An online control method based on intelligent operation data block encryption storage

Abstract

The application discloses an online control method for encrypted storage based on intelligent operation data blocking, relates to the technical fields of intelligent education data processing and data security, and is used for solving the problem that the tail stroke data generated before the first stroke is lifted cannot be accurately supplemented due to the delay in arrival after the first stroke data has been encrypted and storage and the second stroke starts; the late stroke sequence which arrives after the second stroke starts and the sampling time is located before the first stroke lifting event is obtained, the continuous relationship between the late stroke sequence and the first stroke end and the second stroke start is determined to determine the attribution, the associated supplementary encrypted data block is formed under the condition that the first encrypted data block ciphertext content remains unchanged, and the first stroke is reconstructed based on the two encrypted data blocks, so that the integrity, attribution accuracy and storage security of the stroke data are improved.

Description

Technical Field

[0001] This application relates to the fields of smart education data processing and data security technology, and in particular to an online control method for encrypted storage based on smart homework data segmentation. Background Technology

[0002] The intelligent homework system can continuously collect handwriting sampling points, pen-dropping events, and pen-lifting events during the student's writing process using smart pens, dot matrix pens, touch pens, or homework terminals with writing acquisition functions. In order to reduce the storage pressure caused by the centralized uploading of continuous homework data and reduce the risk of homework data being read or tampered with during transmission and storage, the handwriting data can be divided into multiple data blocks according to stroke boundaries, acquisition duration, or data capacity, and each data block can be encrypted and stored separately.

[0003] In one online processing method, the intelligent operation server starts forming the current stroke based on the pen-starting event, updates the current stroke based on the continuously received handwriting sampling points, and ends the current stroke and encrypts and stores the corresponding data block based on the pen-lifting event. In some systems that use handwriting prediction display, the server or operation terminal can also generate predicted handwriting segments for real-time display based on the received end sampling points. After receiving the next pen-starting event, the server starts forming the next stroke and its corresponding data.

[0004] However, handwriting sampling points and pen tip state events may pass through different acquisition channels, buffer queues, or transmission links. Sometimes, the pen lifting event arrives at the server earlier than the tail handwriting sampling point that actually occurs before the pen is lifted. If the server has already encrypted and stored the data block of the previous stroke based on the pen lifting event, and has already started forming the next stroke based on the next pen falling event, the tail handwriting sampling point that arrives subsequently is easily misclassified as the next stroke. If the previous encrypted data block is modified, the ciphertext and its integrity verification information will be changed. If the tail handwriting sampling point is directly discarded, the tail of the previous stroke will be missing. In the case of predictive handwriting display, predictive handwriting that has not been confirmed by real data may continue to participate in display or recognition. Therefore, a method is needed to controllably supplement the tail handwriting data that is generated before the pen is lifted but arrives late, while keeping the previous encrypted data block unchanged. Summary of the Invention

[0005] This application provides an online control method for encrypted storage based on intelligent operation data block segmentation. After the data block corresponding to the first stroke has been encrypted and stored and the second stroke has started, it identifies the handwriting data that was actually generated before the first stroke was lifted but arrived late, and records it by supplementing it with a supplementary encrypted data block associated with the original encrypted data block, thereby avoiding the late handwriting data being incorrectly classified into the second stroke and avoiding the need to modify the first encrypted data block again.

[0006] This application provides an online control method for encrypted storage based on intelligent operation data block segmentation. The method includes: in response to the pen lifting event of the first stroke, forming and encrypting the received first stroke handwriting data into a first encrypted data block. After the second stroke begins, a sequence of late handwritings that arrives after the second stroke begins and whose sampling time is before the pen lifting event is obtained; based on the continuity relationship between the late handwriting sequence and the end of the first stroke and the beginning of the second stroke, the sequence of late handwritings is determined to belong to the first stroke. While keeping the ciphertext content of the first encrypted data block unchanged, the late handwriting sequence and its supplementary recording position in the first stroke are used to form a supplementary encrypted data block associated with the first encrypted data block, and the first stroke is reconstructed based on the first encrypted data block and the supplementary encrypted data block.

[0007] In one possible implementation, the intelligent task server selects candidate handwriting sampling points from the handwriting sampling points received after the second stroke's stroke-starting event. These candidate handwriting sampling points have the same task session identifier, writing page identifier, terminal identifier, and time sequence identifier as the first stroke, and their sampling time is earlier than the time corresponding to the first stroke's stroke-starting event. The candidate handwriting sampling points are sorted according to their sampling time, and the candidate handwriting sampling points whose time interval between adjacent sampling times satisfies the continuous sampling condition are grouped into a late handwriting sequence.

[0008] In one possible implementation, the forward continuity condition includes: the sampling time interval between the end sampling point of the first stroke and the first sampling point of the late stroke sequence is not greater than the continuous sampling time threshold, the spatial distance between the two is not greater than the continuous writing reachable distance, and at least one of the changes in the direction of movement and the changes in writing pressure between the two satisfies the corresponding continuity condition.

[0009] In one possible implementation, the backward exclusion condition includes: the spatial distance between the last sampling point of the late handwriting sequence and the starting sampling point of the second stroke is greater than the starting reach distance of the second stroke, and at least one of the following: the connection relationship of the movement direction between the two or the change in writing pressure indicates that the second stroke is formed by re-writing.

[0010] In one possible implementation, the conditions for completing the supplementary recording include: no preceding handwriting sampling points that have not yet been received are detected, and any of the following conditions are met: a valid tail data transmission completion identifier is received; the last sampling moment of the late handwriting sequence enters the allowable error range before the moment corresponding to the pen lifting event; or the supplementary recording waiting window expires; the tail data transmission completion identifier includes the event identifier of the pen lifting event and the sampling sequence number of the last handwriting sampling point before the pen lifting event, and the intelligent operation server determines that the tail data transmission completion identifier is valid when it has received handwriting sampling points up to the sampling sequence number.

[0011] In one possible implementation, the intelligent operation server or operation terminal generates a predicted handwriting segment for real-time display based on the end sampling point of the first stroke, and stores the predicted handwriting segment as display auxiliary data associated with the first encrypted data block; after determining that the late handwriting sequence belongs to the first stroke, the predicted sub-segment in the predicted handwriting segment corresponding to the late handwriting sequence is invalidated; after the supplementary recording waiting window ends, the remaining part of the predicted handwriting segment that has not been confirmed by the real handwriting sampling point is invalidated.

[0012] In one possible implementation, the supplementary recording status identifier includes a supplementary recording allowed status, a supplementary recording in progress status, and a supplementary recording prohibited status. After the intelligent operation server determines that the late handwriting sequence belongs to the first stroke and meets the supplementary recording completion conditions, it updates the supplementary recording status identifier from the supplementary recording allowed status to the supplementary recording in progress status through a condition update operation. Only supplementary encrypted data blocks are generated from supplementary recording transactions with successful status updates. After the supplementary encrypted data block is successfully persisted, the supplementary recording status identifier is updated to the supplementary recording prohibited status. If persistence fails, the supplementary recording status identifier is restored to the supplementary recording allowed status.

[0013] In one possible implementation, when multiple sets of late handwriting sequences are obtained within the supplementary recording waiting window, the intelligent operation server sorts them according to the sampling time of each late handwriting sequence. Starting from the earliest sampling time, it sequentially selects late handwriting sequences that are continuous with the end sampling point of the first stroke or the previously selected late handwriting sequence and whose sampling time ranges do not overlap, based on the forward continuity condition, until a late handwriting sequence that does not meet the forward continuity condition appears. The selected late handwriting sequences are then merged to form a supplementary data block.

[0014] In one possible implementation, when a late handwriting sequence does not meet the forward continuity condition, or meets the forward continuity condition but still has the possibility of continuous connection with the starting sampling point of the second stroke, the intelligent operation server encrypts and stores the late handwriting sequence to obtain an encrypted data block to be verified. During the period when the late handwriting sequence is in the state of pending verification, it is not used to reconstruct the first or second stroke, and its ownership is re-determined based on the stroke identifier, the preceding sampling point identifier, or the manual confirmation result uploaded by the operation terminal.

[0015] The technical solution provided in this application does not directly determine the stroke attribution based on the order in which the data arrives at the server after the data block corresponding to the first stroke has been encrypted and stored and the second stroke has begun. Instead, it uses the same terminal timing reference, the continuity of the forward trajectory, and the re-stroke characteristics of the second stroke to determine the attribution of the late handwriting sequence. At the same time, late data is collected in the supplementary recording waiting window, and a supplementary encrypted data block is generated all at once after confirming that there are no gaps in the preceding data and that the end triggering condition is met. Therefore, the first encrypted data block and its integrity verification information can be kept unchanged, reducing the possibility that the late handwriting sequence is incorrectly classified as the second stroke and that the tail data is missing due to the premature closure of the supplementary recording entry. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the implementation environment of a smart operation system provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the normal arrival sequence of handwriting events provided in an embodiment of this application; Figure 3 This is a schematic diagram illustrating an example of a pen-lifting event occurring first, followed by a sampling point for the end of the handwriting. Figure 4 This is a flowchart of an online control method for encrypted storage provided in an embodiment of this application; Figure 5 This is a flowchart of a late handwriting sequence filtering method provided in an embodiment of this application; Figure 6 This is a schematic diagram illustrating the determination of forward continuity conditions and backward exclusion conditions provided in an embodiment of this application; Figure 7 This is a schematic diagram of a supplementary recording waiting window and supplementary recording completion conditions provided in an embodiment of this application; Figure 8 This is a schematic diagram of the association structure between a first encrypted data block and a supplementary encrypted data block provided in an embodiment of this application; Figure 9 This is a schematic diagram of a supplementary recording state transition provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the implementation methods of this application will be further described below with reference to the accompanying drawings; it should be understood that the described embodiments are only used to illustrate this application and are not intended to limit the scope of protection of this application.

[0018] like Figure 1As shown, the intelligent work system includes work terminals and intelligent work servers; the work terminals and intelligent work servers can be connected via wired or wireless networks; the work terminals can be smart pens, dot matrix pens, styluses, tablets with touch screens, electronic notebooks, or other terminal devices capable of collecting writing trajectories; the intelligent work servers can be implemented by a single server, or by a server cluster or cloud computing platform.

[0019] In one possible implementation, the work terminal includes handwriting sampling point acquisition function, pen-falling event and pen-lifting event acquisition function, local caching function, and data uploading function; the intelligent work server is used to perform data reception, stroke attribution judgment, data block encrypted storage, late handwriting supplementation control, and stroke reconstruction processing; the above functions can be implemented by software, hardware, or a combination of software and hardware.

[0020] The terminal collects handwriting sampling points, pen-drop events, and pen-lifting events during the user's writing process. The handwriting sampling points include at least the sampling time and coordinates, and may also include at least one of the following: sampling sequence number, writing pressure, pen tip tilt angle, and pen tip azimuth angle. The pen-drop event indicates that the pen tip changes from not contacting the writing surface to contacting the writing surface, and the pen-lifting event indicates that the pen tip changes from contacting the writing surface to not contacting the writing surface.

[0021] The handwriting sampling points, pen-dropping events, and pen-lifting events all carry job session identifiers, writing page identifiers, terminal identifiers, and timer generation identifiers. The job session identifier is used to distinguish different job submission processes, the writing page identifier is used to distinguish different job pages, the terminal identifier is used to distinguish the job terminal uploading data, and the timer generation identifier is used to distinguish different continuous operation stages of the terminal's timing benchmark. The intelligent job server only performs stroke attribution judgment on data where all of the above identifiers are the same.

[0022] Within the same time period, the work terminal uses the same monotonically increasing time reference to record the event times of handwriting sampling points, pen-putting events, and pen-lifting events. When the terminal time reference is reset, wrapped around, or the terminal is restarted, the work terminal generates a new time period identifier, and the intelligent work server does not compare event times across time periods. This avoids errors in judging the relationship before and after pen lifting due to the reuse of terminal time values.

[0023] The first stroke indicates a handwriting capture process that begins with a pen-down event and ends with a subsequent pen-up event; the second stroke indicates a handwriting capture process that begins with the next pen-down event after the pen-up event of the first stroke; the first and second strokes are used to distinguish between two adjacent handwriting capture processes and do not limit the corresponding text, numbers, formulas or graphic content.

[0024] like Figure 2As shown, under normal arrival sequence, the intelligent operation server receives the handwriting sampling point of the first stroke, the lifting event of the first stroke, and the finishing event of the second stroke in sequence. The intelligent operation server can end the first stroke according to the lifting event and encrypt and store the data block corresponding to the first stroke.

[0025] Under the normal arrival sequence, the actual generation order of handwriting events at the work terminal is consistent with the receiving order of the smart work server, that is, the handwriting sampling point of the first stroke, the lifting event of the first stroke, the finishing event of the second stroke, and the subsequent handwriting sampling points of the second stroke are generated and received in sequence.

[0026] like Figure 3 As shown, since coordinate sampling data and pen tip state events can pass through different buffer queues or communication channels, some tail handwriting sampling points in the first stroke that actually occurred before the pen lifting event may arrive at the smart operation server after the pen lifting event and the pen falling event of the second stroke. The arrival time of this tail handwriting sampling point is later than the pen falling event of the second stroke, but its sampling time is earlier than the corresponding time of the pen lifting event of the first stroke.

[0027] In one possible implementation, such as Figure 4 As shown, the encrypted storage online control method may include steps S101 to S106.

[0028] S101. Obtain handwriting sampling points, pen-drop events, and pen-lift events within the same job session; The intelligent task server receives handwriting sampling points, pen-starting events, and pen-lifting events uploaded by the task terminal, and verifies the task session identifier, writing page identifier, terminal identifier, and timer identifier; data with inconsistent identifiers are not included in the timing and continuity judgment of the same stroke.

[0029] The work terminal can send handwriting data according to a fixed upload cycle, or send handwriting data in batches when the local cache reaches a preset capacity, a pen lifting event is detected, or the network is restored; regardless of the sending method, the sampling time represents the actual time when the data is generated at the work terminal, rather than the time when the data arrives at the smart work server.

[0030] S102. Form and encrypt the first data block; In response to the pen-lifting event of the first stroke, the intelligent operation server forms a first data block from the handwriting sampling points that have been received and attributed to the first stroke before the pen-lifting event, and encrypts and stores the first data block to obtain a first encrypted data block. The first data block may include the first stroke identifier, handwriting sampling points, the time corresponding to the pen-lifting event, and the time when the data block was generated.

[0031] The first data block is encrypted and stored using either symmetric encryption or authentication encryption that provides both data confidentiality and integrity verification capabilities. This application does not limit the specific encryption algorithm. The first encrypted data block has a data block identifier and an integrity verification value, which can be a verification value calculated for the ciphertext data.

[0032] In one possible implementation, when encrypting the first data block for authentication, the job session identifier, writing page identifier, terminal identifier, time sequence identifier, first stroke identifier, data block identifier, and the time corresponding to the pen lift event are used as associated authentication information. The associated authentication information does not need to be written into the ciphertext, but it participates in the generation of authentication tags or integrity verification values, so that if any associated authentication information is modified, it cannot pass the integrity verification.

[0033] In one possible implementation, the first encrypted data block records a key identifier and an encryption parameter identifier. The key identifier is used to determine the key version used when encrypting the first data block. When the encryption method requires an initialization vector or random parameters, the intelligent operation server generates unique initialization parameters for the same key and stores the initialization parameters or their acquisition identifier together with the corresponding ciphertext. If the corresponding key cannot be obtained, the initialization parameters are missing, or the key version does not match, decryption and stroke reconstruction are not performed on the first encrypted data block.

[0034] The intelligent operation server sets a supplementary recording status identifier associated with the first encrypted data block and sets its initial state to allow supplementary recording. The supplementary recording status identifier can be stored in the index record of the first encrypted data block or in a separate status record. The supplementary recording status identifier is used to control whether it is allowed to supplement the first stroke with a late stroke sequence by supplementing the encrypted data block, and is not used to modify the ciphertext content in the first encrypted data block.

[0035] S103. Obtain the delayed handwriting sequence within the supplementary recording waiting window; after receiving the second stroke's stroke-starting event, the intelligent job server determines whether the supplementary recording status identifier of the first encrypted data block corresponding to the first stroke is in the allowed supplementary recording state; when the supplementary recording status identifier is in the allowed supplementary recording state, the intelligent job server starts the supplementary recording waiting window for the first encrypted data block; the supplementary recording waiting window is used to continue receiving handwriting sampling points that actually occurred before the first stroke's stroke-starting event but arrived at the intelligent job server with a delay.

[0036] In one possible implementation, when the lifting event of the first stroke arrives, and the finishing event of the second stroke is not received within the preset start delay, but the upload status information sent by the work terminal indicates that there is still tail data of the first stroke that has not been confirmed to be uploaded, the smart work server starts a supplementary recording waiting window for the first encrypted data block.

[0037] The preset start delay can be determined based on the upload cycle of the working terminal and the normal communication delay; the handwriting sampling points received within the supplementary recording waiting window are still processed according to the candidate handwriting sampling point screening rules, forward continuity conditions and supplementary recording completion conditions.

[0038] The intelligent job server determines the duration of the supplementary recording waiting window based on the local cache duration of the job terminal, the historical latency distribution of the communication channel, and the upload cycle of the job terminal. For example, the intelligent job server can determine the basic duration of the supplementary recording waiting window by summing the preset high quantile value of the end-to-end transmission latency under normal network conditions with one upload cycle, and determine the maximum duration of the supplementary recording waiting window based on the maximum local cache duration allowed by the job terminal, two upload cycles, and the network anomaly recovery time, so as to limit the time when the supplementary recording entry is continuously open.

[0039] While the supplementary recording status flag remains in the supplementary recording allowed state, each first encrypted data block corresponds to only one valid supplementary recording waiting window; when the intelligent operation server repeatedly receives a second stroke stroke event carrying the same stroke event flag, the intelligent operation server reuses the supplementary recording waiting window that has already been started for the first encrypted data block, instead of creating a new supplementary recording waiting window.

[0040] The intelligent job server persistently stores the start time, maximum end time, and associated first encrypted data block identifier of the supplementary recording waiting window. When the intelligent job server resumes operation after an interruption, it determines whether the supplementary recording waiting window is still valid based on the relationship between the current time and the maximum end time. If the current time has not reached the maximum end time, the supplementary recording waiting process continues. If the current time reaches or exceeds the maximum end time, the supplementary recording waiting window ends. When a job submission, writing page closure, or timer change is detected, the intelligent job server ends the supplementary recording waiting window early and performs a supplementary recording completion judgment on the handwriting sampling points already received within the supplementary recording waiting window.

[0041] like Figure 5 As shown, the intelligent task server filters candidate handwriting sampling points from the handwriting sampling points received after the second stroke's stroke-starting event. These sampling points have the same task session identifier, writing page identifier, terminal identifier, and time sequence identifier as the first stroke, and their sampling time is earlier than the time corresponding to the first stroke's stroke-starting event.

[0042] When the sampling time of a handwriting sampling point is the same as the time corresponding to the pen lifting event of the first stroke, the intelligent operation server obtains the pen tip contact state corresponding to the handwriting sampling point; when the pen tip is still in contact with the writing surface when the handwriting sampling point is generated, it is used as a candidate boundary point for the first stroke; when the pen tip is in a non-contact state or the pen tip contact state cannot be determined, it is not directly added to the first stroke, but is encrypted, stored, and entered into the verification process; the candidate boundary points are processed according to the same sampling time, continuous sampling judgment, and stroke belonging judgment rules as the candidate handwriting sampling points.

[0043] When the received handwriting sampling point is inconsistent with the job session identifier, writing page identifier, terminal identifier, or timer identifier of the first stroke, or when its sampling time is not earlier than the time corresponding to the pen lifting event of the first stroke, the intelligent job server will not use the handwriting sampling point as a candidate handwriting sampling point, and will continue to receive and filter subsequent handwriting sampling points.

[0044] The intelligent operation server sorts the candidate handwriting sampling points according to the sampling time and compares the time intervals between adjacent candidate handwriting sampling points in turn. When the time intervals between multiple consecutive candidate handwriting sampling points meet the continuous sampling condition, the multiple consecutive candidate handwriting sampling points are grouped into a late handwriting sequence. When the time intervals between adjacent candidate handwriting sampling points do not meet the continuous sampling condition, they are divided into different late handwriting sequences from that position.

[0045] The continuous sampling time threshold can be determined based on the nominal sampling period of the working terminal and the number of consecutively missing sampling points allowed. For example, if the nominal sampling period of the working terminal is about 8 milliseconds and three sampling periods are allowed between adjacent real points, then the duration corresponding to four sampling periods plus the timing error margin can be used as the continuous sampling time threshold. This value is only an example, and the actual value can be determined according to the sampling characteristics of different terminals.

[0046] S104. Determine whether the late handwriting sequence belongs to the first stroke; such as Figure 6 As shown, the intelligent operation server performs forward continuity judgment and backward exclusion judgment respectively; forward continuity judgment is used to determine whether the late handwriting sequence can continue the end of the first stroke, and backward exclusion judgment is used to determine whether the late handwriting sequence should not be the starting part of the second stroke.

[0047] The intelligent task server acquires the end sampling point of the first stroke and the first stroke sampling point of the late stroke sequence; the end sampling point can be the latest real stroke sampling point in the first data block; the intelligent task server determines the sampling time interval and spatial distance between the two.

[0048] The achievable distance for continuous writing can be determined based on the pre-obtained upper limit of writing speed, the sampling time interval between the end sampling point and the first point of the late handwriting sequence, and the coordinate quantization error margin. The upper limit of writing speed can be obtained from the historical writing samples of similar working terminals, or it can be determined based on the user's trial writing data at the beginning of the working session. The longer the time interval, the greater the corresponding achievable distance for continuous writing.

[0049] The change in motion direction can be determined by the angle between the end motion direction formed by at least two consecutive sampling points at the end of the first stroke and the connection direction from the sampling point at the end of the first stroke to the first point of the late handwriting sequence; the direction continuity threshold can be determined by the distribution of the direction angles of continuous stroke samples and the direction calculation error caused by coordinate quantization; the change in writing pressure can be determined by the normalized absolute pressure difference or pressure change trend; the pressure continuity threshold can be determined by the statistical distribution of continuous writing pressure changes on the same terminal and the upper limit of pressure sensor noise.

[0050] Figure 6 In this context, Δt represents the sampling time interval between the end sampling point of the first stroke and the first sampling point of the late stroke sequence, d1 represents the spatial distance between the end sampling point and the first sampling point, θ1 represents the angle between the direction of movement of the end of the first stroke and the direction of connection between the end sampling point and the first sampling point, and d2 represents the spatial distance between the last sampling point of the late stroke sequence and the starting sampling point of the second stroke.

[0051] When the sampling time interval is not greater than the continuous sampling time threshold, the spatial distance is not greater than the continuous writing reachable distance, and at least one of the changes in movement direction and writing pressure satisfies the corresponding continuity condition, it is determined that the late handwriting sequence and the end sampling point of the first stroke satisfy the forward continuity condition. Thus, time and spatial conditions are used as basic conditions, and direction or pressure is used as an enhancement condition to avoid incorrect continuation caused by judging based on just two indicators.

[0052] When the sampling time interval is equal to the continuous sampling time threshold, or when the spatial distance between the end sampling point and the first point of the late handwriting sequence is equal to the reachable distance of continuous writing, the corresponding time basis condition or spatial basis condition is determined to be satisfied.

[0053] When the spatial distance between the end point of the late handwriting sequence and the starting sampling point of the second stroke is equal to the starting reachable distance of the second stroke, not only is the backward exclusion condition determined based on the spatial distance, but also the enhanced judgment is made by combining the connection relationship of the movement direction, the change of writing pressure and the contact state of the pen tip; if it cannot be determined that there are re-writing features, the late handwriting sequence is transferred to the pending verification process.

[0054] The intelligent task server acquires the last handwriting sampling point and the starting sampling point of the second stroke in the late handwriting sequence; the starting sampling point of the second stroke can be the earliest real handwriting sampling point in the second stroke; the starting reachable distance of the second stroke can be determined according to the sampling time interval between the two, the upper limit of writing speed after rewriting, and the coordinate quantization error margin; the upper limit of writing speed can be determined by the historical starting sample after rewriting or the starting sample of the current user.

[0055] When the spatial distance between the end point of the late handwriting sequence and the starting sampling point of the second stroke is greater than the reachable distance of the starting point of the second stroke, and at least one of the following indicates that the second stroke is formed by re-writing, the backward exclusion condition is determined to be met; the re-writing feature can be manifested as the pressure rising again from zero or non-contact state, or there is a significant discontinuity between the starting direction of the second stroke and the ending direction of the late handwriting sequence.

[0056] When a late handwriting sequence satisfies both the forward continuity condition and the backward exclusion condition, the intelligent operation server determines that it belongs to the first stroke; when a late handwriting sequence does not satisfy the forward continuity condition, or has the possibility of being connected to the first stroke and the second stroke, its affiliation is not directly determined, but pending verification is performed.

[0057] S105. Confirm that the supplementary recording is complete and generate a supplementary encrypted data block; after the intelligent operation server determines that the first group of late handwriting sequences belongs to the first stroke, it can continue to receive other candidate handwriting sampling points within the supplementary recording waiting window without immediately closing the supplementary recording entry; such as Figure 7 As shown, when the conditions for completion of supplementary recording are met, the collection of the late tail data of the first stroke is determined to be complete.

[0058] The conditions for completing the supplementary recording include: no unreceived preceding handwriting sampling points are detected, and any of the following termination trigger conditions are met: the intelligent operation server receives a valid tail data transmission completion identifier; the last sampling moment of the late handwriting sequence belonging to the first stroke enters the allowable error range before the time corresponding to the first stroke lifting event; or the supplementary recording waiting window expires; the tail data transmission completion identifier includes the lifting event identifier and the sampling sequence number of the last handwriting sampling point before lifting the pen; the intelligent operation server determines that the identifier is valid only after all handwriting sampling points up to the sampling sequence number have been received.

[0059] The allowable error range can be determined based on the sampling period of the working terminal and the pen lifting event detection error; for example, if the time difference between the end point of the late handwriting sequence and the corresponding moment of the pen lifting event is no greater than one or two sampling periods, it can be considered that the late handwriting has approached the true end position of the first stroke; however, it is still necessary to confirm that there are no previous handwriting sampling points that have not yet been received.

[0060] The number of previous handwriting samples that have not yet been received can be determined based on whether the sampling sequence number is consecutive. When the work terminal does not provide a sampling sequence number, it can be determined by combining the number of sampling points declared in the uploaded batch, the batch integrity check value, and the gap between adjacent sampling times. When any information indicates that there are still data gaps within the late handwriting sequence, it is uncertain whether the conditions for completing the supplementary recording are met.

[0061] The intelligent operation server determines the supplementary recording position of the late handwriting sequence in the first stroke; the supplementary recording position may include the first stroke identifier, the first stroke end sampling point identifier as the precursor, and the sampling time range of the late handwriting sequence; the intelligent operation server writes the late handwriting sequence, the supplementary recording position, the data block identifier of the first encrypted data block, and the integrity check value into the supplementary data block.

[0062] The intelligent operation server encrypts and stores the supplementary data block to obtain a supplementary encrypted data block; the supplementary encrypted data block also has a supplementary data block identifier, a supplementary data block integrity verification value, a key identifier, and a supplementary recording transaction identifier; the supplementary data block integrity verification value at least covers the late handwriting sequence, the supplementary recording position, the first encrypted data block identifier, the first encrypted data block integrity verification value, and the supplementary recording transaction identifier; when any of the above contents changes, the supplementary encrypted data block cannot pass the integrity verification.

[0063] In one possible implementation, the first encrypted data block identifier, the first encrypted data block integrity verification value, the job session identifier, the writing page identifier, and the first stroke identifier are used as the second association authentication information of the supplementary encrypted data block, so as to limit the supplementary encrypted data block to be associated only with the corresponding first encrypted data block.

[0064] like Figure 8 As shown, the first encrypted data block and the supplementary encrypted data block are stored separately. The supplementary encrypted data block is associated with the first encrypted data block through the data block identifier and integrity check value. The generation of the supplementary encrypted data block does not change the ciphertext content of the first encrypted data block.

[0065] The supplementary encrypted data block also has a supplementary data block identifier and a supplementary data block integrity verification value; the supplementary data block integrity verification value at least covers the late handwriting sequence, the supplementary recording position, the first encrypted data block identifier, the first encrypted data block integrity verification value, and the supplementary recording transaction identifier; if any of the above contents changes, the supplementary encrypted data block cannot pass the integrity verification.

[0066] S106, Update the supplementary recording status and reconstruct the first stroke; as follows: Figure 9As shown, the supplementary recording status identifier can include supplementary recording allowed, supplementary recording in progress, and supplementary recording prohibited. After the intelligent operation server determines that the late handwriting sequence belongs to the first stroke and meets the supplementary recording completion conditions, it updates the supplementary recording status identifier from supplementary recording allowed to supplementary recording in progress through a condition update operation. When the supplementary recording completion conditions are not met within the supplementary recording waiting window, the supplementary recording status identifier remains in the supplementary recording allowed state.

[0067] Conditional update operations can be database conditional updates, comparison and swap operations, or state update operations with the same atomic effect; only supplementary transactions that successfully update the state are granted supplementary processing rights for the first encrypted data block, and other concurrent supplementary transactions stop generating supplementary encrypted data blocks.

[0068] Each supplementary recording transaction has a supplementary recording transaction identifier; after the supplementary encrypted data block is successfully persisted, the intelligent operation server updates the supplementary recording status identifier from the supplementary recording processing status to the prohibited supplementary recording status; if persistence fails, the supplementary recording status identifier is restored to the allowed supplementary recording status; when the same supplementary recording transaction is executed repeatedly, the already persisted supplementary encrypted data block can be reused according to the supplementary recording transaction identifier to avoid duplicate supplementary recording.

[0069] When reconstructing the first stroke, the intelligent operation server sequentially verifies the operation session identifier, writing page identifier, terminal identifier, and timer identifier corresponding to the first encrypted data block and the supplementary encrypted data block, verifies the integrity check value of the first encrypted data block, verifies the integrity check value of the supplementary encrypted data block, and verifies whether the first encrypted data block identifier and the first encrypted data block integrity check value recorded in the supplementary encrypted data block are consistent with the first encrypted data block to be reconstructed.

[0070] When all the above verifications pass, the intelligent operation server decrypts the first encrypted data block and the supplementary encrypted data block, and arranges the late handwriting sequence after the end sampling point of the first stroke and before the lifting event of the first stroke according to the supplementary recording position and sampling time, so as to obtain the reconstructed first stroke.

[0071] When the integrity verification of any data block fails, the association information of two data blocks is inconsistent, the supplementary transaction identifier is duplicated but the corresponding data content is inconsistent, or the corresponding key cannot be obtained, the intelligent operation server stops merging the supplementary encrypted data block into the first stroke, marks the corresponding data block as a security verification anomaly, and records the verification failure type, the relevant data block identifier, and the processing time.

[0072] When the supplementary recording waiting window ends and there is no late stroke sequence belonging to the first stroke, the intelligent job server can directly update the supplementary recording status to the prohibited supplementary recording status without generating supplementary encrypted data blocks, so as to avoid keeping the supplementary recording entry open for a long time.

[0073] After the supplementary recording status indicator is already in the prohibited supplementary recording state, if the intelligent operation server receives a handwriting sampling point whose sampling time is earlier than the corresponding pen lifting event, it will not reopen the supplementary recording entry of the first encrypted data block, nor will it modify the already generated supplementary encrypted data block; the intelligent operation server will encrypt and store the handwriting sampling point to form late abnormal data after supplementary recording is closed, and record the first encrypted data block identifier, data arrival time and reason for not supplementing.

[0074] After the supplementary recording is closed, late abnormal data will not participate in the display, recognition, correction or stroke reconstruction of the current first stroke; subsequently, an independent abnormal processing record can be generated based on the manual verification result, but the already stored first encrypted data block and supplementary encrypted data block will not be changed.

[0075] In one possible implementation, to reduce handwriting display latency, the intelligent homework server or homework terminal generates a predicted handwriting segment based on the real handwriting sampling points already received at the end of the first stroke; the predicted handwriting segment has a predicted data identifier and is stored as display auxiliary data associated with the first encrypted data block, and is not used as the student's original homework handwriting for identification, grading or homework evidence.

[0076] After the intelligent operation server determines that the late handwriting sequence belongs to the first stroke, it determines the corresponding predicted sub-segment based on the overlap of the sampling time range and the trajectory position deviation between the late handwriting sequence and the predicted handwriting segment. When the time range overlaps and the trajectory position deviation is not greater than the prediction matching tolerance, the two are determined to correspond, and prediction sub-segment failure information is generated. The prediction matching tolerance can be determined based on the coordinate resolution and the historical error distribution of the prediction model. When reconstructing the first stroke for display, the failed prediction sub-segment is not output. The prediction sub-segment failure information can be written into the supplementary data block and encrypted and stored with the supplementary data block; or it can be stored separately as a display control record associated with the supplementary encrypted data block.

[0077] When the late handwriting sequence only overlaps with a portion of the predicted handwriting segment's time range, only the predicted sub-segment that simultaneously satisfies the time range overlap condition and the trajectory position deviation condition is invalidated; the predicted portion that does not have a trajectory correspondence with the late handwriting sequence remains in an auxiliary display state within the supplementary recording waiting window, and becomes invalid after the supplementary recording waiting window ends because no real handwriting sampling point confirmation is obtained.

[0078] When there are multiple candidate prediction segments that meet the time range overlap condition, the intelligent operation server determines the trajectory position deviation and time overlap ratio between the late handwriting sequence and each candidate prediction segment, and determines the candidate prediction segment with the smallest trajectory position deviation and the highest time overlap ratio as the corresponding prediction segment; if it still cannot be uniquely determined, the candidate prediction segment is not immediately invalidated, but is processed according to the prediction part that has not been confirmed by the real handwriting sampling point after the supplementary recording waiting window ends.

[0079] The term "end of late handwriting data collection" in this application embodiment refers to the intelligent operation server determining that the conditions for completion of supplementary recording are met, or the supplementary recording waiting window ends and no longer receiving candidate handwriting sampling points for the first encrypted data block.

[0080] After the late handwriting data collection is completed, the intelligent task server invalidates the remaining parts of the predicted handwriting segments that have not yet been confirmed by the actual handwriting sampling points; therefore, the predicted content that has not been confirmed by the actual data will not continue to be displayed as valid task handwriting or participate in recognition after the supplementary recording is completed.

[0081] In one possible implementation, since the work terminal can send cached data in multiple upload batches, the intelligent work server obtains multiple sets of late handwriting sequences within the same supplementary recording waiting window; the intelligent work server sorts them according to the sampling time of the first handwriting sampling point in each late handwriting sequence.

[0082] The intelligent task server starts with the earliest late handwriting sequence at the sampling time and determines whether it is continuous with the end sampling point of the first stroke according to the forward continuity condition. For subsequent late handwriting sequences, it determines whether they are continuous with the previously selected late handwriting sequence according to the same forward continuity condition. When a set of late handwriting sequences does not meet the forward continuity condition, it stops selecting subsequent late handwriting sequences to avoid filling the first stroke with irrelevant data across the middle gap.

[0083] When the sampling time ranges of two sets of late handwriting sequences overlap, duplicate points can be avoided by using the sampling sequence number, upload batch identifier, or handwriting sampling point integrity check value. After deduplication, the selected late handwriting sequences with non-overlapping sampling time ranges are merged according to the sampling time to form a supplementary data block.

[0084] When all handwriting sampling points in the subsequent late handwriting sequence already exist in the previous late handwriting sequence, the subsequent late handwriting sequence is determined to be a completely repeated sequence, and no supplementary data block is written again.

[0085] When two sets of late handwriting sequences only partially overlap, the intelligent operation server starts from the first handwriting sampling point of the latter late handwriting sequence, identifies handwriting sampling points that are continuously repeated with the end of the former late handwriting sequence, does not write the identified continuously repeated handwriting sampling points into the supplementary data block again, and takes the first non-repeating handwriting sampling point and the subsequent handwriting sampling points as the valid part to be merged in the latter late handwriting sequence.

[0086] If there are still unreceived sampling numbers between two sets of late handwriting sequences after deduplication, or if the sampling time interval between them is greater than the continuous sampling time threshold, the two sets of late handwriting sequences will not be merged, and subsequent late handwriting sequences will be transferred to the verification process.

[0087] When a late handwriting sequence does not meet the forward continuity condition, it cannot be directly added to the first stroke. When a late handwriting sequence meets the forward continuity condition but still has the possibility of continuous connection with the starting sampling point of the second stroke, its attribution cannot be accurately determined based solely on the current time sequence and trajectory information. For the above two types of pending attribution situations, the intelligent operation server encrypts and stores the late handwriting sequence to obtain an encrypted data block to be verified. For noisy or abnormal data that neither meets the forward continuity condition nor has the possibility of connection with the second stroke, it can be stored in an abnormal isolation data block and will not enter the stroke attribution verification process.

[0088] During the pending verification period, the intelligent operation server will not use the late handwriting sequence for the display, recognition, correction or other data reconstruction processing of the first or second stroke; the encrypted data block to be verified can establish candidate associations with the data blocks corresponding to the first encrypted data block and the second stroke respectively.

[0089] When a late handwriting sequence enters the pending verification state, the supplementary recording waiting window that has already been started for the first encrypted data block is not paused; if it is confirmed that the late handwriting sequence belongs to the first stroke and the supplementary recording status indicator is still in the allowed supplementary recording state before the supplementary recording waiting window ends, it is converted into a sequence to be supplemented.

[0090] When the supplementary recording status flag has been updated to the prohibited supplementary recording status, and it is confirmed that the late handwriting sequence belongs to the first stroke, the supplementary recording entry is not reopened. Instead, a late verification result record associated with the first encrypted data block is generated, and the reconstructed first stroke remains unchanged.

[0091] When the intelligent operation server receives stroke identifiers, precursor sampling point identifiers, or manual confirmation results uploaded by the operation terminal, it re-determines the ownership of the handwriting sequence to be verified. If it is determined to belong to the first stroke and the supplementary recording status is still in the allowed supplementary recording status, it is converted into a sequence to be supplemented. The preset verification period can be determined according to the operation submission deadline, the automatic grading start time, or the system archiving cycle. If the ownership cannot be determined after the preset verification period, the encrypted data block to be verified is archived and the reason for not using it is recorded.

[0092] In this embodiment, the nominal sampling period of the work terminal is 8 milliseconds; the event time of the first stroke lifting event is 1000 milliseconds, and the event time of the second stroke finishing event is 1080 milliseconds; after receiving the second stroke finishing event, the intelligent work server receives three handwriting sampling points with sampling times of 976 milliseconds, 984 milliseconds, and 992 milliseconds, respectively.

[0093] The three handwriting sampling points have the same job session identifier, writing page identifier, terminal identifier, and timer identifier as the first stroke, and the sampling time is earlier than the time corresponding to the pen lifting event of the first stroke; the time interval between the three sampling points is 8 milliseconds, thus forming a set of late handwriting sequences.

[0094] The sampling time of the sampling point at the end of the first stroke is 968 milliseconds. The time interval between the sampling point at the end of the first stroke and the first point of the late stroke sequence is 8 milliseconds. The spatial distance is less than the achievable distance of the corresponding continuous writing, and the direction of movement remains continuous, thus satisfying the forward continuity condition. The spatial distance between the end point of the late stroke sequence and the sampling point at the beginning of the second stroke is greater than the achievable distance at the beginning of the second stroke, and the sampling point at the beginning of the second stroke has the pressure increase characteristic of re-writing, thus satisfying the backward exclusion condition.

[0095] The tail data transmission completion identifier sent by the work terminal carries the first stroke lifting event identifier and the sampling point sequence number with a sampling time of 992 milliseconds; after the intelligent work server confirms that all handwriting sampling points up to the sampling sequence number have been received and there are no gaps in the preceding data, it determines that the supplementary recording completion condition is met, generates a supplementary encrypted data block, and updates the supplementary recording status identifier to the prohibited supplementary recording status; when reconstructing the first stroke, the three late handwriting sampling points are arranged after the end sampling point with a sampling time of 968 milliseconds and before the lifting event.

[0096] In this embodiment, the late handwriting sequence satisfies the temporal and spatial continuity condition with the end of the first stroke, but the spatial distance between the end of the late handwriting sequence and the starting sampling point of the second stroke is also small, and the direction of movement can be connected with the starting direction of the second stroke; the intelligent operation server determines that the late handwriting sequence has the possibility of being continuously connected with the first stroke and the second stroke.

[0097] The intelligent operation server does not directly add the late handwriting sequence into any stroke, but instead encrypts and stores it to obtain an encrypted data block to be verified. Subsequently, the operation terminal uploads the preceding sampling point identifier of the late handwriting sequence, which points to the sampling point at the end of the first stroke. Based on this, the intelligent operation server determines that the late handwriting sequence belongs to the first stroke and enters the supplementary recording process when the supplementary recording status is still in the supplementary recording status.

[0098] In this embodiment, the first batch of late handwriting sequences includes three handwriting sampling points, and the second batch of late handwriting sequences includes two handwriting sampling points. After the first batch of late handwriting sequences arrives, the intelligent operation server determines that it belongs to the first stroke, but because the supplementary recording waiting window has not yet ended, it does not immediately generate a supplementary encrypted data block.

[0099] The second batch of late handwriting sequences then arrives within the supplementary recording waiting window. Its first handwriting sampling point is continuous with the last handwriting sampling point of the first batch of late handwriting sequences according to the forward continuity condition, and the sampling time ranges of the two batches of data do not overlap. The intelligent operation server merges the two batches of late handwriting sequences according to the sampling time. When the supplementary recording waiting window expires, and after confirming that there are no unreceived previous handwriting sampling points based on the sampling sequence number and the uploaded batch verification result, the intelligent operation server generates a supplementary encrypted data block at once and updates the supplementary recording status flag to the prohibited supplementary recording status.

[0100] In one possible implementation, the above method is executed by a smart job server, which includes a processor and a memory. The memory stores at least one computer program, which is loaded and executed by the processor to implement the operations described in the above method embodiments.

[0101] In one possible implementation, a computer-readable storage medium is also provided, wherein at least one computer program is stored therein, the computer program being loaded and executed by a processor to perform the operations described in the above method embodiments.

[0102] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by program instructions related to hardware; the above descriptions are merely exemplary embodiments of this application and are not intended to limit the scope of protection of this application.

Claims

1. A method for encrypted storage and online control of intelligent operation data blocks, characterized in that, include: In response to the pen lifting event of the first stroke, the received first stroke handwriting data is formed and encrypted and stored as the first encrypted data block; After the second stroke begins, obtain the sequence of late handwritings that arrive after the second stroke begins and whose sampling time is before the pen lifting event; Based on the continuity between the late handwriting sequence and the end of the first stroke and the beginning of the second stroke, it is determined that the late handwriting sequence belongs to the first stroke. While keeping the ciphertext content of the first encrypted data block unchanged, the late handwriting sequence and its supplementary recording position in the first stroke are used to form a supplementary encrypted data block associated with the first encrypted data block, and the first stroke is reconstructed based on the first encrypted data block and the supplementary encrypted data block.

2. The method according to claim 1, characterized in that, The acquisition of the late handwriting sequence includes: From the handwriting sampling points received after the second stroke, candidate handwriting sampling points that have the same job session identifier, writing page identifier, terminal identifier, and time sequence identifier as the first stroke, and whose sampling time is earlier than the time corresponding to the pen lifting event of the first stroke, are selected; the candidate handwriting sampling points are sorted according to the sampling time. Candidate handwriting sampling points whose time intervals between adjacent sampling times satisfy the continuous sampling condition are grouped into a set of late handwriting sequences, and different late handwriting sequences are divided at positions where the time intervals do not satisfy the continuous sampling condition.

3. The method according to claim 2, characterized in that, When the late handwriting sequence satisfies the forward continuity condition with the end of the first stroke, it is determined that the late handwriting sequence can continue the first stroke. The forward continuity condition includes: the sampling time interval between the end sampling point of the first stroke and the first sampling point of the late stroke sequence is not greater than the continuous sampling time threshold. The spatial distance between the end sampling point and the first handwriting sampling point is not greater than the continuous writing reachable distance; and at least one of the changes in the direction of movement and the changes in writing pressure between the end sampling point and the first handwriting sampling point satisfies the corresponding continuity condition; The achievable distance for continuous writing is determined based on the sampling time interval, the upper limit of writing speed, and the coordinate quantization error margin.

4. The method according to claim 3, characterized in that, When the late handwriting sequence meets the backward exclusion condition with the starting end of the second stroke, it is determined that the late handwriting sequence does not belong to the second stroke; The backward exclusion condition includes: the spatial distance between the last handwriting sampling point of the late handwriting sequence and the starting sampling point of the second stroke is greater than the initial reachable distance of the second stroke; and At least one of the following, namely the connection relationship of the movement direction between the last handwriting sampling point and the starting sampling point or the change in writing pressure, indicates that the second stroke was formed by a re-stroke. The starting reachable distance of the second stroke is determined based on the sampling time interval between the last stroke sampling point and the starting sampling point, the upper limit of writing speed after rewriting, and the coordinate quantization error margin.

5. The method according to claim 4, characterized in that, The acquisition of the late handwriting sequence includes: In response to the stroke completion event of the second stroke, when the first encrypted data block is in the allowed supplementary recording state, a supplementary recording waiting window is started for the first encrypted data block; within the supplementary recording waiting window, the candidate handwriting sampling points continue to be received; When no preceding handwriting sampling point that has not yet been received is detected, and any of the following termination trigger conditions are met, the supplementary recording completion condition is determined to be met: a valid tail data transmission completion identifier is received; The last sampling moment of the late handwriting sequence falls within the allowable error range before the lifting event of the first stroke; or the supplementary recording waiting window expires; The tail data transmission completion identifier includes the event identifier of the pen lifting event and the sampling sequence number of the last handwriting sampling point before the pen was lifted. The tail data transmission completion identifier is determined to be valid when handwriting sampling points up to the sampling sequence number have been received.

6. The method according to claim 5, characterized in that, The supplementary encrypted data block includes: The late handwriting sequence, the supplementary recording position, the first encrypted data block identifier, the first encrypted data block integrity check value, the supplementary data block identifier, and the supplementary recording transaction identifier; The supplementary encrypted data block has a supplementary data block integrity check value, which at least covers the late handwriting sequence, the supplementary recording position, the first encrypted data block identifier, the first encrypted data block integrity check value, and the supplementary recording transaction identifier; When reconstructing the first stroke, the integrity of the first encrypted data block and the supplementary encrypted data block are verified respectively. When the identifier of the first encrypted data block and the integrity verification value of the first encrypted data block are consistent with the first encrypted data block to be reconstructed, the late handwriting sequence is arranged to the supplementary recording position.

7. The method according to claim 1, characterized in that, The method further includes: Based on the actual handwriting sampling points at the end of the first stroke, a predicted handwriting segment for real-time display is generated, and the predicted handwriting segment is stored as display auxiliary data in association with the first encrypted data block. After determining that the late handwriting sequence belongs to the first stroke, the predicted sub-segment corresponding to the late handwriting sequence is determined based on the overlap of the sampling time range and the trajectory position deviation between the late handwriting sequence and the predicted handwriting segment. Failure information for the predicted sub-segment is generated, and the predicted sub-segment is not output when reconstructing the first stroke for display; After the late handwriting data collection is completed, the remaining parts of the predicted handwriting segment that have not been confirmed by the actual handwriting sampling points are invalidated.

8. The method according to claim 5, characterized in that, The supplementary recording status identifier associated with the first encrypted data block includes supplementary recording allowed status, supplementary recording in progress status, and supplementary recording prohibited status; After determining that the late handwriting sequence belongs to the first stroke and meets the supplementary recording completion condition, the supplementary recording status identifier is updated from the supplementary recording allowed status to the supplementary recording in progress status through a condition update operation; For the same first encrypted data block, only one supplementary record transaction is allowed to obtain supplementary record processing permission through the condition update operation, and other supplementary record transactions stop generating supplementary encrypted data blocks for the first encrypted data block; After the supplementary encrypted data block is successfully persisted, the supplementary recording status flag is updated to the prohibited supplementary recording status; if the supplementary encrypted data block fails to persist, the supplementary recording status flag is restored to the allowed supplementary recording status. When the supplementary recording waiting window ends and there is no late stroke sequence belonging to the first stroke, the supplementary recording status flag is updated to the prohibited supplementary recording status, and the supplementary encrypted data block is not generated.

9. The method according to claim 5, characterized in that, When multiple sets of late handwriting sequences are obtained within the supplementary recording waiting window: The late handwriting sequences are sorted according to their sampling times. When the sampling time ranges of two late handwriting sequences overlap, the handwriting sampling points that are continuously repeated in the later late handwriting sequence and the previous late handwriting sequence are identified. The continuously repeated handwriting sampling points are not written into the supplementary data block again. Starting from the earliest late handwriting sequence at the sampling time, late handwriting sequences that satisfy the forward continuity condition, can be continuously connected with the end sampling point of the first stroke or the previously selected late handwriting sequence, and have no sampling sequence gaps after excluding duplicate handwriting sampling points are selected in sequence. When a late handwriting sequence that does not meet the aforementioned forward continuity condition or has a sampling sequence gap occurs, stop selecting subsequent late handwriting sequences. The selected late handwriting sequences are merged according to the sampling time, and the supplementary encrypted data block is generated based on the merged late handwriting sequences.

10. The method according to claim 4, characterized in that, The method further includes: When the late handwriting sequence satisfies the forward continuity condition but does not satisfy the backward exclusion condition, or does not satisfy the forward continuity condition but still has the possibility of continuous connection with the starting sampling point of the second stroke, the late handwriting sequence is encrypted and stored to obtain an encrypted data block to be verified. During the period when the late handwriting sequence is in the pending verification state, it will not be used to reconstruct the first stroke or the second stroke, and its attribution will be re-determined based on the supplemented uploaded stroke identifier, the predecessor sampling point identifier, or the manual confirmation result. When the late handwriting sequence does not meet the forward continuity condition and does not have the possibility of being connected continuously with the second stroke, the late handwriting sequence is stored in the abnormal isolation data block. When it is confirmed that the handwriting sequence to be verified belongs to the first stroke after the supplementary recording state for the first encrypted data block has been closed, the supplementary recording entry is not reopened, but a late verification result record associated with the first encrypted data block is generated.

Images