A PDF-based signature handwriting verification storage system and method

Abstract

The present application relates to the technical field of data processing, and more particularly to a kind of based on PDF signature handwriting verification storage system and method.The specific inclusion is: obtaining and processing signature handwriting data, constructing dynamic handwriting analysis model to carry out preliminary analysis of handwriting data, obtain the handwriting data with the parameter characteristics of signature handwriting, and carry out data cleaning and interpolation processing;Baseline model is constructed based on the handwriting data after interpolation processing, and signature handwriting data representing the unique style of user is obtained;Signature handwriting data of the unique style of user is compressed and encrypted, and integrated into PDF document;The signature handwriting data of integrated PDF document is recovered and tracked, and anti-tamper verification is carried out through hash function and difference algorithm.The technical problem that the integrity and accuracy of signature handwriting data analysis in prior art are poor, and the security of signature handwriting data is low is solved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a PDF signature handwriting verification and storage system and method. Background Technology

[0002] PDF (Portable Document Format) is an electronic representation of documents, regardless of the document's creation, hardware, operating system, or software application. PDF is an open standard, first released by Adobe Systems in 1993. PDF documents can contain links and buttons, form fields, audio, video, and business logic. An electronic signature is an electronic form of signature used to verify the signer's identity and confirm that the signer has seen and agreed to the content of the signed document. Electronic signature technology typically uses encryption to ensure the security and integrity of the signature. Handwriting storage technology is a technique for recording and storing handwritten characters, commonly used in handwriting input devices such as graphics tablets and touchscreens. Handwriting data can include information such as coordinates, pressure, and speed. Existing PDF signing technologies are typically based on digital certificates. Users sign using a private key and then verify the signature using a public key. This technology ensures the security and integrity of the signature but does not record the signer's actual handwriting.

[0003] There are many methods for note storage. Chinese patent application number "CN202210278374.4", invention title "A Smart Pen Handwriting Data Storage Method and System", mainly includes: a user writing their username and password using a smart pen and sending the obtained username and password to a smart education system and a Kerberos server; the Kerberos server verifying the user's identity, determining whether login to the smart education system is allowed based on the verification result, and establishing an application service connection based on the user's selected handwriting data storage service; after the smart education system establishes a handwriting data storage service with the user through the smart pen, the smart education system interacts with the smart pen to store the handwriting data collected by the smart pen. The system includes modules corresponding to the steps of the method.

[0004] However, the above-mentioned technologies have at least the following technical problems: poor completeness and accuracy in analyzing handwriting data, and low security in protecting the confidentiality of handwriting data. Summary of the Invention

[0005] This invention provides a PDF-based signature handwriting verification and storage system and method, which solves the technical problems of poor integrity and accuracy of signature handwriting data analysis and low confidentiality and security of signature handwriting data in existing technologies, and achieves the technical effect of complete, accurate and secure PDF-based signature handwriting storage.

[0006] The present invention provides a PDF signature handwriting verification and storage system and method, which specifically includes the following technical solutions:

[0007] A PDF signature handwriting verification and storage method includes the following steps:

[0008] S1. Acquire and process signature handwriting data, construct a dynamic handwriting analysis model to perform preliminary handwriting data analysis, obtain handwriting data with signature handwriting parameter features, and perform data cleaning and interpolation processing on the handwriting data with signature handwriting parameter features.

[0009] S2. Based on the interpolated handwriting data, a baseline model is constructed. By constructing a feature space and a graph structure, a new feature space is obtained, and the uniqueness of the user's signature is captured to obtain signature handwriting data that represents the user's unique style.

[0010] S3. Compress and encrypt the user's unique signature handwriting data, and integrate it into a PDF document;

[0011] S4. Recover and track the signature handwriting data of the integrated PDF document, and verify its tamper resistance using hash functions and differential algorithms.

[0012] Preferably, S1 specifically includes:

[0013] The system acquires user signature handwriting data and performs initial smoothing by averaging the acquired signature handwriting data using the moving average method. Then, it applies discrete Gaussian filtering for secondary smoothing to obtain smoothed data. The smoothed data is further subjected to preliminary dynamic analysis using a dynamic handwriting analysis model. At the same time, Tukey's fence method is used for data cleaning. Finally, Bézier curves are used for interpolation to obtain continuous signature handwriting data.

[0014] Preferably, in S1, it further includes:

[0015] The dynamic handwriting analysis model is specifically constructed as follows:

[0016]

[0017] in, It is a smoothed collection of handwriting data, containing information about each point in the user's signature; It is a set of coordinate differences; It is a set of time differences; It is a set of velocities; It is a set of direction angles; It is the set of cumulative curve lengths; It is a set of pressure-weighted speeds for handwriting; It is a measure of signature complexity; It is the output of the model.

[0018] Preferably, S2 specifically includes:

[0019] The baseline model is implemented as follows:

[0020] The first step is feature transformation; the interpolated handwriting data is transformed into temporal flow data, and then the temporal flow data is further transformed into the spectral domain.

[0021] The second step is pattern recognition and feature extraction. Feature vectors are extracted from the spectral data in the spectral domain, and feature selection and feature fusion are further performed. Dimensionality reduction is then carried out using a dimensionality reducer.

[0022] The third step is to construct a baseline model. Based on the feature vectors after dimensionality reduction, a feature space is constructed to obtain the set of data points in the feature space, and a similarity matrix is ​​calculated. Based on the similarity matrix, a graph structure is constructed, and a new feature space is generated. In the new feature space, clustering is used to capture the uniqueness of user signatures.

[0023] Preferably, S3 specifically includes:

[0024] The signature handwriting data with the user's unique style is compressed using a compression algorithm, the compressed signature handwriting data is encrypted, and finally the encrypted signature handwriting data is decrypted and integrated into a PDF document for storage.

[0025] Preferably, S4 specifically includes:

[0026] To perform tamper-proof verification, firstly, a unique hash value is generated from the encrypted signature handwriting data using a hash function and stored. When it is necessary to verify whether the signature handwriting data has been tampered with, the signature handwriting verification and storage system extracts the encrypted signature handwriting data from the integrated PDF document, uses a hash function to generate a hash value again from the extracted encrypted signature handwriting data, and compares the newly generated hash value with the original stored hash value for verification.

[0027] Preferably, in S4, the following is also included:

[0028] When the hash verification is inconsistent, the signature handwriting verification storage system prompts "the signature handwriting data has been tampered with" and initiates a recovery mechanism to recover the signature handwriting data through a differential algorithm and to track the recovered signature handwriting data; when the hash verification is consistent, the tracking is then performed.

[0029] A PDF signature handwriting verification and storage system includes the following components:

[0030] Handwriting capture module, handwriting analysis module, compression and storage module, encryption and security module, anti-tampering verification module, integration module, intelligent recovery and tracking module;

[0031] The handwriting capture module captures the handwriting data of the user's signature through sensors, including coordinates, pressure, and speed information, processes the handwriting data, and then transmits the processed handwriting data to the handwriting analysis module.

[0032] The handwriting analysis module analyzes the handwriting data and transmits the analyzed handwriting data to the compression and storage module.

[0033] The compression and storage module uses a compression algorithm to compress the signature handwriting data and then transmits the compressed signature handwriting data to the encryption and security module.

[0034] The encryption and security module encrypts the compressed signature handwriting data, provides encryption key management and storage, and transmits the encrypted signature handwriting data to the anti-tampering verification module and the integration module.

[0035] The integration module is used to integrate encrypted signature handwriting data with other content of the PDF document, including specifying the signature position, style and display method, supporting user-defined signature integration options to meet the needs of different scenarios, and passing the integrated PDF document to the anti-tampering verification module.

[0036] The anti-tampering verification module uses a hash function to generate a unique hash value for the encrypted signature handwriting data and stores it. When it is necessary to verify whether the signature handwriting data has been tampered with, the signature handwriting verification and storage system extracts the encrypted signature handwriting data from the integrated PDF document, uses a hash function to generate a hash value for the extracted encrypted signature handwriting data again, compares the newly generated hash value with the original stored hash value for verification, and passes the verification result to the intelligent recovery and tracking module.

[0037] The intelligent recovery and tracking module uses intelligent algorithms to track and analyze changes in signature handwriting data, restores it to its original state, and provides users with the function of tracking and retrieving historical signature records;

[0038] This method is applied to PDF signature handwriting verification and storage.

[0039] The beneficial effects of the technical solution of the present invention are:

[0040] 1. This invention performs preliminary dynamic analysis based on smoothed handwriting data, capturing and extracting rich dynamic signature features, such as speed, direction, weighted speed, and complexity. This provides more feature basis for subsequent biometric verification and signature verification. Capturing real-time signature features during the signing process can more accurately determine the authenticity and integrity of the signature.

[0041] 2. This invention ensures the integrity of features extracted from handwriting data by transforming handwriting data to the temporal domain and then to the spectral domain. This transformation can accurately capture subtle differences and uniqueness in signatures. Based on the spectral clustering method, a feature space is adaptively constructed, which can better describe and distinguish the signature styles of different users.

[0042] 3. This invention performs efficient data compression, thereby saving storage space and accelerating data transmission. It uses a secure random number generator to generate keys and IVs, and combined with the use of a hardware security module (HSM), it ensures the security of key management and prevents unauthorized access. Through an advanced PDF editing library, it can ensure seamless integration of signatures with other content in the PDF, and can also personalize the signature style according to the user's needs.

[0043] 4. This invention, through hash verification and differential recovery mechanisms, can ensure that the signature handwriting data in the PDF remains in its original state and has not been tampered with, thereby improving data security and strengthening the protection of signature handwriting data from external attacks and unauthorized access. It also utilizes a relational database to store the historical records of signature handwriting data and metadata, ensuring that every data change is recorded in detail, which facilitates subsequent auditing and verification. Attached Figure Description

[0044] Figure 1 This is a block diagram of a PDF signature handwriting verification and storage system provided in one embodiment of the present invention;

[0045] Figure 2 This is a flowchart of a PDF signature handwriting verification and storage method provided in one embodiment of the present invention. Detailed Implementation

[0046] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. 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.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0048] The following description, in conjunction with the accompanying drawings, details a specific solution for a PDF signature handwriting verification and storage system and method provided by the present invention.

[0049] See attached document Figure 1 The diagram illustrates a block diagram of a PDF signature handwriting verification and storage system according to an embodiment of the present invention. The system includes the following components:

[0050] Handwriting capture module, handwriting analysis module, compression and storage module, encryption and security module, anti-tampering verification module, integration module, intelligent recovery and tracking module;

[0051] The handwriting capture module captures the user's signature handwriting data through a high-precision sensor, including information such as coordinates, pressure, and speed. It processes the handwriting data to improve data quality and then transmits the processed handwriting data to the handwriting analysis module for verification.

[0052] The handwriting analysis module analyzes the handwriting data from the handwriting capture module to ensure the integrity of the signature handwriting, and then passes the analyzed and valid handwriting data to the compression and storage module for further processing.

[0053] The compression and storage module uses an efficient compression algorithm to compress the signature handwriting data to reduce file size and achieve efficient storage. The compressed signature handwriting data is then passed to the encryption and security module to ensure data security.

[0054] The encryption and security module encrypts the compressed signature handwriting data to ensure its security. It provides encryption key management and storage to ensure that only authorized users can decrypt the signature handwriting data. It also passes the encrypted signature handwriting data to the anti-tampering verification module to ensure that the signature handwriting data remains unchanged throughout the entire processing. At the same time, it passes the encrypted signature data to the integration module for integrated processing.

[0055] The integration module provides multiple ways to seamlessly integrate encrypted signature handwriting data with other content in a PDF document, such as specifying the signature location, style, and display method, and supporting user-defined signature integration options to meet the needs of different scenarios. The integrated PDF document is then passed to the anti-tampering verification module to verify the accuracy of the signature handwriting data.

[0056] The anti-tampering verification module uses a hash function to generate a unique hash value for the encrypted signature handwriting data and stores it. When it is necessary to verify whether the signature handwriting data has been tampered with, the system will extract the encrypted signature handwriting data from the integrated PDF document, use the same hash function to generate a hash value for the extracted encrypted signature handwriting data again, compare the newly generated hash value with the original stored hash value, and pass the verification result to the intelligent recovery and tracking module to ensure the integrity of the signature handwriting data.

[0057] The intelligent recovery and tracking module uses intelligent algorithms to track and analyze changes in signature handwriting data, restoring it to its original state to ensure the integrity of the signature. It also provides users with the ability to track and retrieve historical signature records for auditing and verification.

[0058] See attached document Figure 2 The diagram illustrates a flowchart of a PDF signature handwriting verification and storage method according to an embodiment of the present invention, which includes the following steps:

[0059] S1. Acquire and process signature handwriting data, construct a dynamic handwriting analysis model to perform preliminary handwriting data analysis, obtain handwriting data with signature handwriting parameter features, and perform data cleaning and interpolation processing on the handwriting data with signature handwriting parameter features.

[0060] When a user signs their name on a touchscreen or graphics tablet, a high-precision sensor captures the coordinates of each point in real time. ,pressure and time The original handwriting data sequence was obtained. :

[0061]

[0062]

[0063] in, It is the dimension of the handwriting data sequence, the sequence. Any element in can be derived from express, Indicates the first The handwriting data includes the first handwriting data. Each coordinate, pressure, and time. ;

[0064] To compensate for hand tremors or sensor noise, a moving average method is used to average the coordinates and pressure within a window size k to achieve initial smoothing. A discrete Gaussian filter is then applied for secondary smoothing to further smooth the handwriting. The smoothed handwriting data is then obtained. ;

[0065] Furthermore, to conduct more effective handwriting analysis, a dynamic handwriting analysis model is constructed based on the smoothed handwriting data to perform preliminary handwriting data analysis and obtain more parameter features of the signature handwriting; the dynamic handwriting analysis model is specifically constructed as follows:

[0066]

[0067] in, It is a smoothed collection of handwriting data, containing information about each point in the user's signature. Each of them It is a quadruple ; Indicates the first Smoothed handwriting data Indicates the first The coordinates of the smoothed handwriting. Indicates the first The pressure of the smoothed handwriting. Indicates the first The time of the smoothed handwriting; It is a set of coordinate differences, consisting of the coordinate differences between consecutive points, based on the set... Calculation of continuous data points, set any element It can be represented as: ; It is a set of time differences, consisting of the time differences between consecutive points, based on a set. Timestamp calculation, set any element It can be represented as: ; It is a velocity set, describing the movement speed of each part of the signature, calculated using the coordinate difference set A and the time difference set B. any element It can be represented as: , This represents the coordinate difference in the x-direction. This represents the coordinate difference in the y-direction; It is a set of direction angles, describing the direction of movement for each part of the signature, calculated based on the coordinate difference set A. any element It can be represented as: ; It is a set of cumulative curve lengths, describing the total length of the signature, calculated based on the coordinate difference set A. any element It can be represented as: ,in, ;gather It is a pressure-weighted velocity set for handwriting, describing the velocity variation considering pressure, based on velocity. and The pressure in the set is calculated. any element It can be represented as: ; It is a signature complexity metric that describes the relationship between speed and handwriting length, providing a measure of signature complexity based on speed. and cumulative curve length The calculation yielded: ; The output of the model describes some features of the handwriting, including speed, direction, weighted speed, and complexity, and is represented as: ;

[0068] The above dynamic handwriting analysis model is used to dynamically extract key features from handwriting data, and the extracted features are used for biometric verification, signature verification and other processing.

[0069] Dynamic handwriting analysis during handwriting capture can capture real-time features during user signatures, providing more details and features such as speed and direction, which is very helpful for subsequent processing.

[0070] Obtain handwriting data with parametric features of the signature. ;

[0071] The median and interquartile range of global velocity and pressure are calculated for handwriting data with parametric features of signature handwriting. Outliers are removed using Tukey’s fence method to obtain cleaned handwriting data.

[0072] To obtain a continuous handwriting curve and a uniform distribution of data points, interpolation of the data points is required. Using Bézier curves for interpolation results in smoother, more uniform continuous handwriting data.

[0073] The interpolated handwriting data is then structured to obtain a format suitable for storage and transmission, such as JSON, providing accurate handwriting data for subsequent processing.

[0074] S2. Based on the interpolated handwriting data, a baseline model is constructed. By constructing a feature space and a graph structure, a new feature space is obtained, and the uniqueness of the user's signature is captured to obtain signature handwriting data that represents the user's unique style.

[0075] A baseline model is constructed based on the interpolated handwriting data, and this model is used to represent the user's unique signature style. The specific model implementation is as follows:

[0076] The first step is feature transformation;

[0077] First, the interpolated handwriting data is temporally streamed, transforming it into temporally streamed data. Timestamp enhancement is used to associate each data point (e.g., velocity, direction) with a time stamp, resulting in temporally streamed data. Then, a wave-state extractor is used to convert the temporally streamed data into the spectral domain, completing the spectral transformation. The wave-state extractor is represented as follows:

[0078]

[0079] in, It is the representation of the transformed handwriting data in the spectral domain. In this domain, the data of each point represents the change of the original data at a specific frequency and time, describing the dominant wave state component in the spectral domain. It refers to the frequency in the spectral state domain; It is time-flow data; It is a window function used for localized frequency analysis, such as a Gaussian window or a Hamming window; It is a time variable used in the sliding window function during integration. ;

[0080] The second step is pattern recognition and feature extraction.

[0081] Based on the dominant wave state components in the spectral domain, i.e., spectral data, feature vectors are extracted from the spectral data using frequency analysis and energy analysis. Empirical methods are then used to filter the obtained feature vectors, achieving feature selection and resulting in feature fusion of the spectral data. Further, a "dimensionality reducer" is used for dimensionality fusion, reducing the feature vectors to a manageable size. This process employs an iterative method, attempting to remove the least important dimensions in each iteration until the desired number of dimensions is reached. The dimensionality reduction formula (dimensionality reducer) is expressed as:

[0082]

[0083] in, and These are the observations of dimensions i and j in feature fusion, i.e., the elements in the feature vector after feature fusion; and It is the average of dimensions i and j; These are weighting coefficients used to emphasize certain specific observations; Indicates the dimension after feature fusion; It is the covariance between dimension i and dimension j, reflecting the degree of linear relationship between dimension i and dimension j; if A value greater than 0 indicates a positive correlation between them; a value less than 0 indicates a negative correlation between them; and a value of 0 indicates no linear correlation between them.

[0084] Then calculate From the eigenvectors and eigenvalues, we obtain the dimensionality-reduced eigenvectors;

[0085] The third step is baseline model construction;

[0086] Based on the reduced feature vectors, a feature space is constructed, where each point represents a signature sample. Each dimension of the feature vector is used as a coordinate axis to obtain the set of data points in the feature space. Based on this set of data points, a similarity matrix is ​​calculated using a Gaussian kernel function, representing the similarity between any two points. :

[0087]

[0088] in, and These are two data points in the feature space. The width parameter of the Gaussian kernel is used to further obtain the similarity matrix. , ;

[0089] Furthermore, a graph structure is constructed using a similarity matrix. Construct an undirected graph G, where nodes represent data points and edge weights represent the similarity between two data points;

[0090] Spectral clustering is performed using the Laplacian matrix of graph G. First, the degree matrix D and the adjacency matrix A are calculated. Then, the Laplacian matrix L=DA is constructed. Next, the k smallest eigenvectors of L are found. These eigenvectors constitute a new feature space. In this new feature space, each point of the original data points will be mapped to a new location; thus, the new feature space is obtained.

[0091] In the new feature space, the K-means clustering algorithm is used to cluster the data points. The center of each cluster represents the feature vector of the user signature style. The feature vector is a high-dimensional data that contains various minute details and characteristics in the user signature, thereby capturing and describing its uniqueness.

[0092] In summary, once the baseline model is constructed, the feature vectors in the new feature space represent the signature handwriting data that reflects the user's unique style.

[0093] S3. Compress and encrypt the user's unique signature handwriting data, and integrate it into a PDF document;

[0094] First, to efficiently store user-unique signature handwriting data, Wavelet transform is used to compress the user-unique signature handwriting data. Wavelet transform shows high efficiency in processing continuous signature handwriting data. Then, professional technicians adjust the parameters of the compression algorithm according to the characteristics of the user-unique signature handwriting data to maximize the compression ratio. After that, the user-unique signature handwriting data is normalized and filtered to reduce noise and outliers. Finally, the preprocessed handwriting data is compressed using the selected algorithm and parameters.

[0095] Furthermore, to protect the security of the signature handwriting data, the compressed signature handwriting data is encrypted using AES-256 as the encryption algorithm. For enhanced encryption security, GCM (Galois / Counter Mode) is employed, providing integrity protection for the data. Then, a secure random number generator produces a 256-bit key. To ensure the security of this key, a Hardware Security Module (HSM) is used for secure storage, guaranteeing it is not accessed illegally. The encryption process also requires an Initialization Vector (IV), so a new IV is generated using a secure random number generator. Finally, the compressed signature handwriting data is encrypted using AES-256-GCM, the generated key, and the IV. This process can be represented as:

[0096]

[0097] in, It is compressed signature handwriting data; It is the generated key; It is an initialization vector;

[0098] Furthermore, the integration of signature handwriting data with PDF is as follows: First, the encrypted signature handwriting data is integrated into the PDF document. Then, the signature handwriting data is decrypted. Using a symmetric encryption algorithm tool, the decryption key is obtained through the encryption key management system. Finally, the corresponding decryption algorithm is used to decrypt the encrypted signature handwriting data, thereby obtaining the original signature handwriting data.

[0099] Next, the placement of the signature needs to be determined in the PDF. By using a PDF editing library, such as PDF.js or PyPDF2, the PDF document structure is read to find the user-specified or predefined signature placement area. Based on the size and shape of this area, the coordinates (x, y) of the signature are determined, where (x, y) = (specified x-coordinate, specified y-coordinate).

[0100] Subsequently, the signature handwriting data is styled according to the user's needs or preset options. Using image processing libraries such as OpenCV or PIL, the signature handwriting data is loaded as an image, and the signature image is adjusted according to user-specified or preset style parameters, such as size, color, and rotation. Specifically, interpolation algorithms can be used to change the size of the signature, while rotation algorithms can be used to adjust the direction of the signature. After adjustment, the signature handwriting data with the user's desired style is obtained.

[0101] Then, using a PDF editing library, the adjusted signature is inserted at the previously determined location. During this process, it is ensured that the signature is seamlessly integrated with other content in the PDF. For example, the order of layers can be adjusted to ensure that the signature is displayed at the correct level. After integration, a PDF document containing the user's signature handwriting and encrypted signature handwriting data is obtained.

[0102] S4. Recover and track the signature handwriting data of the integrated PDF document, and verify its tamper resistance using hash functions and differential algorithms.

[0103] Furthermore, data recovery and tracking are performed on the integrated PDF document containing the user's signature. First, a PDF parsing library such as PDF.js is used to load the integrated PDF and read the encrypted signature data and other relevant metadata from the PDF document, such as timestamps and signer information. The signature data is then separated from the metadata to prepare for subsequent processing.

[0104] For tamper-proof verification, firstly, a unique hash value is generated for the encrypted signature handwriting data using a hash function and stored. When it is necessary to verify whether the signature handwriting data has been tampered with, the system extracts the encrypted signature handwriting data from the integrated PDF document, generates a new hash value for the extracted encrypted signature handwriting data using the same hash function, compares this newly generated hash value with the originally stored hash value, and passes the comparison verification result to the intelligent recovery and tracking module to ensure the integrity of the signature handwriting data.

[0105] If the verification fails, the system will display a message stating "Signature handwriting data has been tampered with" and initiate a recovery mechanism. First, it retrieves the original version corresponding to the signature handwriting data from the backup database or cloud backup. Then, it uses a differential algorithm to find the differences between the current data and the original data. The specific recovery is based on the following mathematical formula:

[0106]

[0107] in, Represents the original signature handwriting data; This is the current signature handwriting data; The difference or change between the two is represented by a difference algorithm.

[0108] Using the above formula, signature handwriting data can be restored to its original state, and the modified signature handwriting data can be replaced with the restored original data in the PDF for tracking.

[0109] If the verification is consistent, then proceed with the follow-up.

[0110] Furthermore, to track and store the historical records of signature handwriting data, a relational database (such as MySQL or PostgreSQL) is used to store the signature handwriting data and metadata as structured records. Whenever the signature handwriting data is updated or changed, a new entry is created in the database to ensure that each version of the data is properly preserved. By providing a query interface to the database, it is possible to search by date, signer, and other conditions. The change history can be retrieved and viewed at any time. Finally, visualization tools or report generation tools are used to display the change history of the signature handwriting data to users for auditing and verification purposes.

[0111] In summary, a PDF signature handwriting verification and storage system and method have been completed.

[0112] The technical solutions in the above embodiments of the present invention have at least the following technical effects or advantages:

[0113] The order of the embodiments is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0114] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0115] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for verifying and storing signature handwriting on PDF, characterized in that, Includes the following steps: S1. Acquire and process signature handwriting data, construct a dynamic handwriting analysis model to perform preliminary handwriting data analysis, obtain handwriting data with signature handwriting parameter features, and perform data cleaning and interpolation processing on the handwriting data with signature handwriting parameter features. S2. A baseline model is constructed based on the interpolated handwriting data. By constructing a feature space and a graph structure, a new feature space is obtained, and the uniqueness of the user's signature is captured to obtain signature handwriting data representing the user's unique style. The baseline model is specifically implemented as follows: The first step is feature transformation; the interpolated handwriting data is transformed into temporal flow data, and then the temporal flow data is further transformed into the spectral domain. The second step is pattern recognition and feature extraction. Feature vectors are extracted from the spectral data in the spectral domain, and feature selection and feature fusion are further performed. Dimensionality reduction is then carried out using a dimensionality reducer. The third step is to construct a baseline model. Based on the dimensionality-reduced feature vectors, a feature space is constructed to obtain the set of data points in the feature space, and a similarity matrix is ​​calculated. Based on the similarity matrix, a graph structure is constructed, and a new feature space is generated. In the new feature space, clustering is used to capture the uniqueness of user signatures. S3. Compress and encrypt the user's unique signature handwriting data, and integrate it into a PDF document; S4. Recover and track the signature handwriting data of the integrated PDF document, and verify its tamper-proof nature using hash functions and differential algorithms.

2. The PDF signature handwriting verification and storage method according to claim 1, characterized in that, S1 specifically includes: The system acquires user signature handwriting data and performs initial smoothing by averaging the acquired signature handwriting data using the moving average method. Then, it applies discrete Gaussian filtering for secondary smoothing to obtain smoothed data. The smoothed data is further subjected to preliminary dynamic analysis using a dynamic handwriting analysis model. At the same time, Tukey's fence method is used for data cleaning. Finally, Bézier curves are used for interpolation to obtain continuous signature handwriting data.

3. The PDF signature handwriting verification and storage method according to claim 2, characterized in that, The S1 also includes: The dynamic handwriting analysis model is specifically constructed as follows: in, It is a smoothed collection of handwriting data, containing information about each point in the user's signature; It is a set of coordinate differences; It is a set of time differences; It is a set of velocities; It is a set of direction angles; It is the set of cumulative curve lengths; It is a set of pressure-weighted speeds for handwriting; It is a measure of signature complexity; It is the output of the model.

4. The PDF signature handwriting verification and storage method according to claim 1, characterized in that, S4 specifically includes: To perform tamper-proof verification, firstly, a unique hash value is generated from the encrypted signature handwriting data using a hash function and stored. When it is necessary to verify whether the signature handwriting data has been tampered with, the signature handwriting verification and storage system extracts the encrypted signature handwriting data from the integrated PDF document, uses a hash function to generate a hash value again from the extracted encrypted signature handwriting data, and compares the newly generated hash value with the original stored hash value for verification.

5. The PDF signature handwriting verification and storage method according to claim 4, characterized in that, The S4 also includes: When the hash verification is inconsistent, the signature handwriting verification storage system prompts "the signature handwriting data has been tampered with" and initiates a recovery mechanism to recover the signature handwriting data through a differential algorithm and to track the recovered signature handwriting data; when the hash verification is consistent, the tracking is then performed.

6. A PDF signature handwriting verification and storage system, applied to the PDF signature handwriting verification and storage method as described in claim 1, characterized in that, Includes the following parts: Handwriting capture module, handwriting analysis module, compression and storage module, encryption and security module, anti-tampering verification module, integration module, intelligent recovery and tracking module; The handwriting capture module captures the handwriting data of the user's signature through sensors, including coordinates, pressure, and speed information, processes the handwriting data, and then transmits the processed handwriting data to the handwriting analysis module. The handwriting analysis module analyzes the handwriting data and transmits the analyzed handwriting data to the compression and storage module. The compression and storage module uses a compression algorithm to compress the signature handwriting data and then transmits the compressed signature handwriting data to the encryption and security module. The encryption and security module encrypts the compressed signature handwriting data, provides encryption key management and storage, and transmits the encrypted signature handwriting data to the anti-tampering verification module and the integration module. The integration module is used to integrate encrypted signature handwriting data with other content of the PDF document, including specifying the signature position, style and display method, supporting user-defined signature integration options to meet the needs of different scenarios, and passing the integrated PDF document to the anti-tampering verification module. The anti-tampering verification module uses a hash function to generate a unique hash value for the encrypted signature handwriting data and stores it. When it is necessary to verify whether the signature handwriting data has been tampered with, the signature handwriting verification and storage system extracts the encrypted signature handwriting data from the integrated PDF document, uses a hash function to generate a hash value for the extracted encrypted signature handwriting data again, compares the newly generated hash value with the original stored hash value for verification, and passes the verification result to the intelligent recovery and tracking module. The intelligent recovery and tracking module uses intelligent algorithms to track and analyze changes in signature handwriting data, restores it to its original state, and provides users with the function of tracking and retrieving historical signature records.

Images