Block chain evidence storage method and device for electronic paper marking

Abstract

The invention discloses a block chain evidence storage method and device for electronic paper marking, and relates to the technical field of information technology and education, and the method comprises the steps: receiving a test paper scanning image of a test paper and an answer sheet scanning image of an answer sheet, extracting key feature points in the test paper scanning image and the answer sheet scanning image, and storing the key feature points in the test paper scanning image and the answer sheet scanning image; establishing a mapping relationship between the test paper and the answer sheet; encrypting the original data of the test paper and the answer sheet based on a Hash encryption algorithm to generate a unique identifier, and recording the scoring behavior of the answer sheet to obtain scoring process data; and detecting the scoring process data, when it is detected that the scoring process data is an abnormal paper marking behavior, triggering a block chain evidence storage mechanism, recording context information of the abnormal paper marking behavior, and generating an evidence storage record which cannot be tampered. By means of the mode, comprehensive behavior evidence storage in the paper marking process can be achieved, effective supervision and data non-tampering performance are achieved, and paper marking transparency and the intelligent level are improved.

Description

Technical Field

[0001] This application relates to the fields of information technology and education technology, and in particular to a blockchain-based evidence storage method and apparatus for electronic exam marking. Background Technology

[0002] With the rapid development of smart education, the marking process is gradually shifting towards electronic marking. In electronic marking, OCR technology is typically used to scan the exam paper and answer sheet to obtain the corresponding question and answer information for electronic marking. To ensure fairness and impartiality, the marking process needs to be recorded. Currently, recording marking behavior only involves verifying the marker's identity before the marking process begins, usually by verifying the teacher's signature. However, this method cannot provide evidence of the entire marking process, nor can it effectively supervise actions taken during marking, potentially leading to violations such as misconduct.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of this application is to provide a blockchain-based evidence storage method and device for electronic document review, aiming to solve the technical problem of the lack of full recording and effective supervision of document review behavior in the prior art.

[0005] To achieve the above objectives, this application provides a blockchain-based evidence storage method for electronic document review, the method comprising: Receive scanned images of the test paper and the answer sheet, extract key feature points from the scanned images of the test paper and the answer sheet, and establish a mapping relationship between the test paper and the answer sheet; The original data of the test paper and answer sheet are encrypted using a hash encryption algorithm to generate a unique identifier, and the scoring behavior of the answer sheet is recorded to obtain the scoring process data. The scoring process data is inspected. When abnormal marking behavior is detected, a blockchain evidence storage mechanism is triggered to record the context information of the abnormal marking behavior and generate an immutable evidence storage record.

[0006] In one embodiment, the step of establishing the mapping relationship between the test paper and the answer sheet includes: Determine the geometric positional relationship between the question area in the scanned image of the test paper and the answer area in the scanned image of the answer sheet; Based on the geometric positional relationship, key feature points of the question region and the answer region are extracted respectively using an edge detection algorithm; The key feature points are spatially matched to generate a feature point matching matrix containing the feature point mapping relationship. The similarity value between the question region and the answer region is calculated based on the feature point matching matrix. When the similarity value exceeds a preset threshold, the affine transformation matrix parameters are determined based on the similarity value and the feature point matching matrix. Generate the mapping equation from the question region to the answer region based on the affine transformation matrix parameters; A spatial mapping relationship between the test paper and the answer sheet is established based on the mapping equation.

[0007] In one embodiment, the step of encrypting the original data of the test paper and answer sheet based on a hash encryption algorithm includes: Extract the raw data from the test paper and answer sheet, the raw data including image pixel values ​​and text content, and generate a raw data sequence based on the raw data; The original data sequence is hashed using the SHA-256 algorithm to generate a fixed-length hash value. The hash value is used as a unique identifier, and a unique time identifier is obtained by appending timestamp information to the unique identifier. Based on the unique time identifier, an encryption key is generated using an asymmetric encryption algorithm, and the original data sequence is encrypted using the encryption key; The encrypted original data sequence and the unique time identifier are broadcast to a distributed node network, so that each node can verify the logical correlation between the timestamp information and the hash value based on a consensus algorithm and provide feedback. When the number of nodes that pass verification reaches a preset threshold, the unique time identifier and encrypted data are written into an immutable block of the blockchain to complete the encryption process.

[0008] In one embodiment, the step of triggering a blockchain evidence storage mechanism to record the context information of the abnormal grading behavior and generate an immutable evidence storage record when the grading process data is detected as abnormal includes: The scoring operations during the marking process are monitored in real time, and the behavioral characteristics of the scoring operations are extracted. These behavioral characteristics include the scoring time interval, the scoring consistency index, and the scoring deviation value. When the scoring consistency index is lower than a preset threshold or the scoring deviation value is higher than a preset threshold, the scoring operation is determined to be an abnormal marking behavior. The blockchain evidence storage mechanism is triggered to generate an immutable evidence record from the context information of the abnormal marking behavior. The context information includes the timestamp of the marking operation, the identity information of the marker, and the marking result.

[0009] In one embodiment, the step of generating a personalized error report based on the grading markers and the error information includes: Extract incorrect answer data from the incorrect question information, and construct an incorrect question feature vector based on the incorrect answer data. The incorrect question feature vector includes question difficulty coefficient, error type, and error frequency. The K-means clustering algorithm is used to perform cluster analysis on the question difficulty coefficient, the error type, and the error frequency. The clustering results are obtained based on the clusters, cluster centroids, and the feature vectors of the incorrect questions. Based on the clustering results, incorrect questions are categorized and labeled. Then, based on the unique identifier, the original question text, correct answer, and student's response are extracted from a pre-set question database. An error classification description is generated based on the error types in the error feature vector. Based on the knowledge dimensions corresponding to the hierarchical labels, learning strategy suggestions are generated; A personalized error report is generated based on the original question, the correct answer, the student's answer, the error classification explanation, and learning strategy suggestions.

[0010] In one embodiment, before the step of receiving the scanned images of the exam paper and the answer sheet, the method further includes: Determine standardized templates for exam papers and answer sheets, and generate template verification rules; Based on the template verification rules, the scanned images of the test paper and answer sheet are verified for format to ensure that the image quality meets the marking requirements. Noise suppression is applied to the scanned images of the verified test papers and answer sheets to generate denoised scanned images; Dynamic thresholding is performed on the denoised scanned image, and the image pixels in the denoised image are binarized to obtain a binarized image. A Hough transform is performed on the binarized image to determine the baseline tilt angle of the binarized image. Based on the tilt angle, the test paper and the answer sheet are rotated and corrected so that the test paper and the answer sheet are aligned with the target baseline.

[0011] In one embodiment, the blockchain-based evidence storage method for electronic document review further includes: A visual interface for the marking progress is generated based on the query interface of distributed storage nodes. Automatically assign marking tasks through smart contracts; After the marking is completed, a marking result report is generated based on the blockchain network and sent to the relevant parties through an encrypted channel.

[0012] In one embodiment, the step of automatically assigning marking tasks via smart contracts includes: Based on the encrypted test paper data and examiner qualification information stored in the blockchain, task allocation parameters are defined in the smart contract. The task allocation parameters include the question group identifier to be assigned, the examiner skill weight threshold, and the maximum task allocation time. In response to the completion of the test paper encryption event, the smart contract is triggered to filter the set of examiner nodes that meet the qualification conditions according to the task allocation parameters; The question groups in the encrypted test paper data are dynamically associated with the examiner nodes to generate a task allocation record containing question group identifiers, allocation node addresses and timestamps; The task allocation records are written into the distributed ledger of the blockchain, and the integrity and permission compliance of the allocation records are verified by consensus nodes. Once the verification is successful, a task execution instruction containing encrypted question group data and marking rules is sent to the corresponding marking personnel node, and the question group is marked as assigned in the blockchain. The system monitors the progress of marking tasks in real time. If a task times out or a node is detected as abnormal, the smart contract's automatic recycling mechanism is triggered to redistribute incomplete question sets.

[0013] Furthermore, to achieve the above objectives, this application also proposes a blockchain-based evidence storage device for electronic document review, which includes: The mapping relationship establishment module is used to receive scanned images of the test paper and the answer sheet, extract key feature points from the scanned images of the test paper and the answer sheet, and establish a mapping relationship between the test paper and the answer sheet. The data encryption module is used to encrypt the original data of the test paper and answer sheet based on the hash encryption algorithm, generate a unique identifier, record the scoring behavior of the answer sheet, and obtain the scoring process data. An anomaly monitoring module is used to detect the scoring process data. When an abnormal marking behavior is detected in the scoring process data, a blockchain evidence storage mechanism is triggered to record the context information of the abnormal marking behavior and generate an immutable evidence storage record.

[0014] In addition, to achieve the above objectives, this application also proposes a blockchain-based evidence storage device for electronic document review, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. The computer program is configured to implement the steps of the blockchain-based evidence storage method for electronic document review as described above.

[0015] In addition, to achieve the above objectives, the present invention also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the blockchain evidence storage method for electronic document review as described above.

[0016] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the blockchain-based electronic document review method described above.

[0017] This application provides a blockchain-based evidence storage method for electronic marking. By receiving scanned images of the exam paper and answer sheet, key feature points are extracted from both images to establish a mapping relationship between them. The original data of the exam paper and answer sheet are encrypted using a hash encryption algorithm to generate a unique identifier, recording the scoring behavior on the answer sheet to obtain scoring process data. This scoring process data is then monitored; if abnormal marking behavior is detected, a blockchain evidence storage mechanism is triggered, recording the context information of the abnormal marking behavior and generating an immutable evidence record. This method achieves comprehensive behavioral evidence storage, effective supervision, and data immutability throughout the marking process, improving the transparency and intelligence of the marking process. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an embodiment of the blockchain-based evidence storage method for electronic document review in this application. Figure 2 This is a schematic diagram of the process of assigning review tasks via smart contract in an embodiment of the blockchain-based electronic document review method of this application. Figure 3 This is a schematic diagram of the module structure of the blockchain evidence storage device for electronic document review in this application embodiment; Figure 4 This is a schematic diagram of the hardware operating environment involved in the blockchain-based evidence storage method for electronic document review in this application embodiment.

[0021] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The main solution of this application embodiment is as follows: receiving scanned images of the test paper and the answer sheet; extracting key feature points from the scanned images of the test paper and the answer sheet; establishing a mapping relationship between the test paper and the answer sheet; encrypting the original data of the test paper and the answer sheet based on a hash encryption algorithm to generate a unique identifier; recording the scoring behavior of the answer sheet to obtain scoring process data; detecting the scoring process data; and triggering a blockchain evidence storage mechanism when abnormal scoring behavior is detected, recording the context information of the abnormal scoring behavior and generating an immutable evidence storage record.

[0025] Currently, with the rapid development of smart education, the marking process is gradually shifting towards electronic marking. In electronic marking, OCR technology is typically used to scan the exam paper and answer sheet to obtain the corresponding question and answer information for electronic marking. To ensure fairness and impartiality in this marking process, the marking actions need to be recorded. Currently, recording marking behavior only involves verifying the marker's identity before the marking process begins, usually by verifying the teacher's signature, thus achieving the purpose of recording marking behavior. However, this method cannot provide evidence of the entire marking process, nor can it effectively supervise the actions taken during the marking period, leaving the possibility of unauthorized marking practices.

[0026] This application provides a solution that receives scanned images of exam papers and answer sheets, extracts key feature points from both scanned images, and establishes a mapping relationship between the exam papers and answer sheets. The original data of the exam papers and answer sheets are encrypted using a hash encryption algorithm to generate a unique identifier, and the scoring behavior on the answer sheets is recorded to obtain scoring process data. The scoring process data is then monitored; if abnormal scoring behavior is detected, a blockchain-based evidence storage mechanism is triggered to record the context information of the abnormal scoring behavior and generate an immutable evidence record. Through this method, comprehensive behavioral evidence storage, effective supervision, and data immutability of the scoring process can be achieved, improving the transparency and intelligence of the scoring process.

[0027] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, such as a blockchain-based evidence storage device for electronic document review. This embodiment does not specifically limit the scope of the embodiment. The following description uses a blockchain-based evidence storage device for electronic document review as an example to illustrate this embodiment and the subsequent embodiments.

[0028] This application provides a blockchain-based evidence storage method for electronic document review, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the blockchain-based evidence storage method for electronic document review in this application.

[0029] In this embodiment, the blockchain-based evidence storage method for electronic document review includes steps S10 to S30: Step S10: Receive scanned images of the test paper and the answer sheet, extract key feature points from the scanned images of the test paper and the answer sheet, and establish a mapping relationship between the test paper and the answer sheet.

[0030] It should be noted that both the scanned images of the exam paper and the scanned images of the answer sheet are electronic image files converted from paper content using a scanner or document scanner. They typically include handwritten answers, text-based questions, or subjective questions. Key feature points in the scanned images of the exam paper and the answer sheet refer to visually distinctive elements in digital image processing, primarily used for image matching and spatial positioning. In this embodiment, the feature points on the exam paper side include at least the layout structure features such as the coordinates of the question number area and the position of the separator lines; identifying elements such as QR codes / barcodes and page number identifiers; geometric reference points such as corner positioning marks and edge alignment lines; and content features such as the position of the first character of the question stem and the coordinates of special symbols. The feature points on the answer sheet side include at least the coordinate matrix of the multiple-choice question option boxes, the OCR positioning points of the candidate information area, synchronous positioning marks, the starting point of the handwritten handwriting, and the centroid of the filling area. The mapping relationship refers to the correspondence between the feature points of the scanned images of the exam paper and the answer sheet.

[0031] Understandably, the scanning process for exam paper scans and answer sheet scans can be roughly broken down into scanning, positioning and alignment, image segmentation, file naming / numbering, data extraction, and uploading to the marking system or storing on a server.

[0032] In the specific implementation, after receiving the scanned images of the exam paper and the answer sheet, preprocessing operations can be performed on these images. After preprocessing, an improved ORB algorithm can be used for dual-channel feature detection. For feature detection on the exam paper side, the FAST (Features from Accelerated Segment Test) algorithm can be used to accurately locate the corner points of the question area on the exam paper. The corner point position is quickly determined by comparing the intensity difference between a pixel and its neighboring pixels. Based on the formula... Hough line detection is performed to transform the text baselines in the image to the Hough parameter space, enabling the extraction of text baseline features and thus determining the text layout direction and positional relationships. For the positioning pattern characteristics of QR codes, a specialized detection algorithm is employed to accurately locate the QR code vertices.

[0033] For feature detection on the answer sheet side, template matching technology can be used to match a predefined L-shaped corner mark template with the answer sheet image to determine the specific position and orientation of the answer sheet in the image. Morphological operations are used to process the coated area, and the centroid of the connected region is calculated as the feature point of the coated area, which reflects the position of the student's answer mark. A fixed threshold is used to binarize the OMR (Optical Mark Recognition) region, dividing the image pixels into foreground and background, highlighting the position and shape of the recognition.

[0034] Centered on a feature point, pixel pairs within a certain radius are selected for comparison, generating a 256-bit binary feature vector as the BRIEF descriptor. By constructing a multi-layer feature pyramid, features at different scales of the image are extracted and described. During feature matching, efficient matching can be performed between the feature descriptors on the exam paper and answer sheet sides to find corresponding feature point pairs. The specific implementation method is as follows:

[0035] in, It is a homography matrix. and To match point pairs.

[0036] Then, using the random sampling consensus algorithm, based on the number of iterations... (in, The probability of failure. For the probability of success, , Iterate through the process to remove mismatched feature point pairs.

[0037] The homography matrix Capable of describing the geometric transformation relationship between the exam paper and the answer sheet, homography matrix It is possible to solve linear equation systems using singular value decomposition, and the specific calculation formula is as follows:

[0038] When establishing the mapping relationship between the test paper and the answer sheet, the homography matrix can be used. And the formula:

[0039] This establishes a two-way coordinate mapping relationship between the test paper and the answer sheet, enabling the mutual conversion of image coordinate points between the two.

[0040] In one feasible implementation, prior to the step of receiving the scanned images of the exam paper and the answer sheet, the method further includes: Determine standardized templates for exam papers and answer sheets, and generate template verification rules; Based on the template verification rules, the scanned images of the test paper and answer sheet are verified for format to ensure that the image quality meets the marking requirements. Noise suppression is applied to the scanned images of the verified test papers and answer sheets to generate denoised scanned images; Dynamic thresholding is performed on the denoised scanned image, and the image pixels in the denoised image are binarized to obtain a binarized image. A Hough transform is performed on the binarized image to determine the baseline tilt angle of the binarized image. Based on the tilt angle, the test paper and the answer sheet are rotated and corrected so that the test paper and the answer sheet are aligned with the target baseline.

[0041] It should be noted that the template verification rules are a verification mechanism to ensure that the exam paper and answer sheet conform to the standardized design, including three levels of checks: geometric verification, logical verification, and dynamic verification. Failure to verify will trigger a tiered alarm or adaptive correction.

[0042] In the specific implementation, the standardized template generation module determines the format verification rules for the exam paper and answer sheet to ensure that the scanned images meet the marking requirements. This step includes format verification of the scanned images based on the template verification rules, and noise suppression processing of the verified images to generate denoised scanned images. Since the main noise in exam paper images is Gaussian noise and ink diffusion noise, and the common noise in answer sheet images is impulse noise and quantization noise, a hybrid filtering strategy can be used for noise suppression processing, specifically:

[0043] in, The filtered image. For adaptive Wiener filtering, This is a morphological opening and closing operation. This is the noise intensity weighting coefficient.

[0044] During denoising, the scanned exam paper image can be decomposed into a 3-layer wavelet decomposition, and then processed based on an improved Bayesian threshold to obtain a denoised image. The improved Bayesian threshold... for:

[0045] in, Let be the standard deviation of the detail coefficients at level j. This represents the total number of floors.

[0046] When denoising scanned images of answer sheets, a noise probability model can be established. Specifically, it can be expressed as:

[0047] in, The mean of the 5×5 neighborhood.

[0048] Based on this, an improved median filter is used for filtering, and the homomorphic window size is adjusted during the filtering process. for:

[0049] in, The number of noise points detected. The target number of noise points.

[0050] Then, the burrs in the filling are eliminated using a closing operation to obtain the denoised scanned image of the answer sheet. The specific processing procedure is as follows:

[0051] B is the structural element, and a 3×3 cross shape is selected.

[0052] Subsequently, the denoised image pixels are binarized using dynamic thresholding segmentation technology, based on Gaussian weighting, and the dynamic threshold is determined according to the formula. Among them, dynamic threshold The calculation method is as follows:

[0053] in, It is the Gaussian weighted mean of the pixels within the local window. The adjustment constant is set empirically.

[0054] The formula for binarization is:

[0055] After dynamic thresholding, a clear binary image is obtained. Further, the tilt angle of the baseline in the binary image is detected using Hough transform. Based on the tilt angle, the scanned images of the exam paper and answer sheet are rotated and corrected to align with the target baseline. Specifically, Canny edge detection is used to enhance the baseline features of the clear binary image obtained after dynamic thresholding, based on the polar coordinate equation of Hough transform. Detecting straight lines, where The distance from the straight line to the origin. The angle between the line and the x-axis. These are the image space coordinates. Filtering angles within... Within the range of baselines, calculate the average tilt angle. Generate rotation matrix The image is rotated using an affine transformation to align the baseline with the horizontal axis. This process ensures the accuracy of subsequent feature point extraction.

[0056] Rotation matrix for:

[0057] in, center point The center of the image.

[0058] In one feasible implementation, the step of establishing the mapping relationship between the test paper and the answer sheet includes: Determine the geometric positional relationship between the question area in the scanned image of the test paper and the answer area in the scanned image of the answer sheet; Based on the geometric positional relationship, key feature points of the question region and the answer region are extracted respectively using an edge detection algorithm; The key feature points are spatially matched to generate a feature point matching matrix containing the feature point mapping relationship. The similarity value between the question region and the answer region is calculated based on the feature point matching matrix. When the similarity value exceeds a preset threshold, the affine transformation matrix parameters are determined based on the similarity value and the feature point matching matrix. Generate the mapping equation from the question region to the answer region based on the affine transformation matrix parameters; A spatial mapping relationship between the test paper and the answer sheet is established based on the mapping equation.

[0059] It should be noted that the affine transformation parameters are a set of parameters determined by optimizing the residuals using the least squares method. These parameters are used to construct the mapping equation between the question region and the answer region. The mapping equation is the mathematical expression of the correspondence between the question region and the answer region.

[0060] In the specific implementation, the question area in the scanned image of the exam paper is determined. Answer area in the scanned image of the answer sheet When determining the geometric positional relationship between them, it is assumed that the two can be associated through an affine transformation model, i.e., a mapping relationship exists:

[0061] in, , ,parameter Control rotation, scaling, and shear transformations. This represents the translation amount.

[0062] The contour information of the two regions is extracted using an edge detection algorithm, and corner detection is performed using Harris corner detection based on the response function. Extracting the set of key feature points and .

[0063]

[0064] in, , For image gradient, Gaussian window The size is .

[0065] The initial matching set is generated by selecting matching pairs that satisfy the minimum distance ratio condition. And calculate matching similarity. Among them, matching similarity Represented as:

[0066] like hour, The affine transformation parameters are solved using the least squares method based on matching point pairs. The optimization objective is:

[0067] Solve the system of linear equations to obtain the optimal parameters. Substitute the parameters into the affine transformation equations to establish... arrive mapping relationship And through bilinear interpolation , By calculating the distance between adjacent pixels, pixel-level alignment of the image is achieved, thereby accurately associating the spatial coordinates of the question and answer areas.

[0068] Step S20: Encrypt the original data of the test paper and answer sheet based on the hash encryption algorithm to generate a unique identifier, record the scoring behavior of the answer sheet, and obtain the scoring process data; It should be noted that the unique identifier is a special marker used to annotate the original data. The scoring process data includes the operational data in the marking process, including marking actions, reading records, and marking results.

[0069] Understandably, during the grading process, the original data is encrypted using the SHA-256 hash encryption algorithm to generate a unique identifier. and By leveraging the collision resistance and irreversibility of hashing to ensure data uniqueness and privacy, each operation during the scoring process is recorded using a hash chain. Initial chain head binding This ensures the immutability of operational history. Data integrity is verified by comparing real-time hash values ​​with stored values, and non-repudiation auditing is achieved through digital signatures and compliance algorithm selection, ultimately meeting the requirements for secure storage, transparent traceability, and compliance of educational data.

[0070] In one feasible embodiment, the step of encrypting the original data of the test paper and answer sheet based on the hash encryption algorithm includes: Extract the raw data from the test paper and answer sheet, the raw data including image pixel values ​​and text content, and generate a raw data sequence based on the raw data; The original data sequence is hashed using the SHA-256 algorithm to generate a fixed-length hash value. The hash value is used as a unique identifier, and a unique time identifier is obtained by appending timestamp information to the unique identifier. Based on the unique time identifier, an encryption key is generated using an asymmetric encryption algorithm, and the original data sequence is encrypted using the encryption key; The encrypted original data sequence and the unique time identifier are broadcast to a distributed node network, so that each node can verify the logical correlation between the timestamp information and the hash value based on a consensus algorithm and provide feedback. When the number of nodes that pass verification reaches a preset threshold, the unique time identifier and encrypted data are written into an immutable block of the blockchain to complete the encryption process.

[0071] In the specific implementation, when encrypting the original data of the exam paper and answer sheet, a fixed-length hash value is generated by hashing the original data sequence based on the SHA-256 algorithm. Then, a timestamp is concatenated to generate unique time identifiers (UIDs) for the exam paper and answer sheet, respectively. Using the RSA-2048 asymmetric encryption algorithm, a key pair is dynamically generated based on the UID. A temporary key midpoint is generated using the UID and salt value. A public key (PK) and a private key (SK) are then generated using elliptic curve cryptography. The public key is used to encrypt the original data, forming an encrypted data packet. This data packet is broadcast to a distributed node network. Each node verifies the logical correlation using a consensus algorithm, recalculates the hash value locally, verifies the recalculated hash value, and checks the validity of the timestamp. Upon successful verification, a confirmation signal is sent. When a preset threshold condition is met, the encrypted data packet is written into a new block of the blockchain, forming an immutable block, thus completing the data encryption process.

[0072] Step S30: Detect the scoring process data. When abnormal marking behavior is detected in the scoring process data, trigger the blockchain evidence storage mechanism to record the context information of the abnormal marking behavior and generate an immutable evidence storage record. It should be noted that abnormal marking behavior includes violations such as data tampering, answer replacement, and log tampering, as well as abnormal behavior such as excessively fast marking speed.

[0073] Understandably, in order to achieve real-time detection and reliable evidence preservation of abnormal behavior during the marking process, the marking process data is first subjected to multi-dimensional anomaly detection through predefined rules. When an anomaly is detected, a smart contract is immediately triggered to extract the anomaly context information and generate an encrypted evidence preservation package through asymmetric encryption. Based on the Practical Byzantine Fault Tolerance (PBFT) consensus protocol, the evidence preservation request is broadcast to the blockchain node network. After each node verifies the timestamp logic and data integrity, if the number of verified nodes meets the verification conditions, the evidence preservation data is written into a new block according to the Merkle tree structure, generating a unique on-chain identifier and automatically associating the examiner's digital certificate and historical behavior trajectory through a smart contract, forming an immutable audit chain to ensure that it is traceable afterward and that judicial evidence is effective.

[0074] In one feasible implementation, the step of triggering a blockchain evidence storage mechanism when abnormal marking behavior is detected in the marking process data, recording the context information of the abnormal marking behavior, and generating an immutable evidence storage record includes: The scoring operations during the marking process are monitored in real time, and the behavioral characteristics of the scoring operations are extracted. These behavioral characteristics include the scoring time interval, the scoring consistency index, and the scoring deviation value. When the scoring consistency index is lower than a preset threshold or the scoring deviation value is higher than a preset threshold, the scoring operation is determined to be an abnormal marking behavior. The blockchain evidence storage mechanism is triggered to generate an immutable evidence record from the context information of the abnormal marking behavior. The context information includes the timestamp of the marking operation, the identity information of the marker, and the marking result.

[0075] In its implementation, the scoring process is monitored in real time, recording the scoring time interval, scoring consistency index, and scoring deviation value. The scoring time interval is determined based on the time difference between scoring operations; if the time difference between two scoring operations is less than a preset time interval, the current scoring operation is marked as hasty. The scoring consistency index measures the stability of scoring within the same question group. , ,in, It is the standard deviation of the rating. It's an average score. If stability is measured... If the score is less than the preset value, it indicates a discrepancy in the scoring criteria. The scoring deviation value refers to the degree of deviation between the grader's score and the overall average score, denoted as _____. .

[0076]

[0077] in, The rater For the question The rating, For the title The overall average score.

[0078] like If the value exceeds the preset value, it is determined to be an abnormal deviation.

[0079] When the scoring consistency index falls below a preset threshold or the scoring deviation exceeds a preset threshold, the scoring operation is determined to be an abnormal marking behavior. At this time, the blockchain evidence storage mechanism can be triggered to generate an immutable evidence record of the abnormal marking behavior, whereby the context information includes the timestamp of the scoring operation, the identity information of the scorer, and the scoring result.

[0080] In one feasible implementation, the blockchain-based evidence storage method for electronic document review further includes: Incorrect questions are graded and marked based on a multi-dimensional clustering algorithm; The incorrect question information is determined based on the unique identifier; A personalized error report is generated based on the graded tags and the error information.

[0081] It should be noted that the graded marking is used to annotate incorrect questions, and is a special mark used to describe the type of incorrect question, such as the difficulty level of the question. Incorrect question information includes the question stem, the answer, and statistical information about the error.

[0082] In the specific implementation, incorrect question information is generated into an incorrect question set, which can be represented as: , The title content, Tags for knowledge points For error types (such as calculation errors, conceptual misunderstandings). Time allotted for answering questions. Error frequency is defined. A multi-dimensional vector is constructed based on the knowledge point relevance, error type weight, time efficiency coefficient, and error frequency score. Specifically, it is represented as follows: Knowledge point relevance ; Error type weight ; Time efficiency coefficient , This represents the average time to answer similar questions. Error frequency score

[0083] DBSCAN (density clustering) was used to analyze the standardized feature matrix. Perform clustering and generate cluster labels. The grading is based on the average error frequency within the cluster and the importance of the knowledge point. Each incorrect question has a unique identifier, so the specific incorrect question information can be determined based on the unique identifier. Therefore, when generating a personalized incorrect question report, the unique identifier can be used directly to determine the incorrect question information and the resulting grading label to generate a personalized incorrect question report.

[0084] In one feasible implementation, the step of generating a personalized error report based on the grading markers and the error information includes: Extract incorrect answer data from the incorrect question information, and construct an incorrect question feature vector based on the incorrect answer data. The incorrect question feature vector includes question difficulty coefficient, error type, and error frequency. The K-means clustering algorithm is used to perform cluster analysis on the question difficulty coefficient, the error type, and the error frequency. The clustering results are obtained based on the clusters, cluster centroids, and the feature vectors of the incorrect questions. Based on the clustering results, incorrect questions are classified and marked, and the original text of the question, the correct answer, and the student's answer are extracted from the preset question database based on the unique identifier. Generate an error classification description based on the error types in the aforementioned error feature vector; Based on the knowledge dimensions corresponding to the hierarchical labels, learning strategy suggestions are generated; A personalized error report is generated based on the original question, the correct answer, the student's answer, the error classification explanation, and learning strategy suggestions.

[0085] In the specific implementation, incorrect answer data is extracted from the incorrect question information, and an incorrect question feature vector is constructed based on the incorrect answer data. The incorrect question feature vector includes the question difficulty coefficient. Error Types and error frequency Therefore, the eigenvector can be represented as , These can represent conceptual errors, calculation errors, and logical errors, respectively. Based on the K-means clustering algorithm, cluster analysis is performed on the question difficulty coefficient, the error type, and the error frequency. Clustering results are obtained based on the clusters, cluster centroids, and the error feature vectors. Errors are then categorized and labeled according to the clustering results. Based on the unique identifier, the original question text, correct answer, and student response are extracted from a pre-set question database. An error classification description is generated based on the error type in the error feature vector. Learning strategy suggestions are generated based on the knowledge dimension corresponding to the categorization label. A personalized error report is generated based on the original question text, the correct answer, the student response, the error classification description, and the learning strategy suggestions.

[0086] In one feasible implementation, the blockchain-based evidence storage method for electronic document review further includes: A visual interface for the marking progress is generated based on the query interface of distributed storage nodes. Automatically assign marking tasks through smart contracts; After the marking is completed, a marking result report is generated based on the blockchain network and sent to the relevant parties through an encrypted channel.

[0087] It should be noted that during the marking process, the progress is displayed visually to facilitate tracking. Specifically, each distributed storage node has a built-in query interface to obtain the marking status in real time. Simultaneously, upon obtaining the marking progress, marking tasks are automatically allocated according to smart contracts to ensure collaborative completion of the marking tasks. After marking is completed, a marking result report is generated in the blockchain network. This report includes a summary of the marking process for that distributed node and a compilation of the marking results. The marking result report is then sent to relevant parties via an encrypted channel, including administrators, grade-level managers, and school administrators.

[0088] In one feasible implementation, the steps of automatically assigning marking tasks via smart contracts include: Based on the encrypted test paper data and examiner qualification information stored in the blockchain, task allocation parameters are defined in the smart contract. The task allocation parameters include the question group identifier to be assigned, the examiner skill weight threshold, and the maximum task allocation time. In response to the completion of the test paper encryption event, the smart contract is triggered to filter the set of examiner nodes that meet the qualification conditions according to the task allocation parameters; The question groups in the encrypted test paper data are dynamically associated with the examiner nodes to generate a task allocation record containing question group identifiers, allocation node addresses and timestamps; The task allocation records are written into the distributed ledger of the blockchain, and the integrity and permission compliance of the allocation records are verified by consensus nodes. Once the verification is successful, a task execution instruction containing encrypted question group data and marking rules is sent to the corresponding marking personnel node, and the question group is marked as assigned in the blockchain. The system monitors the progress of marking tasks in real time. If a task times out or a node is detected as abnormal, the smart contract's automatic recycling mechanism is triggered to redistribute incomplete question sets.

[0089] In the specific implementation, refer to Figure 2 , Figure 2This is a flowchart illustrating the process of assigning marking tasks to a smart contract. The blockchain stores encrypted exam paper data and marker qualification information. The smart contract stores task allocation parameters, including the assigned question group identifier, the marker skill weight threshold, and the maximum task allocation time. After the exam paper encryption event is completed, the smart contract is triggered, allowing markers who meet the qualification requirements to be selected based on the task parameters. Since the markers' network access information is fixed during marking, the set of marker nodes can be obtained. Then, each question group in the encrypted exam paper data is dynamically associated with the marker nodes, generating a task allocation record containing the question group identifier, the allocation node address, and a timestamp. This task allocation record is written to the blockchain's distributed ledger, and the integrity and permission compliance of the allocation record are verified by consensus nodes. Upon successful verification, a task execution instruction containing the encrypted question group data and marking rules is sent to the corresponding marker node, and the question group identifier is marked as assigned in the blockchain. The marking task execution progress is monitored in real time. If a task timeout or node anomaly is detected, the smart contract's automatic recycling mechanism is triggered to reallocate unfinished question groups.

[0090] This embodiment provides a blockchain-based evidence storage method for electronic marking. It receives scanned images of the exam paper and answer sheet, extracts key feature points from both images, and establishes a mapping relationship between them. The original data of the exam paper and answer sheet are encrypted using a hash encryption algorithm to generate a unique identifier. The scoring behavior on the answer sheet is recorded to obtain scoring process data. This scoring process data is then monitored. When abnormal marking behavior is detected, a blockchain evidence storage mechanism is triggered, recording the context information of the abnormal marking behavior and generating an immutable evidence record. This method achieves comprehensive behavioral evidence storage, effective supervision, and data immutability throughout the marking process, improving the transparency and intelligence of the marking process.

[0091] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the blockchain-based evidence storage method for electronic document review in this application. Any simple modifications based on this technical concept are within the scope of protection of this application.

[0092] This application also provides a blockchain-based evidence storage device for electronic document review; please refer to [link / reference]. Figure 3 The blockchain-based evidence storage device for electronic document review includes: The mapping relationship establishment module 10 is used to receive the scanned image of the test paper and the scanned image of the answer sheet, extract the key feature points in the scanned image of the test paper and the scanned image of the answer sheet, and establish the mapping relationship between the test paper and the answer sheet; The data encryption module 20 is used to encrypt the original data of the test paper and answer sheet based on the hash encryption algorithm, generate a unique identifier, record the scoring behavior of the answer sheet, and obtain the scoring process data. The anomaly monitoring module 30 is used to detect the scoring process data. When the scoring process data is detected as abnormal marking behavior, the blockchain evidence storage mechanism is triggered to record the context information of the abnormal marking behavior and generate an immutable evidence storage record.

[0093] In one feasible implementation, the mapping relationship establishment module 10 is further used to determine the geometric positional relationship between the question area in the scanned image of the test paper and the answer area in the scanned image of the answer sheet; Based on the geometric positional relationship, key feature points of the question region and the answer region are extracted respectively using an edge detection algorithm; The key feature points are spatially matched to generate a feature point matching matrix containing the feature point mapping relationship. The similarity value between the question region and the answer region is calculated based on the feature point matching matrix. When the similarity value exceeds a preset threshold, the affine transformation matrix parameters are determined based on the similarity value and the feature point matching matrix. Generate the mapping equation from the question region to the answer region based on the affine transformation matrix parameters; A spatial mapping relationship between the test paper and the answer sheet is established based on the mapping equation.

[0094] In one feasible implementation, the data encryption module 20 is further used to extract the original data of the test paper and answer sheet, the original data including image pixel values ​​and text content, and to generate an original data sequence based on the original data; The original data sequence is hashed using the SHA-256 algorithm to generate a fixed-length hash value. The hash value is used as a unique identifier, and a unique time identifier is obtained by appending timestamp information to the unique identifier. Based on the unique time identifier, an encryption key is generated using an asymmetric encryption algorithm, and the original data sequence is encrypted using the encryption key; The encrypted original data sequence and the unique time identifier are broadcast to a distributed node network, so that each node can verify the logical correlation between the timestamp information and the hash value based on a consensus algorithm and provide feedback. When the number of nodes that pass verification reaches a preset threshold, the unique time identifier and encrypted data are written into an immutable block of the blockchain to complete the encryption process.

[0095] In one feasible implementation, the anomaly monitoring module 30 is also used to monitor the scoring operations during the marking process in real time and extract the behavioral characteristics of the scoring operations, including the scoring time interval, scoring consistency index and scoring deviation value. When the scoring consistency index is lower than a preset threshold or the scoring deviation value is higher than a preset threshold, the scoring operation is determined to be an abnormal marking behavior. The blockchain evidence storage mechanism is triggered to generate an immutable evidence record from the context information of the abnormal marking behavior. The context information includes the timestamp of the marking operation, the identity information of the marker, and the marking result.

[0096] In one feasible implementation, the personalized error reporting module 40 is also used to classify and label errors based on a multi-dimensional clustering algorithm; The incorrect question information is determined based on the unique identifier; A personalized error report is generated based on the graded tags and the error information.

[0097] In one feasible implementation, the personalized error reporting module 40 is further used to extract the error answer data from the error information and construct an error feature vector based on the error answer data. The error feature vector includes the question difficulty coefficient, error type, and error frequency. The K-means clustering algorithm is used to perform cluster analysis on the question difficulty coefficient, the error type, and the error frequency. The clustering results are obtained based on the clusters, cluster centroids, and the feature vectors of the incorrect questions. Based on the clustering results, incorrect questions are classified and marked, and the original text of the question, the correct answer, and the student's answer are extracted from the preset question database based on the unique identifier. Generate an error classification description based on the error types in the aforementioned error feature vector; Based on the knowledge dimensions corresponding to the hierarchical labels, learning strategy suggestions are generated; A personalized error report is generated based on the original question, the correct answer, the student's answer, the error classification explanation, and learning strategy suggestions.

[0098] In one feasible implementation, the mapping relationship establishment module 10 is also used to determine the standardized templates of the test paper and answer sheet, and generate template verification rules; Based on the template verification rules, the scanned images of the test paper and answer sheet are verified for format to ensure that the image quality meets the marking requirements. Noise suppression is applied to the scanned images of the verified test papers and answer sheets to generate denoised scanned images; Dynamic thresholding is performed on the denoised scanned image, and the image pixels in the denoised image are binarized to obtain a binarized image. Perform a Hough transform on the binarized image to determine the baseline tilt angle of the binarized image. Based on the tilt angle, rotate and correct the test paper and the answer sheet to align them with the target baseline. In one feasible implementation, the marking allocation module 50 is also used to generate a marking progress visualization interface based on the query interface of the distributed storage node. Automatically assign marking tasks through smart contracts; After the marking is completed, a marking result report is generated based on the blockchain network and sent to the relevant parties through an encrypted channel.

[0099] In one feasible implementation, the marking allocation module 50 is further used to define task allocation parameters in a smart contract based on the encrypted test paper data and marking personnel qualification information stored in the blockchain. The task allocation parameters include the identifier of the question group to be assigned, the skill weight threshold of the marking personnel, and the maximum task allocation time. In response to the completion of the test paper encryption event, the smart contract is triggered to filter the set of examiner nodes that meet the qualification conditions according to the task allocation parameters; The question groups in the encrypted test paper data are dynamically associated with the examiner nodes to generate a task allocation record containing question group identifiers, allocation node addresses and timestamps; The task allocation records are written into the distributed ledger of the blockchain, and the integrity and permission compliance of the allocation records are verified by consensus nodes. Once the verification is successful, a task execution instruction containing encrypted question group data and marking rules is sent to the corresponding marking personnel node, and the question group is marked as assigned in the blockchain. The system monitors the progress of marking tasks in real time. If a task times out or a node is detected as abnormal, the smart contract's automatic recycling mechanism is triggered to redistribute incomplete question sets.

[0100] The blockchain-based evidence storage device for electronic document review provided in this application, employing the blockchain-based evidence storage method for electronic document review in the above embodiments, can solve the technical problem of the lack of full recording and effective supervision of document review behavior in the prior art. Compared with the prior art, the beneficial effects of the blockchain-based evidence storage device for electronic document review provided in this application are the same as the beneficial effects of the blockchain-based evidence storage method for electronic document review provided in the above embodiments, and other technical features in the blockchain-based evidence storage device for electronic document review are the same as the features disclosed in the methods of the above embodiments, and will not be repeated here.

[0101] This application provides a blockchain-based evidence storage device for electronic document review. The device includes: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the blockchain-based evidence storage method for electronic document review described in Embodiment 1 above.

[0102] The following is for reference. Figure 4 The diagram illustrates a structural schematic of a blockchain-based evidence storage device suitable for implementing electronic document review in the embodiments of this application. The blockchain-based evidence storage device for electronic document review in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The blockchain-based evidence storage device for electronic document review shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0103] like Figure 4As shown, the blockchain-based evidence storage device for electronic document review may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the blockchain-based evidence storage device for electronic document review. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, LCDs (Liquid Crystal Displays), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard drives, etc.; and communication devices 1009. Communication device 1009 allows the blockchain-based electronic document review device to communicate wirelessly or wiredly with other devices to exchange data. Although blockchain-based electronic document review devices with various systems are shown in the figures, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0104] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0105] The blockchain-based evidence storage device for electronic document review provided in this application, employing the blockchain-based evidence storage method for electronic document review described in the above embodiments, can solve the technical problems of blockchain-based evidence storage for electronic document review. Compared with the prior art, the beneficial effects of the blockchain-based evidence storage device for electronic document review provided in this application are the same as the beneficial effects of the blockchain-based evidence storage method for electronic document review provided in the above embodiments, and other technical features in this blockchain-based evidence storage device for electronic document review are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0106] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0107] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0108] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the blockchain-based evidence storage method for electronic document review in the above embodiments.

[0109] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory or Flash Memory), optical fibers, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0110] The aforementioned computer-readable storage medium may be included in the blockchain-based evidence storage device for electronic document review; or it may exist independently and not be installed in the blockchain-based evidence storage device for electronic document review.

[0111] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the blockchain evidence storage device for electronic marking, the blockchain evidence storage device for electronic marking causes the electronic marking blockchain evidence storage device to: receive scanned images of the test paper and the answer sheet, extract key feature points from the scanned images of the test paper and the answer sheet, and establish a mapping relationship between the test paper and the answer sheet. The original data of the test paper and answer sheet are encrypted using a hash encryption algorithm to generate a unique identifier, and the scoring behavior of the answer sheet is recorded to obtain the scoring process data. The scoring process data is inspected. When abnormal marking behavior is detected, a blockchain evidence storage mechanism is triggered to record the context information of the abnormal marking behavior and generate an immutable evidence storage record.

[0112] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0113] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. 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. It should also be noted that 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 consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0114] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0115] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned blockchain-based evidence storage method for electronic document review, thereby solving the technical problem of blockchain-based evidence storage for electronic document review. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the blockchain-based evidence storage method for electronic document review provided in the above embodiments, and will not be repeated here.

[0116] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the blockchain-based electronic document review method described above.

[0117] The computer program product provided in this application can solve the technical problem of blockchain-based evidence storage for electronic document review. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the blockchain-based evidence storage method for electronic document review provided in the above embodiments, and will not be repeated here.

[0118] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A blockchain-based evidence storage method for electronic document review, characterized in that, The blockchain-based evidence storage method for electronic document review includes: Receive scanned images of the test paper and the answer sheet, extract key feature points from the scanned images of the test paper and the answer sheet, and establish a mapping relationship between the test paper and the answer sheet; The original data of the test paper and answer sheet are encrypted using a hash encryption algorithm to generate a unique identifier, and the scoring behavior of the answer sheet is recorded to obtain the scoring process data. The scoring process data is inspected. When abnormal marking behavior is detected, a blockchain evidence storage mechanism is triggered to record the context information of the abnormal marking behavior and generate an immutable evidence storage record.

2. The method as described in claim 1, characterized in that, The steps for establishing the mapping relationship between the test paper and the answer sheet include: Determine the geometric positional relationship between the question area in the scanned image of the test paper and the answer area in the scanned image of the answer sheet; Based on the geometric positional relationship, key feature points of the question region and the answer region are extracted respectively using an edge detection algorithm; The key feature points are spatially matched to generate a feature point matching matrix containing the feature point mapping relationship. The similarity value between the question region and the answer region is calculated based on the feature point matching matrix. When the similarity value exceeds a preset threshold, the affine transformation matrix parameters are determined based on the similarity value and the feature point matching matrix. Generate the mapping equation from the question region to the answer region based on the affine transformation matrix parameters; A spatial mapping relationship between the test paper and the answer sheet is established based on the mapping equation.

3. The method as described in claim 1, characterized in that, The steps for encrypting the original data of the exam paper and answer sheet based on the hash encryption algorithm include: Extract the raw data from the test paper and answer sheet, the raw data including image pixel values ​​and text content, and generate a raw data sequence based on the raw data; The original data sequence is hashed using the SHA-256 algorithm to generate a fixed-length hash value. The hash value is used as a unique identifier, and a unique time identifier is obtained by appending timestamp information to the unique identifier. Based on the unique time identifier, an encryption key is generated using an asymmetric encryption algorithm, and the original data sequence is encrypted using the encryption key; The encrypted original data sequence and the unique time identifier are broadcast to a distributed node network, so that each node can verify the logical correlation between the timestamp information and the hash value based on a consensus algorithm and provide feedback. When the number of nodes that pass verification reaches a preset threshold, the unique time identifier and encrypted data are written into an immutable block of the blockchain to complete the encryption process.

4. The method as described in claim 1, characterized in that, The step of triggering a blockchain evidence storage mechanism to record the context information of the abnormal marking behavior and generate an immutable evidence storage record when the marking process data is detected as abnormal includes: The scoring operations during the marking process are monitored in real time, and the behavioral characteristics of the scoring operations are extracted. These behavioral characteristics include the scoring time interval, the scoring consistency index, and the scoring deviation value. When the scoring consistency index is lower than a preset threshold or the scoring deviation value is higher than a preset threshold, the scoring operation is determined to be an abnormal marking behavior. The blockchain evidence storage mechanism is triggered to generate an immutable evidence record from the context information of the abnormal marking behavior. The context information includes the timestamp of the marking operation, the identity information of the marker, and the marking result.

5. The method as described in claim 1, characterized in that, The blockchain-based evidence storage method for electronic document review also includes: Incorrect questions are graded and marked based on a multi-dimensional clustering algorithm; The incorrect question information is determined based on the unique identifier; A personalized error report is generated based on the graded tags and the error information.

6. The method as described in claim 5, characterized in that, The step of generating a personalized error report based on the graded markers and the error information includes: Extract incorrect answer data from the incorrect question information, and construct an incorrect question feature vector based on the incorrect answer data. The incorrect question feature vector includes question difficulty coefficient, error type, and error frequency. The K-means clustering algorithm is used to perform cluster analysis on the question difficulty coefficient, the error type, and the error frequency. The clustering results are obtained based on the clusters, cluster centroids, and the feature vectors of the incorrect questions. Based on the clustering results, incorrect questions are classified and marked, and the original text of the question, the correct answer, and the student's answer are extracted from the preset question database based on the unique identifier. Generate an error classification description based on the error types in the aforementioned error feature vector; Based on the knowledge dimensions corresponding to the hierarchical labels, learning strategy suggestions are generated; A personalized error report is generated based on the original question, the correct answer, the student's answer, the error classification explanation, and learning strategy suggestions.

7. The method as described in claim 1, characterized in that, Before the step of receiving the scanned images of the exam paper and the answer sheet, the method further includes: Determine standardized templates for exam papers and answer sheets, and generate template verification rules; Based on the template verification rules, the scanned images of the test paper and answer sheet are verified for format to ensure that the image quality meets the marking requirements. Noise suppression is applied to the scanned images of the verified test papers and answer sheets to generate denoised scanned images; Dynamic thresholding is performed on the denoised scanned image, and the image pixels in the denoised image are binarized to obtain a binarized image. A Hough transform is performed on the binarized image to determine the baseline tilt angle of the binarized image. Based on the tilt angle, the test paper and the answer sheet are rotated and corrected so that the test paper and the answer sheet are aligned with the target baseline.

8. The method as described in claim 1, characterized in that, The blockchain-based evidence storage method for electronic document review also includes: A visual interface for the marking progress is generated based on the query interface of distributed storage nodes. Automatically assign marking tasks through smart contracts; After the marking is completed, a marking result report is generated based on the blockchain network and sent to the relevant parties through an encrypted channel.

9. The method as described in claim 8, characterized in that, The steps for automatically assigning marking tasks via smart contracts include: Based on the encrypted test paper data and examiner qualification information stored in the blockchain, task allocation parameters are defined in the smart contract. The task allocation parameters include the question group identifier to be assigned, the examiner skill weight threshold, and the maximum task allocation time. In response to the completion of the test paper encryption event, the smart contract is triggered to filter the set of examiner nodes that meet the qualification conditions according to the task allocation parameters; The question groups in the encrypted test paper data are dynamically associated with the examiner nodes to generate a task allocation record containing question group identifiers, allocation node addresses and timestamps; The task allocation records are written into the distributed ledger of the blockchain, and the integrity and permission compliance of the allocation records are verified by consensus nodes. Once the verification is successful, a task execution instruction containing encrypted question group data and marking rules is sent to the corresponding marking personnel node, and the question group is marked as assigned in the blockchain. The system monitors the progress of marking tasks in real time. If a task times out or a node is detected as abnormal, the smart contract's automatic recycling mechanism is triggered to redistribute incomplete question sets.

10. A blockchain-based evidence storage device for electronic document review, characterized in that, The blockchain-based evidence storage device for electronic document review includes: The mapping relationship establishment module is used to receive scanned images of the test paper and the answer sheet, extract key feature points from the scanned images of the test paper and the answer sheet, and establish a mapping relationship between the test paper and the answer sheet. The data encryption module is used to encrypt the original data of the test paper and answer sheet based on the hash encryption algorithm, generate a unique identifier, record the scoring behavior of the answer sheet, and obtain the scoring process data. An anomaly monitoring module is used to detect the scoring process data. When an abnormal marking behavior is detected in the scoring process data, a blockchain evidence storage mechanism is triggered to record the context information of the abnormal marking behavior and generate an immutable evidence storage record.

Images