Digital signature generation and authentication system and method

Abstract

This invention discloses a digital signature generation and authentication system and method, wherein the system includes a first terminal and a second terminal; the first terminal includes: a first acquisition module, which generates a first graphical interface and, in response to a user's signature executed on the first graphical interface, acquires first vector coordinate data signed on the first graphical interface; a second acquisition module, which samples the user's facial features; a data processing module, which extracts a first key parameter from the first vector coordinate data, uses the facial features as a second key parameter, and encrypts the first vector coordinate data using the first key parameter and the second key parameter to obtain key data; a storage module, which stores the key data; and a communication module, which transmits the key data to the second terminal; the second terminal includes a third acquisition module, which verifies the user's facial features; and a generation module, which verifies the facial features, decrypts the key data using the second key parameter, and then requests the first terminal to generate a digital copy for the second graphical interface.

Description

Technical Field

[0001] This invention relates to the field of encryption authentication technology, specifically to a digital signature generation and authentication system. Background Technology

[0002] Currently, mainstream digital signature systems on the market primarily employ Public Key Infrastructure (PKI) technology, using asymmetric encryption algorithms to generate and verify signatures. However, these systems have significant technical shortcomings: First, simple cryptographic protection is insufficient to defend against Advanced Persistent Threat (APT) attacks, where attackers can obtain users' private keys through social engineering or malware; second, traditional digital signatures lack binding to user biometrics, making it difficult to effectively verify whether the signing action was performed by the legitimate user, resulting in a high risk of signature misuse; third, signature data in existing systems is typically stored and transmitted in vector coordinate form, which, if intercepted, can be directly reused by attackers to forge signatures.

[0003] To enhance security, some systems attempt to combine biometric identification technology with digital signatures. For example, some solutions authenticate users by collecting dynamic behavioral characteristics (such as pen pressure and speed) during the signing process. However, these methods still have significant shortcomings: on the one hand, dynamic behavioral characteristics are easily affected by factors such as user status and device differences, resulting in unstable recognition accuracy; on the other hand, this feature data is usually stored in plaintext or simple encryption, offering limited security. Other technologies attempt to combine facial recognition with digital signatures, but most only treat facial recognition as an independent authentication step, lacking deep integration with the signature data itself, and thus failing to fundamentally prevent signature data from being tampered with or forged.

[0004] Furthermore, current technologies for protecting signature data largely rely on traditional encryption algorithms, lacking innovative processing of the data's form. When signature data exists in its raw vector coordinate form, even after encryption, an attacker can still directly obtain the complete signature information if the decryption key is leaked, causing the entire security system to collapse. Simultaneously, existing systems often directly use the raw signature data for comparison during the signature verification process, which not only increases the risk of data leakage but also allows attackers to obtain sensitive information by analyzing the verification process.

[0005] Therefore, there is an urgent need for a digital signature generation and authentication system that can deeply integrate user biometrics and signature data, possess advanced anti-counterfeiting capabilities, and effectively protect the security of original signature data, in order to solve the security risks existing in the current technology and meet the growing demand for security authentication. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a digital signature generation and authentication system to improve the security of digital signature usage.

[0007] A first aspect of the present invention includes a first terminal and a second terminal, wherein the first terminal is a generation terminal and the second terminal is an authentication terminal;

[0008] The first terminal includes:

[0009] The first acquisition module is configured to generate a first graphical interface and, in response to a user's signature performed on the first graphical interface, acquire the first vector coordinate data of the signature on the first graphical interface.

[0010] The second acquisition module is configured to sample the user's facial features;

[0011] The data processing module is configured to extract typical and atypical values ​​from the first vector coordinate data as a first key parameter, extract feature points and non-feature points from the facial features as a second key parameter, and perform encryption on the first vector coordinate data using the first key parameter and / or the second key parameter to obtain key data;

[0012] A storage module is configured to store the key data;

[0013] The communication module is configured to transmit the key data to the second terminal;

[0014] The second terminal includes:

[0015] The third acquisition module is configured to sample the user's facial features, extract the second key parameters, and return them to the first terminal for verification.

[0016] The generation module is configured to, after the facial feature verification is passed, decrypt the key data using the first key parameter and / or the second key parameter via the first terminal and then generate a digital copy to the second graphical interface.

[0017] As a preferred embodiment, the typical value includes the coordinates of the first preset type position of the signature, and the atypical value includes a preset number of other coordinates selected in the path of the signature besides the typical value.

[0018] In a preferred embodiment, the feature points include pixels at a second preset type location of the facial features, and the non-feature points include pixels randomly selected from the facial features.

[0019] As a preferred embodiment, the first acquisition module is configured as follows:

[0020] When generating the first graphical interface, the interface display area is adjusted so that the signature is located in the center of the first graphical interface;

[0021] Store the original vector coordinates of the signature when it is generated in the first graphical interface, and store the coordinate transformation values ​​of the typical value and the atypical value when the signature is located at the center position;

[0022] The original vector coordinates and the coordinate transformation values ​​are stored as the first vector coordinate data.

[0023] As a preferred embodiment, the data processing module is configured as follows:

[0024] Convert the first vector coordinate data into a first pixel matrix;

[0025] The signature is divided into multiple geometric parts by a segmented preprocessing method, and the pixel coordinates of each geometric part are redistributed.

[0026] The pixel coordinates of each of the aforementioned geometric parts are merged to obtain a pixel set, and different geometric numbers and coordinate values ​​are recorded.

[0027] The coordinate transformation values ​​are used as part of the encrypted pixel matrix, wherein the coordinate transformation values ​​include the typical values ​​and the atypical values.

[0028] Write the geometric sequence number, coordinate value, and coordinate transformation value into the encrypted pixel matrix, and delete redundant coordinate data in the first pixel matrix.

[0029] As a preferred embodiment, the data processing module is configured as follows:

[0030] Noise pixels and their virtual coordinates are generated according to a preset ratio of the pixel set;

[0031] The noise pixels are used as the confusion coordinate values ​​of adjacent data in each geometric part;

[0032] A block image is generated based on the obfuscated coordinate values, and the position of the block image in the first pixel matrix coordinate system is obtained;

[0033] Obtain the coordinates of the noise pixels at the boundary of the block image;

[0034] Obtain the updated coordinates of each pixel after it is written into the block image, and update the pixel set accordingly.

[0035] As a preferred embodiment, the data processing module is configured as follows:

[0036] The pixel data of the facial features are preprocessed and then written into the block image as part of the obfuscated data;

[0037] Generate a second key parameter based on the obfuscated data;

[0038] The second key parameter is combined with the first key parameter to perform a pixel migration operation on the first pixel matrix to encrypt the first vector coordinate data.

[0039] As a preferred embodiment, the generation module is configured as follows:

[0040] Scan the facial features of the user to be authenticated to obtain facial feature data;

[0041] The scanned facial feature data is compared with the facial features in the second key parameters to verify the validity and authenticity of the scanned facial feature data;

[0042] When the verification is successful, the key data is decrypted using the second key parameter;

[0043] A digital copy is generated based on the decryption result, and its authenticity is verified by comparing the digital copy with the signature data in the second graphical interface.

[0044] The digital copy mentioned therein is a non-vector copy that has been decrypted and subjected to pixel migration.

[0045] As a preferred embodiment, the data processing module is configured as follows:

[0046] Record the offset value between the first position coordinates when the first terminal samples the user's facial features and the second position coordinates when the second terminal samples the user's facial features;

[0047] The offset value is used as an additional key parameter, which, together with the first key parameter and the second key parameter, participates in the encryption process of the first vector coordinate data.

[0048] The added key parameters are stored in the key data in encrypted form.

[0049] A second aspect of the present invention provides a method for generating and authenticating digital signatures, comprising the following steps:

[0050] In the first terminal, a first graphical interface is generated, and in response to the signature performed by the user in the first graphical interface, the first vector coordinate data of the signature in the first graphical interface is obtained;

[0051] At the first terminal, the user's facial features are sampled;

[0052] Extract typical and atypical values ​​from the first vector coordinate data as the first key parameters;

[0053] Feature points and non-feature points are extracted from the facial features and used as the second key parameters;

[0054] The first vector coordinate data is encrypted using the first key parameter and the second key parameter to obtain key data;

[0055] The key data is transmitted to the second terminal;

[0056] On the second terminal, the user's facial features are verified;

[0057] Decrypt the key data using the second key parameter;

[0058] Request the first terminal to generate a digital copy to the second graphical interface;

[0059] The digital copy mentioned therein is a non-vector copy that has been decrypted and subjected to pixel migration.

[0060] Compared with the prior art, the present invention has the following advantages:

[0061] This invention deeply binds the vector coordinate data of the signature to the user's facial biometric features. During the signature generation stage, the system simultaneously collects the user's signature trajectory and facial features; during the authentication stage, both the signature data and the real-time collected facial features must be verified simultaneously, effectively preventing the risk of signature misuse by others.

[0062] This invention innovatively employs pixel migration technology to process signature data, converting the original vector coordinates into an encrypted pixel matrix. This process not only includes conventional encryption operations but also achieves a fundamental transformation of the data format through pixel migration guided by facial features, making it difficult for attackers to reverse engineer the original signature.

[0063] The digital copy generated during the authentication process in this invention is a decrypted and pixel-shifted non-vector copy, detached from the original vector coordinate system. This design fundamentally prevents the possibility of obtaining the original signature data through the authentication process. Even if the authentication terminal is compromised, attackers cannot obtain the original signature information that can be used for forgery, thus achieving the security goal of authentication without data leakage.

[0064] This invention utilizes feature points and non-feature points of facial features as dynamic obfuscation values, ensuring that the encryption parameters generated for each signature are associated with the user's current biometric status. Because facial features exhibit subtle random variations, even multiple signatures from the same user will generate different encryption parameters, effectively preventing attacks based on fixed keys and improving the system's dynamic security. Attached Figure Description

[0065] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0066] Figure 1 This is a schematic diagram of the system provided in an embodiment of the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] The digital signature generation and authentication system of the present invention, such as Figure 1 As shown, it includes a first terminal and a second terminal, wherein the first terminal is a generation terminal and the second terminal is an authentication terminal.

[0069] The first terminal includes a first acquisition module, a second acquisition module, a data processing module, a storage module, and a communication module; the second terminal includes a third acquisition module and a generation module.

[0070] In this embodiment, the first acquisition module is configured to generate a first graphical interface and, in response to a user's signature performed on the first graphical interface, acquire and store the vector coordinate data of the signature. Specifically, the first acquisition module displays a blank canvas on the touchscreen as the first graphical interface. When the user performs a signature operation using a finger or stylus, the system samples and captures the handwriting trajectory, accurately recording the coordinates and timestamps of each sampling point to form an original vector coordinate data sequence. When generating the first graphical interface, the first acquisition module dynamically adjusts the interface display area to automatically center the signature at the center of the first graphical interface and stores the original vector coordinates of the signature when the first graphical interface is generated. Simultaneously, the system calculates the coordinate transformation values ​​of typical and atypical values ​​when the signature is located at the center position using a coordinate transformation algorithm. The typical values ​​include the coordinates of the starting point, major turning points, and ending point of the signature, while the atypical values ​​include other coordinate points selected in the signature path besides the typical values. The coordinate transformation value is specifically represented as the offset vector between the original global coordinates and the local coordinates after center alignment, for example, the offset (+136, +176) of the starting point from the original coordinates (120, 80) to the center coordinates (256, 256). The first acquisition module stores the original vector coordinates and the coordinate transformation values ​​together as the first vector coordinate data, ensuring that the coordinate transformation values ​​serve as the metadata basis for the subsequent encrypted pixel matrix.

[0071] The second acquisition module is configured to sample the user's facial features and acquire feature points and non-feature points of the facial features. In this embodiment, the second acquisition module captures the user's facial image at a rate of thirty frames per second using a front-facing camera. It employs a deep learning-based facial keypoint detection algorithm to accurately locate sixty-eight facial feature points, including key anatomical points such as the inner endpoint of the eyebrow, the center of the pupil, the tip of the nose, the corners of the mouth, and the jawline. Simultaneously, a pseudo-random number generator uniformly selects two hundred non-feature points from the facial image as random pixels. These non-feature points strictly avoid feature areas (such as the eyes or lips) and are distributed in non-key areas such as the cheeks and forehead. Feature points include pixels at the second preset type of facial feature location, and non-feature points include randomly selected pixels from the facial features. The system assigns high brightness values ​​(255 levels) to feature points and medium brightness values ​​(128 levels) to non-feature points, forming a hierarchical facial feature dataset.

[0072] The data processing module is configured to extract typical and atypical values ​​from the first vector coordinate data as a first key parameter, and extract feature points and non-feature points from facial features as a second key parameter, and then perform encryption on the first vector coordinate data using the first and second key parameters. The specific implementation is as follows:

[0073] First, the data processing module converts the first vector coordinate data into the first pixel matrix. The system maps the signature trajectory to a 500x500 pixel two-dimensional coordinate system, sets the pixel resolution based on the density distribution of the original vector coordinate data, and converts the coordinates of each sampling point into the corresponding pixel position to form an initial pixel matrix. In this matrix, the signature trajectory is represented by white pixels (level 255), the background is filled with black pixels (level 0), and the timestamp information is implicitly encoded through the pixel grayscale gradient.

[0074] Secondly, the data processing module employs a segmented preprocessing approach, dividing the signature into multiple geometric parts. Based on the curvature variation characteristics of typical values, the system dynamically divides the signature path into five continuous geometric parts: the starting segment (from the starting point to the first major turning point), the first turning point segment (from the first major turning point to the middle segment), the middle segment (the smoothest area of ​​the path), the second turning point segment (from the middle segment to the last major turning point), and the ending segment (from the last major turning point to the ending point). For each geometric part, the system independently sets the origin of the local coordinate system as its starting point coordinates, and reassigns the global coordinates to local coordinates. For example, the global coordinates (120, 80) of the starting segment are remapped to the local coordinates (0, 0), eliminating the dependency on global coordinates.

[0075] Next, the data processing module merges the pixel coordinates of each geometric part to obtain a pixel set, recording only the different geometric indices and coordinate values. The system traverses all geometric parts, identifies pixels with the same local coordinate values, and merges them into a single data item. Each data item contains a geometric indice (e.g., the starting segment is marked as G1) and the corresponding local coordinate value (e.g., (50, 30)), while removing duplicate coordinates. This process uses a hash table to achieve efficient indexing, compressing the original vector data to 40% of its original volume.

[0076] Subsequently, the data processing module integrates the coordinate transformation values ​​as a core component of the encrypted pixel matrix. The coordinate transformation values ​​include a set of offset vectors for typical and atypical values, such as the offset of the starting point (+136, +176) and the average offset of the atypical values ​​(+100, +150). The system writes the geometric index, local coordinate values, and coordinate transformation values ​​together into the metadata area of ​​the encrypted pixel matrix, forming a three-dimensional data structure: the first dimension stores the geometric index, the second dimension stores the local coordinate values, and the third dimension stores the coordinate transformation values, ensuring a strong binding relationship between the coordinate transformation values ​​and the pixel set.

[0077] Then, the data processing module generates a pixel set. The pixel set is not a simple aggregation based on the same coordinate values, but a structured data set constructed through segmented preprocessing, containing a complete mapping of geometric indices, local coordinate values, and coordinate transformation values. The system removes redundant global coordinate data in the first pixel matrix, retaining only the core information of the pixel set, further compressing the data volume to 25% of the original volume.

[0078] Furthermore, the data processing module generates noise pixels and their virtual coordinates according to a preset ratio of the pixel set. The system sets the noise ratio to 15%, and defines the noise pixels as the scrambled coordinate values ​​of adjacent data in each geometric part: specifically, at the boundary of the geometric part (such as the boundary area between the starting segment and the first turning segment), the local coordinate transition points of adjacent geometric parts are extracted, and virtual coordinate points are generated as scrambled coordinate values; for example, the midpoint (202, 152) between the local coordinates of the end point of the starting segment (200, 150) and the local coordinates of the starting point of the first turning segment (205, 155) is selected as the scrambled coordinate value to blur the boundary features.

[0079] The data processing module generates block images based on obfuscated coordinate values. Each block image is composed of obfuscated coordinate values, and each block image corresponds to a geometric boundary region (such as the intersection of the starting segment and the first turning segment), and its position is accurately located in the first pixel matrix coordinate system. The system calculates the coordinates of the noise pixels at the boundary of the block image and uses these coordinates as the core elements of the obfuscated data, such as the coordinates of the upper left corner (180, 130) and the lower right corner (220, 170) of the block image.

[0080] Subsequently, the data processing module obtains the updated coordinate positions of each pixel after it is written into the block image. The system adjusts the pixel coordinates through a dynamic displacement algorithm, which combines the overall distribution characteristics of facial features: first, it calculates the geometric center coordinates of the facial feature points, and then generates a displacement vector field based on the feature point density distribution; for example, when facial feature points are concentrated in the left half of the region, the displacement vector field is shifted to the left by 5%, so that the updated coordinate positions are dynamically associated with the facial feature data; the updated coordinate positions are written into the pixel set in real time, forming a dynamic coordinate mapping in the encryption process.

[0081] Crucially, the data processing module preprocesses the pixel data of facial features and writes it into the block image as part of the obfuscated data. Specifically, the system converts the feature points and non-feature points of facial features into standardized pixel vectors: histogram equalization is used to eliminate illumination differences, and affine transformation is used to normalize the scale to a uniform size, generating facial feature pixel data with fixed dimensions. This data is directly embedded into the pixel filling layer of the block image as obfuscated data. For example, the normalized facial feature pixel values ​​(such as the brightness value of the nose tip of 200) replace the noise pixel values ​​at the block image boundary, and are deeply fused with the obfuscated coordinate values.

[0082] Based on the aforementioned obfuscated data, the data processing module generates a second key parameter. The second key parameter includes a structural description of the obfuscated data (such as the embedding position of facial feature pixels), pixel migration rules (such as the calculation formula for the displacement vector field), and dynamic displacement algorithms (such as the feature point density weight coefficient). Its essence is entirely derived from the preprocessing results of facial features, ensuring that the second key parameter is strongly bound to the user's biometric features.

[0083] Finally, the data processing module combines the second key parameter with the first key parameter to perform pixel migration operation on the first pixel matrix. Based on the obfuscated data in the second key parameter, the system performs coordinate rearrangement and pixel value permutation on the pixel set: first, facial feature pixel data is injected into the migration process as an obfuscation factor, for example, using the normalized coordinates (100, 120) of the nose tip as a seed point to generate a random permutation sequence; then, the pixel set undergoes nonlinear coordinate rearrangement, causing the local coordinates (50, 30) of geometric index G1 to migrate to a new position (180, 210); after the migration is completed, an encrypted pixel matrix is ​​generated. This matrix is ​​completely detached from the original vector coordinate system, retaining only the encrypted pixel set data, while the original vector coordinate information is permanently destroyed.

[0084] The storage module stores the encrypted key data in a hardware-level secure encryption area, and the communication module transmits the encrypted key data to the second terminal through a preset transmission protocol.

[0085] On the second terminal side, the third acquisition module receives the encrypted key data and samples the user's facial features. The generation module performs the following operations:

[0086] First, the facial features of the user to be authenticated are scanned to obtain facial feature data. In this embodiment, the camera of the second terminal captures the user's real-time facial image with the same parameters and extracts sixty-eight feature points and two hundred non-feature points that are completely consistent with those of the first terminal.

[0087] Then, the scanned facial feature data is compared with the facial features in the second key parameters to verify the validity and authenticity of the scanned facial feature data. Validity verification includes checking image sharpness (by edge sharpness index), lighting conditions (by grayscale variance threshold), and integrity (by feature point coverage); authenticity verification uses multimodal liveness detection technology, confirming it as a real human face through micro-expression analysis (detecting blink frequency) and depth perception (infrared 3D imaging).

[0088] When verification is successful, authorization is granted to decrypt the encrypted pixel matrix using the second key parameter. The decryption process is the inverse operation of the encryption process, specifically including: recalculating the obfuscated data structure based on the facial features obtained from the scan, restoring the original geometric indices and local coordinate values ​​of the pixel set through inverse pixel migration operations, and then reconstructing the first vector coordinate data; however, the reconstruction process strictly depends on real-time facial features, and static data cannot be used to complete the decryption.

[0089] Based on the decryption result, the generation module generates a digital copy and verifies its authenticity by comparing the digital copy with the signature data to be authenticated. The digital copy is a non-vector copy that has undergone decryption and pixel migration. Specifically, the decrypted pixel set is converted into a 400x400 pixel binary dot matrix image, where white pixels represent the signature trajectory and black pixels represent the background. Because facial feature obfuscation data is injected during the pixel migration operation, the pixel positions in the digital copy have been permanently rearranged and have no direct geometric mapping relationship with the original signature. For example, the starting point (120, 80) in the original signature is represented as a discrete pixel (180, 210) in the copy. However, this mapping depends on real-time facial features during authentication (such as the coordinates of the tip of the nose), making it impossible to reverse engineer the original vector coordinate data. Even if an attacker obtains this copy, they cannot map it back to the original signature through coordinate reverse engineering or image matching, thus completely blocking replay attacks and data theft risks, providing security far exceeding that of traditional vector signatures.

[0090] In this embodiment, the data processing module also records the offset value between the first position coordinates when the first terminal samples the user's facial features and the second position coordinates when the second terminal samples the user's facial features. This offset value originates from the change in the position of the facial features in the image coordinate system during the two sampling processes. For example, due to the tilt of the user's head, the tip of the nose may shift from the first terminal coordinates (300, 200) to the second terminal coordinates (310, 195). The system uses this offset value as an additional key parameter, which, together with the first key parameter and the second key parameter, participates in the encryption process of the first vector coordinate data. The additional key parameter is embedded in the key data in the form of an encrypted hash, enabling cross-terminal authentication to have dynamic anti-offset capability.

[0091] In one specific embodiment of the present invention, a user completes a signature operation on a tablet computer (first terminal). A first acquisition module ensures the signature is located at the center of the interface and stores the coordinate transformation value, while a second acquisition module simultaneously captures facial features. A data processing module performs segmented preprocessing, cutting the signature into five geometric parts. After redistributing the coordinates of each geometric part, only the geometric sequence number and local coordinate values ​​are recorded, and the coordinate transformation values ​​are integrated into the metadata area of ​​the encrypted pixel matrix. The facial feature pixel data, after normalization preprocessing, is written as obfuscated data into a block image. The block image is composed of obfuscated coordinate values ​​of adjacent boundaries of geometric parts (such as the midpoint at the intersection of the starting segment and the first turning segment), used to generate the second key parameter. The system generates an encryption key and transmits it to an authentication terminal (such as an authentication device at a bank counter) through a secure channel. When a user goes to the bank to conduct business, a second terminal scans their facial features, and a third acquisition module verifies the facial features and calculates the coordinate offset values ​​of the two sampling positions. After the generation module verifies the data, it uses the second key parameter to decrypt the key data. However, the generated digital copy is a non-vector bitmap image that cannot be mapped back to the original signature: all pixel positions in the copy are dynamically rearranged driven by facial features. For example, the continuous strokes of the original signature are represented as discrete pixel clusters in the copy, and the pixel distribution pattern dynamically changes with each authentication of facial features. This design ensures that even if the copy is intercepted, attackers cannot reconstruct the original signature trajectory because the copy does not contain any original vector coordinate information, and there is no reversible mapping relationship between its pixel layout and the original signature, thus providing security protection.

[0092] The foregoing description and accompanying drawings fully illustrate embodiments of this disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operations may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terminology used in this application is for describing embodiments only and is not intended to limit the claims. As used in the description of embodiments and claims, the singular forms “a,” “an,” and “the” are intended to equally include the plural forms unless the context clearly indicates otherwise. Similarly, the term “and / or” as used in this application means including one or more of the associated listed items and all possible combinations thereof. Additionally, when used in this application, the term "comprise" and its variations "comprises" and / or "comprising" refer to the presence of stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof. Without further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes said element. In this document, each embodiment may focus on the differences from other embodiments, and similar or identical parts between embodiments can be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, the relevant parts can be referred to the description of the method section.

[0093] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented using electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to achieve the described functions, but such implementation should not be considered beyond the scope of the embodiments of this disclosure. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the described devices, apparatuses, and units can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0094] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, function, and operation of possible implementations of apparatus, methods, and computer program products according to embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than those disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based device that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A digital signature generation and authentication system, characterized in that, It includes a first terminal and a second terminal, wherein the first terminal is a generating terminal and the second terminal is an authentication terminal; The first terminal includes: The first acquisition module is configured to generate a first graphical interface and, in response to a user's signature performed on the first graphical interface, acquire the first vector coordinate data of the signature on the first graphical interface. When generating the first graphical interface, the system dynamically adjusts the display area to automatically center the signature at the center of the first graphical interface and stores the original vector coordinates of the signature when the first graphical interface is generated. Simultaneously, the system calculates the coordinate transformation values ​​of typical and atypical values ​​when the signature is at the center position using a coordinate transformation algorithm. Typical values ​​include the coordinates of the signature's starting point, major turning points, and ending point, while atypical values ​​include other coordinate points selected along the signature path besides the typical values. The coordinate transformation value is specifically represented as the offset vector between the original global coordinates and the local coordinates after center alignment. The original vector coordinates and the coordinate transformation value are combined and stored as the first vector coordinate data. The second acquisition module is configured to sample the user's facial features; The data processing module is configured to extract typical and atypical values ​​from the first vector coordinate data as a first key parameter, and extract feature points and non-feature points from the facial features as a second key parameter. The feature points include pixels at key locations of the facial features, and the non-feature points include pixels randomly selected from non-key regions of the facial features. The first vector coordinate data is encrypted using the first key parameter and the second key parameter to obtain key data; the specific process includes: The data processing module converts the first vector coordinate data into a first pixel matrix; A segmented preprocessing method is adopted to cut the signature into multiple geometric parts and redistribute the pixel coordinates of each geometric part; the pixel coordinates of each geometric part are merged to obtain a pixel set, which only records the different geometric indices and coordinate values; the pixel set is a structured data set constructed by segmented preprocessing, containing a complete mapping of geometric indices, local coordinate values ​​and coordinate transformation values. The data processing module generates noise pixels and their virtual coordinates according to a preset ratio of the pixel set; The noise pixels are used as the confusion coordinate values ​​of adjacent data in each geometric part; Block images are generated based on confused coordinate values, and each block image corresponds to a geometric boundary region. The data processing module preprocesses the pixel data of facial features and writes it into the block image as part of the obfuscated data; The updated coordinate positions of each pixel are obtained after being written to the block image; the pixel coordinates are adjusted through a dynamic displacement algorithm. First, the geometric center coordinates of the facial feature points are calculated, and then a displacement vector field is generated based on the feature point density distribution; the updated coordinate positions are dynamically associated with the facial feature data; the updated coordinate positions are written to the pixel set in real time, forming a dynamic coordinate mapping in the encryption process. The data processing module combines the second key parameter with the first key parameter to perform pixel migration operation on the first pixel matrix; the system performs coordinate rearrangement and pixel value permutation on the pixel set based on the obfuscated data in the second key parameter; after the migration is completed, an encrypted pixel matrix is ​​generated as key data; A storage module is configured to store the key data; The communication module is configured to transmit the key data to the second terminal; The second terminal includes: The third acquisition module receives the encrypted key data and samples the user's facial features; The generation module scans the facial features of the user to be authenticated, obtains the scanned facial feature data, and verifies the validity and authenticity of the facial feature data; When the verification is successful, the received key data is decrypted. The decryption process is the inverse operation of the encryption process. Specifically, it includes: recalculating the obfuscated data structure based on the facial features obtained by scanning, restoring the original geometric sequence number and local coordinate value of the pixel set through the inverse pixel migration operation, and then reconstructing the first vector coordinate data. Based on the decryption result, the generation module generates a digital copy and verifies its authenticity by comparing the digital copy with the signature data to be authenticated. The digital copy is a non-vector copy that has been decrypted and subjected to pixel migration. Specifically, the decrypted pixel set is converted into a binary dot matrix image, where white pixels represent the signature trajectory and black pixels represent the background. Because facial feature obfuscation data is injected into the pixel migration operation, the pixel positions in the digital copy have been permanently rearranged and have no direct geometric mapping relationship with the original signature.

2. The digital signature generation and authentication system according to claim 1, characterized in that, The data processing module is configured as follows: Record the offset value between the first position coordinates when the first terminal samples the user's facial features and the second position coordinates when the second terminal samples the user's facial features; The offset value is used as an additional key parameter, which, together with the first key parameter and the second key parameter, participates in the encryption process of the first vector coordinate data. The added key parameters are stored in the key data in encrypted form.

Images