array and timer data during the user's handwritten signature data, preprocess the handwritten feature parameters, and then perform asymmetric encryption on the handwritten signature data based on the preprocessed handwritten feature parameters to obtain the encrypted user electronic signature.

[0047] Exemplarily, the electronic signature module can be an electronic version with only a handwritten signature function, such as a graphics tablet, or a mobile device with a touch screen (such as a tablet computer or a smart phone). Integrating a pressure sensor array under the graphics tablet or touch screen can detect the pressure distribution during signature in real time. The pressure sensor array converts the detected pressure data into an electrical signal and transmits it to the processing module through an interface. The processing module synchronizes the pressure data with the signature image and records the pressure changes of each stroke. The timer is connected to the pressure sensor array and uses a high-precision timer to record the start and end times of each stroke during the signature process. The timer works synchronously with the pressure sensor array to ensure that the time data corresponds one-to-one with the pressure data. Finally, the time data and the pressure data are combined to form complete dynamic handwritten signature feature data.

[0048] The second encryption module extracts handwritten feature parameters, such as stroke feature parameters, writing speed parameters, pressure feature parameters, etc., from the pressure sensor array and timer data during the user's handwritten signature data, preprocesses the handwritten feature parameters, and then performs asymmetric encryption on the handwritten signature data based on the preprocessed handwritten feature parameters to obtain the encrypted user electronic signature. In this embodiment, the preprocessed handwritten feature parameters are used to perform asymmetric encryption on the handwritten signature data in order to decrypt it later using the pre-stored second user handwritten feature parameters, and determine whether the handwritten signature data has signature entry authorization or whether it is a proxy signature by judging whether decryption can be completed. The specific decryption method can be to call all the pre-stored second user handwritten feature parameters in the storage module and use these second user handwritten feature parameters as the decryption key elements of the asymmetric decryption algorithm for decryption in turn. If decryption is successful, it is considered that the input handwritten signature meets the entry authorization conditions.

[0049] Further, since there are still acts of signing on behalf of others. For example, both user A and user B have the authorization for handwritten signature input. User A signs the name of user B and inputs it into the device. In this case, since the handwritten feature parameters of user A can match the handwritten feature parameters of the user stored in the storage module, it may be impossible to identify this act of signing on behalf of others. Therefore, the decrypted user handwritten signature can also be subjected to name recognition to obtain the signed user name, and the signed user name is compared with the user name corresponding to the second user handwritten feature parameter for decryption to determine whether it is a signature on behalf of others.

[0050] As an alternative implementation, the electronic signature is obtained by converting a physical seal. According to the three-dimensional contour data and texture detail data of the physical seal, the consistency with the characteristics represented by the electronic signature data is determined. As Figure 2 shown, it includes: S101, constructing a three-dimensional seal digital model according to the three-dimensional contour data and texture detail data of the physical seal;

[0051] S102, determining multiple target projection angles based on the geometric characteristics of the three-dimensional seal digital model;

[0052] S103, projecting the seal part of the three-dimensional seal digital model onto a two-dimensional plane at multiple target projection angles according to the projection algorithm of computer graphics to obtain multiple two-dimensional projections;

[0053] S104, performing feature pooling on the multiple two-dimensional projections and performing feature fusion on the multiple two-dimensional projections to obtain the fusion feature of the seal part in the three-dimensional seal digital model;

[0054] S105, extracting multiple types of feature data in the electronic signature according to the target feature extraction algorithm, including contour features and texture semantic features;

[0055] S106, determining the contour features in the fusion feature, and determining the consistency between the contour features in the electronic signature and the corresponding contour features in the fusion feature according to the first target algorithm;

[0056] S107, determining the texture semantic features in the fusion feature, and determining the consistency between the texture semantic features in the electronic signature and the corresponding texture semantic features in the fusion feature according to the second target algorithm;

[0057] S108, determining the consistency of the physical seal and the electronic signature features according to the consistency between the contour features and the consistency between the texture semantic features.

[0058] Exemplarily, first, the collected three-dimensional contour data is denoised to remove the noise points generated due to scanning errors, and a filtering algorithm is used to smooth the data. For the texture detail photos, operations such as grayscale conversion and contrast enhancement are performed to improve the clarity of the texture information. Then, a three-dimensional modeling software is used to import the preprocessed three-dimensional contour data and texture detail data, and a three-dimensional model is constructed in the software according to the actual shape and size of the physical seal. The texture is mapped onto the surface of the model. Then, a geometric property analysis is performed on the constructed three-dimensional seal digital model to determine key geometric elements such as the axis of symmetry and the main planes of the model. For example, for a square seal, the planes where the four sides and the diagonals are located are the key analysis objects. Therefore, according to the geometric properties, with the center of the model as the reference point, the projection angle can be calculated through a mathematical algorithm. Specifically, based on the principle of uniform distribution, multiple equally spaced points can be selected within a certain range for projection. For a square seal, the positions where the points corresponding to the midpoints of the four sides of the square seal are vertically extended by 20 cm can be used, and the depression angle of 30° can be used as the four projection angles to ensure that the projection information of the model is obtained from multiple directions. For a circular seal, the points can be determined by equally dividing the circumferential line into six parts, and then vertically extended from these points, and the depression angle of 30° can be used as the six projection angles. The determination method of the projection angle in this embodiment is not limited, and those skilled in the art can determine it according to needs.

[0059] Using the orthographic projection algorithm in computer graphics, at the selected multiple target projection angles, the seal part of the three-dimensional seal digital model is input into the projection algorithm for calculation. The calculation results are rendered onto a two-dimensional plane to generate two-dimensional projection images. The multiple two-dimensional projection images are the seal parts of the three-dimensional seal digital model mapped from multiple angles. In this way, during subsequent comparison, it can adapt to the situation where due to the environment and personal habits, the first user actually stamps on the paper with different force application directions, resulting in distortion of the stamped impression. In this embodiment, by mapping the seal part of the three-dimensional seal digital model from multiple angles and comparing the seal features of the three-dimensional seal digital model fused with the mappings from multiple angles with the electronic seal data entered by the first user for consistency, it can avoid the situation where due to personal force application habits or environmental factors, the electronic seal data entered by the first user is distorted and cannot match the two-dimensional projection of the three-dimensional seal digital model mapped at a certain fixed angle, thereby reducing the matching accuracy.

[0060] For multiple two-dimensional projection images, the max pooling or average pooling method is adopted. Max pooling selects the maximum value within the pooling region as the output feature, highlighting the key features in the image; average pooling calculates the average value within the pooling region as the output feature, which can smooth the image features. The multiple two-dimensional projection feature maps after pooling are fused. Specifically, the channel concatenation method can be used to connect different feature maps in the channel dimension to form a fused feature tensor, or the weighted fusion method can be used to assign different weights according to the importance of each two-dimensional projection, and then add the weighted feature maps to obtain the fused feature.

[0061] The target feature extraction algorithm includes edge detection algorithms such as the Canny algorithm and the Sobel algorithm. Taking the Canny algorithm as an example, first, Gaussian filtering is performed on the electronic signature image to reduce noise, then the gradient magnitude and direction of the image are calculated, the edges are refined through non-maximum suppression, and finally, the image contour is determined by double-threshold detection and edge tracking to extract the contour features and texture semantic features.

[0062] After extracting the contour features and texture semantic features, the first target algorithm and the second target algorithm are used to calculate the consistency between the texture semantic features in the electronic signature and the corresponding fused features. Specifically, the first target algorithm can adopt a shape matching algorithm such as the matching algorithm based on Fourier descriptors. The Fourier descriptors of the contour features extracted from the electronic signature are compared with the Fourier descriptors of the contour features in the fused features to calculate their similarity. A similarity threshold is set. If the similarity is higher than the threshold, it is considered that the contour features are consistent; otherwise, they are inconsistent. The second target algorithm can use the cosine similarity algorithm. The texture semantic feature vectors extracted from the electronic signature and the texture semantic feature vectors in the fused features are normalized, and then their cosine similarity is calculated. Similarly, a threshold is set, and the consistency of the texture semantic features is determined according to the comparison result of the similarity and the threshold. In this embodiment, the first target algorithm and the second target algorithm are not limited, and those skilled in the art can determine them according to needs.

[0063] As an alternative implementation, the electronic signature is an encrypted user electronic signature, and preprocessing is performed on the handwritten feature parameters, including: querying the numerical range set where each handwritten feature parameter is located in the mapping table, and using the number of each numerical range set as the preprocessed handwritten feature parameter;

[0064] Determine the consistency with the characteristics represented by the electronic signature data according to the handwriting feature parameters of the second user, including: query the set of numerical ranges where the handwriting feature parameters of the second user are located in the mapping table, use the number of each set of numerical ranges as a decryption key element, and perform decryption using the asymmetric decryption method. When the decryption is successful, it indicates that the characteristics represented by the electronic signature data are consistent with the handwriting feature parameters of the second user; when the decryption is unsuccessful, it indicates that the characteristics represented by the electronic signature data are inconsistent with the handwriting feature parameters of the second user.

[0065] Exemplarily, in order to avoid differences in the handwriting feature parameters of the first user due to changes in the environment or personal status in this embodiment, resulting in inability to match the handwriting feature parameters of the second user in the storage module. Therefore, in this embodiment, before the encryption process in the second encryption module, preprocessing is performed on the handwriting feature parameters. The specific method is to expand the range of the handwriting feature parameters. Taking the stroke feature parameters as an example of the handwriting feature parameters, query the set where the stroke inclination angle is located, and use the set number as the stroke inclination angle value. For example, if a certain stroke inclination angle is 15°, and it is included in the set of 14° - 16°, and the set number is 10, then the number 10 can be used as the stroke inclination angle value. It should be noted that in this case, the handwriting feature parameters of the second user stored in the storage module are also obtained in the same way.

[0066] After preprocessing the handwriting feature parameters in the above manner, use the second encryption module to perform asymmetric encryption on the handwriting signature data with the preprocessed handwriting feature parameters as the encryption key element to obtain the encrypted user electronic signature. On this basis, in this embodiment, use the corresponding numerical range number of the handwriting feature parameters of the second user in the storage module as the decryption key element and perform decryption using the asymmetric decryption method. If decryption is successful, it indicates that the characteristics represented by the electronic signature data are consistent with the handwriting feature parameters of the second user; when the decryption is unsuccessful, it indicates that the characteristics represented by the electronic signature data are inconsistent with the handwriting feature parameters of the second user. It should be noted that the sorting of the handwriting feature parameter numbers in the encryption process and the sorting of the handwriting feature parameter numbers in the decryption process are pre-agreed.

[0067] As an optional implementation manner, a signature information collection device for electronic signature also includes: when the characteristics represented by the electronic signature data are consistent with the handwriting feature parameters of the second user, add the second user number to the electronic signature data, and send the electronic signature data with the second user number added to the target device. The second user number is used to represent the identity of the second user. By synchronously sending the second user number to the target device, the signature user can be effectively traced.

[0068] As an optional implementation manner, a signature information collection device for electronic signature also includes:

[0069] The split seal signature recognition module is used to recognize whether the electronic signature data entered by the first user is a split seal signature;

[0070] The split seal signature optimization module is used to optimize the split seal signature image when it is recognized as a split seal signature.

[0071] Exemplarily, before recognition, the electronic signature data entered by the first user is first preprocessed. Through grayscale operation, the color image is converted into a grayscale image to simplify subsequent calculations. The noise interference in the image is removed using a noise reduction algorithm to improve the image quality. Image enhancement technology is adopted to enhance the contrast of the image and highlight the signature details. Then, feature extraction is performed. Since the edges of split seal signatures usually have unique continuity and fracture characteristics, an edge detection algorithm is used to extract the edge information of the image. Finally, based on the extracted features, a machine learning or deep learning model is used for recognition. In terms of machine learning, a support vector machine (SVM) model can be trained. The extracted features are used as input, and the SVM is trained with a large number of labeled split seal signatures and non-split seal signature samples to learn the feature differences between the two, so as to achieve classification recognition. In terms of deep learning, a convolutional neural network (CNN) model is constructed. The CNN is used to automatically extract the deep features of the image. For example, classic network structures such as VGG and ResNet are used, and through training with a large amount of image data, it can accurately determine whether the image is a split seal signature. When a split seal is recognized, due to the possible fracture or blurred areas where the split seal crosses the paper gap, an image repair algorithm is used for repair. For example, a repair method based on partial differential equations analyzes the pixel information around the fracture area and fills and repairs the fracture according to a certain mathematical model to make the signature lines continuous and complete.

[0072] As an alternative implementation, a signature information collection device for signature digitization further includes:

[0073] The electromagnetic shielding structure is arranged on the outer surface layer of the main body of the signature information collection device for signature digitization, and includes an alternating conductive layer and magnetic conductive layer, which is used to block or attenuate external electromagnetic waves.

[0074] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A signature information collection device for electronic signature, characterized in that: include: An electronic signature input module, used to receive electronic signature data input by a first user, where the electronic signature data includes a handwritten signature and an image signature; A storage module, used to store handwriting feature parameters of a second user or three-dimensional contour data and texture detail data of a physical signature, where the second user is the user represented by the electronic signature; An electronic signature consistency determination module, used to determine the consistency with the features represented by the electronic signature data based on the handwriting feature parameters of the second user or the three-dimensional contour data and texture detail data of the physical signature; The communication module is used to transmit the electronic signature data to the target device after the signature passes the consistency comparison; The electronic signature is converted from the physical seal. Based on the three-dimensional contour data and texture detail data of the physical seal, the consistency of the features represented by the electronic signature data is determined, including: Construct a three-dimensional digital model of the signature based on the three-dimensional contour data and texture detail data of the physical signature; Based on the geometric characteristics of the three-dimensional signature digital model, multiple target projection angles are determined; According to the projection algorithm of computer graphics, the signature part of the three-dimensional signature digital model is projected onto a two-dimensional plane at multiple target projection angles to obtain multiple two-dimensional projections; Perform feature pooling on multiple two-dimensional projections, perform feature fusion on multiple two-dimensional projections, and obtain the fusion features of the signature part in the three-dimensional signature digital model; According to the target feature extraction algorithm, multiple types of feature data are extracted from the electronic signature, including contour features and texture semantic features; Determine the contour feature in the fused feature, and determine the consistency between the contour feature in the electronic signature and the contour feature in the corresponding fused feature according to the first target algorithm; Determine the texture semantic features in the fused features, and determine the consistency between the texture semantic features in the electronic signature and the texture semantic features in the corresponding fused features according to the second target algorithm; The consistency between the physical signature and the electronic signature features is determined based on the consistency between the contour features and the consistency between the texture semantic features.

2. A signature information collection device for electronic signature according to claim 1, characterized in that: Also includes: A biometric information collection module, used to collect the biometric information of the first user who enters the electronic signature data; A first encryption module, used to encrypt and encode the electronic signature data and the first user's biometric data to obtain target electronic signature data; Among them, the communication module is used to transmit the target electronic signature data to the target device when the signature passes the consistency comparison.

3. The signature information collection device for electronic signature according to claim 1, characterized in that: Electronic signature entry module, including: Automatic fill light module, used to automatically fill light to the scanning area according to the current light; A scanning module is used to scan the scanning area to obtain a signature image; The automatic distortion correction module is used to automatically correct the signature image.

4. The signature information collection device for electronic signature according to claim 1, characterized in that: Electronic signature entry module, including: Electronic signature module, used to receive the user's handwritten signature data; The pressure sensor array is used to collect the pressure of different areas when the user writes a signature; A timer, connected to the pressure sensor array, for recording the time when pressure is generated in different areas when the user writes a signature; The second encryption module is used to extract handwriting feature parameters from the pressure sensor array and timer data when the user writes the handwriting signature data, pre-process the handwriting feature parameters, and then asymmetrically encrypt the handwritten signature data according to the pre-processed handwriting feature parameters to obtain an encrypted user electronic signature.

5. The signature information collection device for electronic signature according to claim 4, characterized in that: The electronic signature is an encrypted user electronic signature, which pre-processes the handwriting feature parameters, including: Query the value range set of each handwriting feature parameter in the mapping table, and use the number of each value range set as the preprocessed handwriting feature parameter; Determining the consistency of the handwriting feature parameters of the second user with the features represented by the electronic signature data includes: Query the mapping table for the set of numerical ranges where the handwriting feature parameters of the second user are located, use the number of each numerical range set as a decryption key element, and use an asymmetric decryption method to decrypt. If the decryption is successful, it means that the features represented by the electronic signature data are consistent with the handwriting feature parameters of the second user; When the decryption fails, it means that the characteristics represented by the electronic signature data are inconsistent with the handwriting characteristic parameters of the second user.

6. The signature information collection device for electronic signature according to claim 5, characterized in that: Also includes: When the feature represented by the electronic signature data is consistent with the handwriting feature parameter of the second user, the second user number is added to the electronic signature data, and the electronic signature data with the second user number added is sent to the target device.

7. The signature information collection device for electronic signature according to claim 1, characterized in that: Also includes: A saddle signature recognition module, used to recognize whether the electronic signature data entered by the first user is a saddle signature; The saddle signature optimization module is used to optimize the saddle signature image when it is identified as a saddle signature.

8. The signature information collection device for electronic signature according to claim 1, characterized in that: Also includes: The electromagnetic shielding structure is arranged on the outer surface of the main body of the signature information collection device for electronic signature, including alternating conductive layers and magnetic conductive layers, which are used to block or attenuate external electromagnetic waves.

Images