A test paper mixed scanning method and system
By dynamically adjusting the number of threads for identification of test papers and assigning subjective questions in real time, the problems of random order and cross-machine scanning of test papers are solved, realizing instant scan and efficient utilization of resources are achieved, and the school's marking efficiency and resource utilization are improved.
Patent Information
- Application Number
- CN202411171082.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-26
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-26
AI Technical Summary
In the process of online school paper marking, test papers are prone to missed collection, wrong papers, staged paper collection or deliberately disruption, resulting in out-of-order, resulting in client being unable to identify them according to the entire scan, affecting the efficiency of marking, and scanning across machines. It cannot be combined into a whole, which increases the burden and cost of the server.
Use dynamic thread calculation formula to adjust the number of threads to identify the test paper based on hardware resources, allocate subjective questions to score priority teachers in real time, and download and store test questions slices using asynchronous methods to realize the sweep and read test papers, make full use of local machine resources, and reduce services End pressure.
It realizes scanning and reading immediately when the test paper is incomplete, improves the efficiency of marking papers, makes full use of local machine resources, reduces server costs, supports mixed scanning scenarios for cross-sectional test papers, and improves resource utilization.
Smart Images

Figure CN118672793B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of examination paper marking, and in particular relates to an examination paper mixed scanning method and system. Background Art
[0002] When schools grade papers online, they usually scan each paper together in order. The scanning and recognition client starts from the first paper and takes N papers as a whole to recognize the paper, split the questions, and combine the questions across pages. However, the following scenarios may occur during the process of collecting and arranging the papers:
[0003] 1. Missing papers: Sometimes due to negligence, some papers are not collected, which affects the subsequent marking.
[0004] 2. Wrong test paper: When sorting the test papers, the last test paper of the previous student may be swapped with the first test paper of the next student, resulting in the wrong test paper.
[0005] 3. Collecting papers in stages: The teacher may collect the first paper first, and then collect the second paper at the same time, which increases the possibility of confusion among the papers.
[0006] 4. Deliberately shuffling the test paper: In order to prevent cheating, the test paper may be deliberately shuffled.
[0007] 5. Cross-machine scanning: The test papers should be divided into multiple batches and scanned on different machines to speed up the scanning process.
[0008] These scenarios will cause the test papers to be out of order, and the client will not be able to scan and identify the test papers as a whole, which will seriously affect the subsequent teacher's marking work. The existing scanning and recognition method solves this problem by having the client recognize the test papers in stages and then combine them into a whole. First, the out-of-order test papers are scanned locally, and the test number of each is first identified. After all the recognition is completed, the test papers are combined into a whole on the client locally, and then the whole copy is identified, split and combined, and finally the complete scan is completed. This technical solution can solve the scanning scenario where the test papers are all on one machine. If a test paper is scanned across machines, the test papers cannot be combined into a whole copy, and the complete scan cannot be completed.
[0009] Another scanning and recognition method is that the client is only responsible for scanning, and the scanned image files of the test paper are uploaded to the server. The server determines whether a test paper has been scanned completely, and then performs the recognition work of the entire test paper on the server. This solution can solve mixed scanning and cross-machine scanning problems, but the recognition function is completely migrated to the server, and the pressure of local machine dispersion is all transferred to the server, wasting the local computing power resources of the scanning machine and increasing the cost and instability of the server system.
[0010] The present invention designs a method for mixed scanning of test papers, which meets the scenario where schools need mixed scanning. The scanned sheets can be marked without waiting for the complete test paper to be scanned, thereby realizing instant scanning and marking, and improving marking efficiency. Summary of the invention
[0011] In order to solve the above technical problems, the present invention provides a test paper mixed scanning method and system, which are used to solve the technical problems in the prior art.
[0012] In the first aspect, the invention provides the following technical solution, a method for mixed scanning of test papers, the method comprising:
[0013] The number of test paper recognition threads is dynamically adjusted based on the current hardware resource utilization rate using a dynamic thread calculation formula, and the test paper is scanned based on the adjusted number of threads to form a test paper image;
[0014] Identify the test questions in each of the test paper images and divide them into question block images, determine whether the test questions in all the question block images are complete test questions, and if the test questions in the question block images are complete test questions, divide the question block images into objective question images and subjective question images;
[0015] The priority function and the heartbeat detection of the examiner are used to calculate the priority of each examiner based on the examiner's schedule, personal ability, title and number of tasks. Based on the examiner's priority and the question dimension to which the subjective question picture is to be assigned, the subjective question picture is assigned to the corresponding examiner in real time for scoring;
[0016] The standard answers set for the objective questions are compared with the options of the objective questions, and the scores of the objective question pictures are obtained based on the comparison results. The scores of all the objective question pictures and the subjective question pictures corresponding to the test paper picture are summed up to obtain the student's total score.
[0017] Compared with the prior art, the beneficial effects of the present application are as follows: by allocating the subjective question images to the corresponding grading teachers for grading in real time based on the priority of the grading teachers and the dimensions of the test questions to which the subjective question images are to be allocated, if the test paper is incomplete, it will not affect the grading of the test questions of the scanned test papers, thus achieving true scan-and-read, without having to wait for a certain number of copies to be scanned or for the complete data of the students to be scanned before grading assignment, thereby improving the school's grading efficiency; by dynamically adjusting the number of test paper recognition threads based on the current hardware resource utilization rate using a dynamic thread calculation formula, the computing power resources of the local machine are fully utilized to give priority to a part of the recognition work, and in the case where some test questions span across sheets, a small amount of server resources are used to combine and identify the test questions across sheets, thereby supporting the school's mixed scanning scenarios of test papers, improving resource utilization, and reducing service costs.
[0018] Furthermore, the dynamic thread calculation formula includes:
[0019]
[0020] In the formula, Optimal Threads represents the optimal number of threads, N represents the number of CPU cores, M represents the total amount of available memory, and T represents the memory consumption of each thread. Indicates the current CPU usage. Represents the current memory usage, Expressed as the maximum allowed CPU usage, Expressed as the maximum allowed memory usage.
[0021] Furthermore, after the step of determining whether the test questions in all the question block images are complete test questions, the method further includes:
[0022] If the test question in the question block image is not a complete test question, a portion of the test question across sheets of the question block image is cut out to form test question slices, and the test question slices are stored;
[0023] The test question slices are downloaded and stored based on the local area network in an asynchronous manner, and the downloaded test question slices are stored in the disk directory of the corresponding student.
[0024] Determine whether all the test question slices corresponding to a test question are stored in the disk directory. If all the test question slices corresponding to a test question are stored in the disk directory;
[0025] Then, all the test question slices are merged using an image conversion tool to form a merged test question;
[0026] Determine whether the fusion test question is an objective question. If the fusion test question is an objective question, perform filling identification on the fusion test question to identify and obtain filling options for the fusion test question.
[0027] Furthermore, after the step of forming the test paper image, the method further includes:
[0028] Obtaining a first coordinate of a positioning block of a standard answer sheet, and calculating a search area based on the first coordinate and an expansion coefficient;
[0029] Obtaining second coordinates of four corner points of the positioning block of the test paper image based on the search area using a contour detection algorithm;
[0030] Calculate a first scaling factor based on the second coordinate and the first coordinate using a scaling calculation formula;
[0031] Calculating the third coordinates of the black and white page number block of the test paper image based on the first scaling factor and the coordinates of the black and white page number block of the standard answer sheet;
[0032] Obtaining blocks of black and white page number blocks of the test paper image based on the third coordinates, and obtaining page numbers of the test paper image based on the pixel ratio of each block using binary numbers and decimal numbers;
[0033] Recognize the test number of all the scanned test paper images, and after the recognition is completed, determine the student and the number of the test paper images corresponding to the student;
[0034] If the number of test papers does not correspond to the number of examinees, it is checked whether all test papers have been scanned. If all test papers have been scanned, the corresponding students will be marked as missing.
[0035] Further, after the step of obtaining the first scaling factor, the method includes:
[0036] The second scaling factor is calculated based on the auxiliary positioning point of the test paper image using the scaling calculation formula;
[0037] Calculating a difference value based on the second scaling factor and the first scaling factor using a relative difference formula;
[0038] It is determined whether the difference value is greater than a threshold value. If the difference value is greater than the threshold value, the second scaling factor is used to calculate the page number.
[0039] Furthermore, the scaling calculation formula includes:
[0040]
[0041] In the formula, Expressed as the second scaling factor or the first scaling factor, , , and Respectively expressed as the horizontal and vertical coordinates of the two second coordinates, , and Respectively represented as the abscissa and ordinate of the two first coordinates;
[0042] The relative difference formula includes:
[0043]
[0044] In the formula, represents the first scaling factor, The second scaling factor, Expressed as difference value.
[0045] Furthermore, the priority function includes:
[0046] priority = w1 * availability + w2 * rank - w3 * task_count
[0047] Among them, w1 represents the weight of the timetable availability; w2 represents the weight of the examiner's personal ability and title; w3 represents the weight of the number of tasks; availability represents that if the current time is within the examiner's available time period, availability=1, otherwise availability=0; rank represents the examiner's personal ability and title rank, ranging from 1 to 5; task_count represents the number of tasks for the current examiner;
[0048] The heartbeat detection of paper marking includes:
[0049] Determine whether the marking teacher has not made new markings for more than the priority lowering threshold, if the marking teacher has not made new markings for more than the priority lowering threshold;
[0050] It is determined whether the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking. If the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking, the marking teacher will be removed from the priority queue and the marking teacher and the administrator will be notified.
[0051] In a second aspect, the invention provides the following technical solution, a test paper mixed scanning system, the system comprising:
[0052] A forming module is used to dynamically adjust the number of test paper recognition threads based on the current hardware resource utilization rate by using a dynamic thread calculation formula, and scan the test paper based on the adjusted number of threads to form a test paper image;
[0053] A distinguishing module is used to identify the test questions in each of the test paper images and divide them into question block images, determine whether the test questions in all the question block images are complete test questions, and if the test questions in the question block images are complete test questions, distinguish the question block images into objective question images and subjective question images;
[0054] The scoring module is used to calculate the priority of each marking teacher based on the marking teacher's class schedule, personal ability, title and number of tasks by using the priority function and marking heartbeat detection, and to assign the subjective question picture to the corresponding marking teacher in real time for scoring based on the marking teacher's priority and the question dimension to which the subjective question picture is to be assigned;
[0055] A summing module is used to compare the standard answers set for objective questions with the options of the objective questions, obtain the score of the objective question picture based on the comparison result, and sum the scores of all the objective question pictures and the subjective question pictures corresponding to the test paper picture to obtain the student's total score.
[0056] In a third aspect, the invention provides the following technical solution: a computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the test paper mixed scanning method as described above when executing the computer program.
[0057] In a fourth aspect, the invention provides the following technical solution: a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the test paper mixed scanning method as described above. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0059] Figure 1 A flow chart of a test paper mixed scanning method provided by a first embodiment of the present invention;
[0060] Figure 2 A structural block diagram of a test paper mixed scanning system provided by a second embodiment of the present invention;
[0061] Figure 3 A schematic diagram of the hardware structure of a computer provided for the third embodiment of the present invention.
[0062] The embodiments of the present invention will be further described below with reference to the accompanying drawings. DETAILED DESCRIPTION
[0063] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout are the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the embodiments of the present invention, and should not be construed as limiting the present invention.
[0064] In the description of the embodiments of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.
[0065] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0066] Embodiment 1
[0067] In the first embodiment of the present invention, see Figure 1 , a test paper mixed scanning method, comprising the following steps S01 to S04:
[0068] S01, dynamically adjusting the number of test paper recognition threads based on the current hardware resource usage rate using a dynamic thread calculation formula, and scanning the test paper based on the adjusted number of threads to form a test paper image;
[0069] Specifically, the dynamic thread calculation formula includes:
[0070]
[0071] In the formula, Optimal Threads represents the optimal number of threads, N represents the number of CPU cores, M represents the total amount of available memory, and T represents the memory consumption of each thread. Indicates the current CPU usage. Represents the current memory usage, Expressed as the maximum allowed CPU usage, Expressed as the maximum allowed memory usage.
[0072] In this embodiment, a scanner is used to scan the test paper. The scanning client first detects the hardware configuration and storage of the computer to ensure smooth operation during the scanning and recognition process. Based on the current hardware resource utilization, the number of test paper recognition threads is dynamically adjusted to improve the recognition speed. The default number of threads is 10 (indicates that 10 threads will simultaneously recognize and process the scanned image). The dynamic thread calculation formula is:
[0073]
[0074] Where N is the number of CPU cores, M is the total available memory (unit: MB), T is the memory consumption of each thread (unit: MB), : Current CPU usage (between 0 and 1), : Current memory usage (between 0 and 1), : Maximum allowed CPU usage (between 0 and 1), : Maximum allowed memory usage (between 0 and 1).
[0075] Specifically, after the step of forming the test paper image, the method further includes:
[0076] S11, obtaining a first coordinate of a positioning block of a standard answer sheet, and calculating a search area based on the first coordinate and an expansion coefficient;
[0077] S12, using a contour detection algorithm based on the search area to obtain second coordinates of four corner points of the positioning block of the test paper image;
[0078] S13, using a scaling calculation formula to calculate based on the second coordinate and the first coordinate to obtain a first scaling factor;
[0079] S14, calculating the third coordinates of the black and white page number block of the test paper image based on the first scaling factor and the coordinates of the black and white page number block of the standard answer sheet;
[0080] S15, obtaining blocks of black and white page number blocks of the test paper image based on the third coordinates, and obtaining page numbers of the test paper image based on the pixel ratio of each block using binary numbers and decimal numbers;
[0081] S16, identifying the test numbers of all the scanned test paper images, and after the identification is completed, determining the student and the number of the test paper images corresponding to the student;
[0082] S17, if the number of test papers does not correspond to the number of test takers, then check whether all the test papers have been scanned. If all the test papers have been scanned, the corresponding students will be marked as missing.
[0083] Optionally, after the step of obtaining the first scaling factor, the method includes:
[0084] The second scaling factor is calculated based on the auxiliary positioning point of the test paper image using the scaling calculation formula;
[0085] Calculating a difference value based on the second scaling factor and the first scaling factor using a relative difference formula;
[0086] It is determined whether the difference value is greater than a threshold value. If the difference value is greater than the threshold value, the second scaling factor is used to calculate the page number.
[0087] Specifically, the scaling calculation formula includes:
[0088]
[0089] In the formula, Expressed as the second scaling factor or the first scaling factor, , , and Respectively expressed as the horizontal and vertical coordinates of the two second coordinates, , and Respectively represented as the abscissa and ordinate of the two first coordinates;
[0090] The relative difference formula includes:
[0091]
[0092] In the formula, represents the first scaling factor, The second scaling factor, Expressed as difference value.
[0093] In this embodiment, after scanning, the client obtains two front and back pictures generated for each test paper. The front side has a fixed student information column and test prompt related information area. The original answer sheet picture is used to capture the fixed information column area from the original answer sheet picture as a template image. The template image is searched for in the two scanned pictures using the opencv template matching algorithm to obtain the searched matrix information. The matrix includes matching position information and similarity scores. Since the features of the test information column area are relatively obvious, the score threshold is set to 0.8 here. If the score is greater than or equal to 0.8, the picture is front, otherwise the picture is back.
[0094] After confirming the front image, search for the positioning blocks. The positioning blocks are black squares located at the four corners of the answer sheet, which are used to locate the valid area of the answer sheet. From the information of the standard answer sheet, we get the standard coordinate positions of the four positioning blocks. In order to avoid searching for the positioning blocks on the entire image, first confirm the positioning block search area and search separately in this area. Taking the upper left corner positioning block as an example, the upper left coordinates of the upper left rectangle are x1, y1, the width and height of the rectangular positioning block are w and h respectively, and the original coordinates of the four corner points are:
[0095] Upper left corner: (x1, y1), upper right corner: (x1+w, y1), lower left corner: (x1, y1+h) and lower right corner: (x1+w, y1+h);
[0096] Although the client automatically sets the scanner parameters to 100% zoom before scanning, there will still be zooming and pixel offset due to the accuracy of the scanner itself, image processing software or hardware limitations. In order to consider zooming and rotation, first calculate an expanded rectangular boundary. Set the expansion factor to scale_factor=1.2, and the four corner points after expansion are calculated as follows:
[0097] (1) Coordinates of the center point of the extended rectangle: center_x = x1+ , center_y = y1+ ;
[0098] (2) Width after expansion: w1 = w*scale_factor;
[0099] (3) Height after expansion: h1 = h*scale_factor;
[0100] (4) Upper left corner after expansion: lt_x = center_x - ,lt_y = center_y- ;
[0101] (5) Upper right corner after expansion: rt_x = center_x + ,lt_y = center_y- ;
[0102] (6) Lower left corner after expansion: lb_x = center_x - ,lt_y = center_y+ ;
[0103] (7) Lower right corner after expansion: lb_y = center_x + ,lt_y = center_y+ ;
[0104] In this way, the search area of one positioning block is determined, and the search areas of the other three positioning blocks are obtained in the same way. Then, the contour detection algorithm of OpenCV is used to find the positioning block in the search area, and the coordinates of the upper left corner positioning block of the scanned test paper (x1, y1) are obtained.
[0105] After obtaining the four positioning points of the scanned test paper image, the distance between the two positioning points on the horizontal axis and the distance between the two positioning points on the vertical axis are calculated, and then compared with the two distances of the original answer sheet to obtain the image zoom factor. Assume that the standard answer sheet positioning points are: (x1, y1), (x2, y2), (x3, y3), (x4, y4);
[0106] Scan the image positioning points: (x1', y1'), (x2', y2'), (x3', y3'), (x4', y4'). If the scanned answer sheet is uniformly scaled, a set of distances can be calculated for comparison. The distance calculation formula is:
[0107]
[0108] The distance d1 between the two top positioning points of the standard answer sheet and the distance d2 between the two top positioning points of the scanned answer sheet are calculated by the formula, and the scaling ratio scale = d2 / d1 is obtained. If the four positioning points are not obtained on the test paper, the other two positioning points are calculated based on the geometric characteristics of the rectangle using the affine transformation matrix according to at least two positioning points that have been obtained. If the calculated coordinates exceed the image boundary, it means that the image scan difference is too large and the next step of recognition cannot be performed. The user is prompted to identify abnormalities and manual review of the abnormalities of the test paper is required.
[0109] After the positioning point recognition is completed, the page number recognition is performed. Here, the page uses binary black and white blocks to mark the page number, and only the front mark is required. The page number block has three small rectangular blocks. The third block is black, indicating 1, and the first two are white, indicating 0. The binary 001 corresponds to 1 in decimal, indicating the first page. The default is three blocks, which can support the page number marking of 4 answer sheets, meeting the card quantity scenarios of most exams. If necessary, it can be flexibly expanded to 4 blocks, which can support the page numbers of 16 answer sheets. Compared with the traditional digital page number at the bottom, the recognition of binary page numbers only needs to be calculated through image processing and simple recognition. Digital page numbers also require additional ORC for digital recognition. Including many scenarios, the visual format and position of the bottom page number may vary due to different designs of the answer sheet, which increases the complexity of recognition. If there is pollution around the number to generate noise, it will especially affect the accuracy of the page number. Even if the page number area of the binary black and white block is defiled, as long as the block is still visible, the accuracy of recognition will not be greatly affected. The binary black and white page number block is inside the rectangle of the four positioning blocks. According to the scaling ratio and affine transformation matrix A obtained in the previous step, the coordinates of the black and white page number block on the scanned image are calculated using the coordinates of the black and white page number block on the standard answer sheet. Then the page number is identified. The recognition of the page number here is the same as the recognition of the positioning block. The rectangular blocks are first identified by contour detection, and then the proportion of black pixels in each block is calculated. If the proportion is greater than the set threshold of 80%, it is represented as a black block and marked as 1. Otherwise, it is a white block and marked as 0. According to the order of the x coordinates, it is combined into a binary number and finally converted into a decimal number to get the page number.
[0110] After identifying the page number, the question blocks are identified and segmented. The question blocks are identified based on the four points of the rectangle of each question block on the standard answer sheet. The scaling and affine transformation matrices calculated previously are used to calculate the corresponding coordinates of the four points on the scanned image, and then each question block image is segmented based on these coordinates.
[0111] If the scanning is performed with uniform scaling and the scaling ratios of each position are consistent, the above image can be better segmented into the pictures of each question block. In reality, differences in student scanning environments, hardware, and paper conditions will lead to inconsistent horizontal or vertical scaling ratios. If a uniform scaling ratio is still used to calculate the coordinates on the scanned test paper, it will lead to deviations. Auxiliary positioning points are added here to calculate whether the horizontal or vertical scaling ratios are consistent. After identifying the auxiliary positioning points on the left, the distance formula is used to calculate the distance between the two sets of auxiliary positioning points and the upper left corner positioning point, and multiple scaling ratios are obtained respectively. x , scale y , and then use the relative difference formula:
[0112]
[0113] Calculate the scaling difference. If the scaling difference is less than the set threshold of 0.01 (1%), the scaling is considered uniform. Then the previous method can be used to calculate the coordinate mapping of other positioning points. If it exceeds the threshold, it is considered that the scaling difference in the horizontal or vertical axis is large.
[0114] The scaling difference is large across the entire horizontal and vertical axis span, but if the two points are close, the scaling of the two point areas is also very close. At this time, to calculate the mapping of the quasi-card image coordinates to the scanned card image coordinates, you need to find a nearest auxiliary point to calculate the horizontal and vertical scaling ratios of the auxiliary point. Then use the scaling ratio of the auxiliary point to perform point mapping calculations to obtain the points on the scanned image. Minimize the point coordinate recognition deviation caused by uneven scaling as much as possible.
[0115] After the recognition is completed, the student and the test paper number are determined, and then the test questions on the test paper number are recognized and cut into small question block images.
[0116] In this embodiment, the scanned test paper image is identified by the test number and page number on the local client, and the student and the number of test papers are determined after the identification is completed. The detection device is used to check whether the student's test paper is scanned completely. If any test paper is not reported, the corresponding student is marked as missing. If all the student test papers are reported, the corresponding student is marked as scanned, which is convenient for the scanning teacher to check the complete status of the student test paper in real time in the scanning progress module to avoid omissions.
[0117] S02, identifying the test questions in each of the test paper images and dividing them into question block images, determining whether the test questions in all the question block images are complete test questions, and if the test questions in the question block images are complete test questions, dividing the question block images into objective question images and subjective question images;
[0118] Specifically, after the step of determining whether the test questions in all the question block images are complete test questions, the method further includes:
[0119] S21, if the test question in the question block image is not a complete test question, then partially cutting out the test question across sheets of the question block image to form test question slices, and storing the test question slices;
[0120] S22, downloading the stored test question slices based on the local area network in an asynchronous manner, and storing the downloaded test question slices in the disk directory of the corresponding student.
[0121] S23, judging whether all the test question slices corresponding to a test question are stored in the disk directory, if all the test question slices corresponding to a test question are stored in the disk directory;
[0122] S24, using an image conversion tool to merge all the test question slices to form a merged test question;
[0123] S25, determining whether the fusion test question is an objective question. If the fusion test question is an objective question, performing filling identification on the fusion test question to identify and obtain filling options for the fusion test question.
[0124] In this embodiment, if the test questions are incomplete across sheets, the part of the test questions (test question slices) that span across sheets will be cut out, and the state of the test questions will be marked as the state of spanning sheets to be completed. After each sheet is identified, all the partial test questions images on a test paper are stored in the minio object storage, and then the identified question number, test question slice image address, test question complete status, option data, test number information, test paper sheet number and other information are encapsulated into a JSON data structure and reported to the test question data receiving module of the test server, without considering and waiting for whether the student's other sheets have been scanned.
[0125] After receiving the data, the test data receiving module will first classify the question blocks. If the question blocks marked as normal are objective questions, they will be added to the scoring module in real time through MQ messages; if they are subjective questions, they will be added to the question block allocation module. Question blocks marked as incomplete across sheets will be added to the question block merging module.
[0126] The question block merging module will monitor all reported incomplete question block data. After receiving the question block message, the question block merging module will first download the partial image of this question block from the minio object storage in an asynchronous manner. Considering the download speed and bandwidth, to avoid the influence of the teacher's access speed to the marking system, the download address is mapped to the LAN here, and the image is quickly downloaded using the intranet IP. The downloaded image is stored in the disk directory named by the corresponding student ID for cache. The cache directory design structure is: test ID> student ID> question number> sheet number. If the corresponding multiple test papers of a student's cross-sheet test paper are scanned, the cross-sheet test paper image merging task will be triggered. The merging task finds all the partial slices of the test paper from the above cache directory, and combines multiple partial images in the order of sheets using imagemagick (image conversion tool) to obtain a complete question block and upload it to the minio object storage. If the merged question block is an objective question, it is necessary to perform an objective question filling recognition. After identifying the student's filled-in options, it will occur to the scoring module. If the merged question block is a subjective question, the question block is added to the question block allocation module through an MQ message for allocation.
[0127] The specific application scenarios are as follows:
[0128] Test paper image storage:
[0129] The marking system can store the segmented test paper images in MinIO, and each image can be managed and retrieved through a unique object key.
[0130] Concurrent access: MinIO's high concurrent processing capability allows multiple examiners to access and mark the test papers at the same time, improving the overall marking efficiency.
[0131] Automated processing: By integrating with MinIO, the marking system can automatically upload, classify and process test images, reducing the time and errors of manual operations.
[0132] Data backup and recovery: MinIO supports data backup and recovery functions. The examination paper marking system can regularly back up examination paper images to ensure data security and recoverability.
[0133] Minio object storage is introduced as follows:
[0134] 1. Efficient data structures and algorithms;
[0135] Object storage architecture: MinIO uses an object storage architecture, which does not rely on complex directory structures like traditional file systems. Each object has a unique key, and access operations are performed through key-value pairs, which simplifies the data management and access process.
[0136] Erasure Coding: MinIO uses Erasure Coding for data protection, which not only provides efficient data redundancy, but is also more efficient than traditional replication technology when recovering data. Erasure Coding splits data and distributes it across multiple disks, which can increase data reading speed while ensuring data security.
[0137] 2. Parallel processing;
[0138] Parallel I / O operations: MinIO supports parallel read and write operations and can handle multiple I / O requests at the same time. For the storage and reading of a large number of small images, MinIO can make full use of hardware resources to improve data throughput and access speed.
[0139] Multi-threaded processing: MinIO has designed a multi-threaded processing mechanism internally, which can fully utilize hardware performance in a multi-core CPU environment, perform concurrent data processing, and further improve the system's response speed.
[0140] 3. Efficient underlying storage engine;
[0141] Memory caching: MinIO uses memory caching technology to cache frequently accessed data in memory, reducing the number of disk accesses and thus speeding up data reading.
[0142] SSD support: For high performance requirements, MinIO can be configured to use SSD as the storage medium to further improve data reading and writing speeds.
[0143] 4. Network optimization;
[0144] Efficient network protocol: MinIO uses efficient HTTP / HTTPS protocols for data transmission and optimizes the network layer to reduce data transmission delays and improve data transmission efficiency.
[0145] Load balancing: In distributed deployment, MinIO can achieve load balancing, distribute requests to different nodes, avoid the performance bottleneck of a single node, and thus improve the throughput of the overall system.
[0146] 5. Data sharding and distribution;
[0147] Data sharding: MinIO divides large files into multiple small blocks for storage, and each small block can be accessed independently. For small images, multiple images can be combined into a large file for storage, which can reduce the overhead of metadata management and improve access efficiency.
[0148] Distributed storage: In cluster mode, MinIO can distribute data to different storage nodes to achieve distributed storage and access of data and improve the concurrent performance of data storage and retrieval.
[0149] Among them, using JSON to store the test paper recognition results, including the question block location, page number, answer, segmented image address, client recognition version number, etc., has the following advantages:
[0150] 1. Simplicity and readability;
[0151] Structured data: JSON uses key-value pairs to store data, making the data structure clear and easy to read and understand.
[0152] Concise format: Compared with formats such as XML, JSON has a more concise syntax, which reduces data redundancy and makes the file smaller and easier to read.
[0153] 2. Easy to parse and generate;
[0154] Wide support: JSON is a lightweight data exchange format. Almost all programming languages provide native or third-party libraries to support JSON parsing and generation. Whether it is front-end JavaScript or back-end languages such as Python, Java, Go, etc., JSON data can be easily processed.
[0155] Automated tools: There are a large number of tools and libraries that can help generate and parse JSON data, thus simplifying development work.
[0156] 3. Flexibility and scalability;
[0157] Dynamic structure: JSON supports nested objects and arrays, and can flexibly represent complex hierarchical structures. In test paper recognition, information such as question blocks, questions, and cutting paths can be organized hierarchically.
[0158] Extensibility: If new fields or data structures need to be added in the future, they can be easily added in JSON without affecting the existing data format.
[0159] 4. Cross-platform and language independent;
[0160] Cross-platform: JSON is a platform-independent data format that works across different operating systems and devices. Whether it's Windows, Linux, or a mobile device, JSON works consistently.
[0161] Language-independent: JSON is not dependent on a specific programming language, which allows data to be seamlessly transferred and shared between different systems and applications.
[0162] 5. Efficient transmission and storage;
[0163] Compactness: JSON is relatively compact and occupies less bandwidth and storage space than formats such as XML, which helps save resources during network transmission and storage.
[0164] Serialization and deserialization: JSON serialization and deserialization operations are very efficient and suitable for scenarios that require high-frequency data exchange.
[0165] S03, using the priority function and the heartbeat detection of the paper marking, based on the class schedule, personal ability, title and number of tasks of the paper marking teacher, the priority of each paper marking teacher is calculated, and based on the priority of the paper marking teacher and the question dimension of the subjective question picture to be assigned, the subjective question picture is assigned to the corresponding paper marking teacher in real time for scoring;
[0166] Specifically, the priority function includes:
[0167] priority = w1 * availability + w2 * rank - w3 * task_count
[0168] Among them, w1 represents the weight of the timetable availability; w2 represents the weight of the examiner's personal ability and title; w3 represents the weight of the number of tasks; availability means that if the current time is within the examiner's available time period, availability=1, otherwise availability=0; rank represents the examiner's personal ability and title rank, ranging from 1 to 5; task_count represents the number of tasks for the current examiner.
[0169] Specifically, the heartbeat detection of paper marking includes:
[0170] Determine whether the marking teacher has not made new markings for more than the priority lowering threshold, if the marking teacher has not made new markings for more than the priority lowering threshold;
[0171] It is determined whether the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking. If the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking, the marking teacher will be removed from the priority queue and the marking teacher and the administrator will be notified.
[0172] In this embodiment, the question block allocation module uses a priority queue and a heartbeat detection to manage the examination marking tasks of the examiner. After receiving the new question block, the question block allocation module allocates the examination marking tasks in real time according to the question dimension based on the teacher's priority. The priority queue designs a weighted priority function: priority = w1 * availability + w2 * rank - w3 * task_count, which comprehensively considers the differences in the schedules of different teachers, the different comprehensive abilities of teachers, and the number of question blocks assigned to teachers, where w1: the weight of the schedule availability; w2: the weight of the teacher's personal ability and title; w3: the weight of the number of tasks; availability: if the current time is within the teacher's available time period, availability = 1, otherwise availability = 0; rank: the teacher's personal ability and title level, ranging from 1 to 5; task_count: the number of tasks of the current teacher. In this way, the priority of each teacher can be reasonably calculated according to the teacher's schedule, personal ability and title, and the number of tasks, to ensure the fairness and efficiency of task allocation. The marking heartbeat detects the marking status of the teacher, sets the marking priority lowering threshold T1, and sends a text message to remind the marking teacher and the test administrator if the teacher does not make a new review within T1. Set the marking priority kick-out threshold T2, T2>T1. If the teacher does not make a new review within T2, to avoid the teacher's review affecting the overall marking progress, the teacher will be removed from the priority queue and the teacher and administrator will be notified. The administrator can then add the kicked-out teacher back to the priority queue when the teacher can mark the paper.
[0173] S04, comparing the standard answers set for the objective questions with the options for the objective questions, obtaining the score of the objective question picture based on the comparison result, and summing up the scores of all the objective question pictures and the subjective question pictures corresponding to the test paper picture to obtain the student's total score.
[0174] In this embodiment, the scoring module is divided into two modules, single-question real-time scoring and student total score scoring. Single-question real-time scoring receives objective questions and subjective questions marked by teachers, where the objective questions are identified by the recognition engine. The scoring module directly compares the answers set for the objective questions with the identified options to calculate the scores. The scores of subjective questions are based on the scores directly marked by the teachers. If the test questions are double-evaluated, the final score is the average score of the two teachers without exceeding the score difference threshold S1. If it exceeds S1, the test questions need to be assigned to the arbitration teacher for final scoring. The student total score is based on the single-question scoring, and it is determined whether all the test questions of the student to whom the test question belongs have been marked completely. If complete, the student marking status is updated and all single-question scores are summed to obtain the student's total score.
[0175] In summary, the client and server are mixed for recognition, making full use of the computing power of the local machine to prioritize part of the recognition work. When some test questions are across sheets, a small amount of server resources are used to combine and recognize the test questions across sheets, which supports the mixed scanning scenario of school test papers, improves resource utilization, and reduces service costs. If a student's entire test paper is incomplete, it will not affect the grading of the scanned test papers, realizing true instant scanning and grading. There is no need to wait for a certain number of copies to be scanned or the student's complete data to be scanned before grading and assigning the test papers, which improves the school's grading efficiency.
[0176] Embodiment 2
[0177] like Figure 2 As shown, in a second embodiment of the present invention, a test paper mixed scanning system is provided, the system comprising:
[0178] Forming module 10, for dynamically adjusting the number of test paper recognition threads based on the current hardware resource utilization rate by using a dynamic thread calculation formula, scanning the test paper based on the adjusted number of threads to form a test paper image;
[0179] A distinguishing module 20 is used to identify the test questions in each of the test paper images and divide them into question block images, determine whether the test questions in all the question block images are complete test questions, and if the test questions in the question block images are complete test questions, distinguish the question block images into objective question images and subjective question images;
[0180] The scoring module 30 is used to calculate the priority of each marking teacher based on the marking teacher's schedule, personal ability, title and number of tasks by using the priority function and marking heartbeat detection, and to assign the subjective question picture to the corresponding marking teacher in real time for scoring based on the marking teacher's priority and the question dimension to which the subjective question picture is to be assigned;
[0181] The summing module 40 is used to compare the standard answers set for the objective questions with the options of the objective questions, obtain the score of the objective question picture based on the comparison result, and sum the scores of all the objective question pictures and the subjective question pictures corresponding to the test paper picture to obtain the student's total score.
[0182] The test paper mixed scanning system provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference can be made to the corresponding contents in the aforementioned method embodiment.
[0183] Embodiment 3
[0184] like Figure 3 As shown, in the third embodiment of the present invention, the embodiment of the present invention provides the following technical solution, a computer, including a memory 202, a processor 201, and a computer program stored in the memory 202 and executable on the processor 201, and the processor 201 implements the test paper mixed scanning method as described above when executing the computer program.
[0185] Specifically, the processor 201 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
[0186] Among them, the memory 202 may include a large-capacity memory for data or instructions. For example, but not limitation, the memory 202 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 202 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 202 may be inside or outside the data processing device. In a specific embodiment, the memory 202 is a non-volatile memory. In a specific embodiment, the memory 202 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.
[0187] The memory 202 may be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 201 .
[0188] The processor 201 implements the above-mentioned test paper mixed scanning method by reading and executing the computer program instructions stored in the memory 202.
[0189] In some embodiments, the computer may further include a communication interface 203 and a bus 200. Figure 3 As shown, the processor 201, the memory 202, and the communication interface 203 are connected via a bus 200 and communicate with each other.
[0190] The communication interface 203 is used to implement communication between the modules, devices, units and / or equipment in the embodiment of the present application. The communication interface 203 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.
[0191] The bus 200 includes hardware, software or both, and couples the components of the computer to each other. The bus 200 includes but is not limited to at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example and not limitation, bus 200 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, bus 200 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.
[0192] Embodiment 4
[0193] In the fourth embodiment of the present invention, in combination with the above-mentioned test paper mixed scanning method, the embodiment of the present invention provides the following technical solution, a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned test paper mixed scanning method is implemented.
[0194] Those skilled in the art will appreciate that the logic and / or steps described in the flowchart as data or in other ways herein, for example, can be considered as a sequenced data table of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses.
[0195] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.
[0196] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0197] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0198] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A method for mixed scanning of test papers, characterized in that: The method comprises: The number of test paper recognition threads is dynamically adjusted based on the current hardware resource utilization rate using a dynamic thread calculation formula, and the test paper is scanned based on the adjusted number of threads to form a test paper image; Identify the test questions in each of the test paper images and divide them into question block images, determine whether the test questions in all the question block images are complete test questions, and if the test questions in the question block images are complete test questions, divide the question block images into objective question images and subjective question images; The priority function and the heartbeat detection of the examiner are used to calculate the priority of each examiner based on the examiner's schedule, personal ability, title and number of tasks. Based on the examiner's priority and the question dimension to which the subjective question picture is to be assigned, the subjective question picture is assigned to the corresponding examiner in real time for scoring; Comparing the standard answers set for the objective questions with the options of the objective questions, obtaining the score of the objective question picture based on the comparison result, and summing the scores of all the objective question pictures and the subjective question pictures corresponding to one of the test paper pictures to obtain the total score of the student; The dynamic thread calculation formula includes: In the formula, Optimal Threads represents the optimal number of threads, N represents the number of CPU cores, M represents the total amount of available memory, and T represents the memory consumption of each thread. Indicates the current CPU usage. Represents the current memory usage, Expressed as the maximum allowed CPU usage, Expressed as the maximum allowed memory usage; The priority function includes: priority = w1 * availability + w2 * rank - w3 * task_count Among them, w1 represents the weight of the timetable availability; w2 represents the weight of the examiner's personal ability and title; w3 represents the weight of the number of tasks; availability represents that if the current time is within the examiner's available time period, availability=1, otherwise availability=0; rank represents the examiner's personal ability and title rank, ranging from 1 to 5; task_count represents the number of tasks for the current examiner; The heartbeat detection of paper marking includes: Determine whether the marking teacher has not made new markings for more than the priority lowering threshold, if the marking teacher has not made new markings for more than the priority lowering threshold; It is determined whether the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking. If the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking, the marking teacher will be removed from the priority queue and the marking teacher and the administrator will be notified.
2. The test paper mixed scanning method according to claim 1, characterized in that: After the step of determining whether the test questions in all the question block images are complete test questions, the method further includes: If the test question in the question block image is not a complete test question, a portion of the test question across sheets of the question block image is cut out to form test question slices, and the test question slices are stored; Downloading the stored test question slices based on the local area network in an asynchronous manner, and storing the downloaded test question slices in the disk directory of the corresponding student; Determine whether all the test question slices corresponding to a test question are stored in the disk directory. If all the test question slices corresponding to a test question are stored in the disk directory; Then, all the test question slices are merged using an image conversion tool to form a merged test question; Determine whether the fusion test question is an objective question. If the fusion test question is an objective question, perform filling identification on the fusion test question to identify and obtain filling options for the fusion test question.
3. The test paper mixed scanning method according to claim 1, characterized in that: After the step of forming the test paper picture, the method further includes: Obtaining a first coordinate of a positioning block of a standard answer sheet, and calculating a search area based on the first coordinate and an expansion coefficient; Obtaining second coordinates of four corner points of the positioning block of the test paper image based on the search area using a contour detection algorithm; Calculate a first scaling factor based on the second coordinate and the first coordinate using a scaling calculation formula; Calculating the third coordinates of the black and white page number block of the test paper image based on the first scaling factor and the coordinates of the black and white page number block of the standard answer sheet; Obtaining blocks of black and white page number blocks of the test paper image based on the third coordinates, and obtaining page numbers of the test paper image based on the pixel ratio of each block using binary numbers and decimal numbers; Recognize the test number of all the scanned test paper images, and after the recognition is completed, determine the student and the number of the test paper images corresponding to the student; If the number of test papers does not correspond to the number of examinees, it is checked whether all test papers have been scanned. If all test papers have been scanned, the corresponding students will be marked as missing.
4. The test paper mixed scanning method according to claim 3, characterized in that: After the step of obtaining the first scaling factor, the method comprises: The second scaling factor is calculated based on the auxiliary positioning point of the test paper image using the scaling calculation formula; Calculating a difference value based on the second scaling factor and the first scaling factor using a relative difference formula; It is determined whether the difference value is greater than a threshold value. If the difference value is greater than the threshold value, the second scaling factor is used to calculate the page number.
5. The test paper mixed scanning method according to claim 4, characterized in that: The scaling calculation formula includes: In the formula, Expressed as the second scaling factor or the first scaling factor, , , and Respectively expressed as the horizontal and vertical coordinates of the two second coordinates, , and Respectively represented as the abscissa and ordinate of the two first coordinates; The relative difference formula includes: In the formula, represents the first scaling factor, The second scaling factor, Expressed as difference value.
6. A test paper mixed scanning system, characterized in that: The system comprises: A forming module is used to dynamically adjust the number of test paper recognition threads based on the current hardware resource utilization rate by using a dynamic thread calculation formula, and scan the test paper based on the adjusted number of threads to form a test paper image; A distinguishing module is used to identify the test questions in each of the test paper images and divide them into question block images, determine whether the test questions in all the question block images are complete test questions, and if the test questions in the question block images are complete test questions, distinguish the question block images into objective question images and subjective question images; The scoring module is used to calculate the priority of each marking teacher based on the marking teacher's class schedule, personal ability, title and number of tasks by using the priority function and marking heartbeat detection, and to assign the subjective question picture to the corresponding marking teacher in real time for scoring based on the marking teacher's priority and the question dimension to which the subjective question picture is to be assigned; A summing module, used for comparing the standard answer set based on the objective question with the options of the objective question, obtaining the score of the objective question picture based on the comparison result, and summing the scores of all the objective question pictures and the subjective question pictures corresponding to the test paper picture to obtain the total score of the student; The dynamic thread calculation formula includes: In the formula, Optimal Threads represents the optimal number of threads, N represents the number of CPU cores, M represents the total amount of available memory, and T represents the memory consumption of each thread. Indicates the current CPU usage. Represents the current memory usage, Expressed as the maximum allowed CPU usage, Expressed as the maximum allowed memory usage; The priority function includes: priority = w1 * availability + w2 * rank - w3 * task_count Among them, w1 represents the weight of the timetable availability; w2 represents the weight of the examiner's personal ability and title; w3 represents the weight of the number of tasks; availability represents that if the current time is within the examiner's available time period, availability=1, otherwise availability=0; rank represents the examiner's personal ability and title rank, ranging from 1 to 5; task_count represents the number of tasks for the current examiner; The heartbeat detection of paper marking includes: Determine whether the marking teacher has not made new markings for more than the priority lowering threshold, if the marking teacher has not made new markings for more than the priority lowering threshold; It is determined whether the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking. If the marking teacher has exceeded the priority kick-out threshold and has not conducted new marking, the marking teacher will be removed from the priority queue and the marking teacher and the administrator will be notified.
7. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the test paper mixed scanning method as described in any one of claims 1 to 5 is implemented.
8. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the test paper mixed scanning method as described in any one of claims 1 to 5 is implemented.
