An optimized method for leaving marks in the marking of examination papers, a storage medium and a device
By constructing a hash library of exam paper images, the accuracy and flexibility of marking and recording are achieved, solving the problems of misalignment and inflexibility in the marking and recording process of the marking machine, and supporting the unified printing of exam papers from multiple classes.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 读书郎教育科技有限公司
- Filing Date
- 2025-07-09
- Publication Date
- 2026-05-26
AI Technical Summary
Existing grading machines are prone to misalignment or errors during the grading and recording process, and cannot grade multiple classes' papers at once and then print out a unified record, making the operation inflexible.
A hash library for printed images is built, which matches the grading labels with hash values to support automatic recognition and printing of front and back images, achieving a unified operation for grading and record keeping.
It improves the accuracy and flexibility of marking and recording, supports unified printing of test papers from multiple classes, reduces operational requirements, and minimizes the problem of misaligned markings.
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent education technology, specifically to an optimized method, storage medium, and device for printing correction and feedback traces on exam papers. Background Technology
[0002] With the rapid development of information technology in education, and the significant assistance of large-scale models, intelligent grading of exams and assignments has become a reality. Exam / assignment grading traceability refers to the process where, after students complete their answers on paper exams / assignments, teachers scan the papers / assignments into the grading system, complete the grading on the system, and finally print the teacher's grading marks on the corresponding paper exam / assignment. Currently, there are grading machines on the market that integrate exam / assignment grading traceability; however, these grading machines have the following problems with grading trace printing:
[0003] 1. The two steps of scanning and grading homework and printing homework grading records must ensure that the same order, front and back, and orientation of the paper are consistent in order to leave accurate records; otherwise, misaligned or incorrect records will occur.
[0004] 2. Typically, a teacher teaches multiple classes. The current grading machine only supports scanning and grading the test papers / homework of one class and then immediately printing the results for that class. It cannot scan and grade the test papers / homework of multiple classes at once and then perform the printing operation uniformly, which is not flexible enough. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention aims to provide a method, storage medium, and device for optimizing roll-to-roll correction and marking printing.
[0006] To achieve the above objectives, the present invention adopts the following technical solution:
[0007] A method for optimizing the printing of correction and annotation traces on printed paper includes the following steps:
[0008] S1. Scan all the paper papers to be graded to obtain a set of paper paper images;
[0009] S2. Perform image preprocessing on each image in the set of printed images to obtain a first set of printed images; the image preprocessing includes uniformly scaling all printed images to a preset first size.
[0010] S3. Calculate the hash value corresponding to each paper image in the first paper image set, store the hash value corresponding to each paper image in the database and create an index to build a paper image hash library. In the paper image hash library, the corresponding paper image and its correction label can be located and accessed by searching through the hash value. Each paper paper includes a front image and a back image. When building the paper image hash library, the front image and back image of the same paper paper are further associated.
[0011] S4. The teacher corrects each paper image in the first set of paper images and generates correction marks left by the teacher on the paper images as correction labels for each paper image; the correction labels contain data of each correction mark and the relative position of each correction mark data on the paper image.
[0012] S5. Print the correction mark data from the correction label of each sheet image onto the corresponding side of the paper sheet:
[0013] S5.1 Place the paper roll to be printed into the paper feed slot of the printer;
[0014] S5.2. Use the printer to capture the image of the current side of each paper roll to be printed, and obtain the paper feed image set;
[0015] S5.3. Preprocess each paper image in the paper image set in the same way as step S2, and then calculate the hash value of the preprocessed paper image in the same way as step S3 to obtain the hash value of the paper image.
[0016] S5.4 For each paper image, calculate the Hamming distance between the hash value of the paper image and the hash values corresponding to each paper image in the hash database.
[0017] S5.5 For each paper input image, obtain the correction label of the paper image corresponding to the minimum Hamming distance, and print the correction mark data of each correction mark in the obtained correction label to the corresponding position on the front of the corresponding paper paper according to its relative position on the paper image. At the same time, obtain the correction label of another paper image associated with the paper image corresponding to the minimum Hamming distance, and print the correction mark data of each correction mark in the obtained correction label to the corresponding position on the other side of the corresponding paper paper according to its relative position on the other paper image.
[0018] Furthermore, in step S2, the image preprocessing also includes converting all roll images into preset level grayscale images.
[0019] Further, in step S3, the hash value of each image in the first set of images is calculated using the differential hashing algorithm, which serves as the fingerprint of each image.
[0020] Furthermore, in step S4, the marking trace data includes symbols and / or evaluation statements used by teachers to mark the test papers.
[0021] Furthermore, in step S5.5, before printing, the paper roll surface is first checked for inversion based on the paper feed image. If the paper roll surface is inverted, the data of each correction mark in the correction label is inverted before being printed on the current surface of the paper roll surface.
[0022] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.
[0023] The present invention also provides a computer device, including a processor and a memory, wherein the memory is used to store a computer program; and the processor is used to execute the computer program to implement the above-described method.
[0024] The beneficial effects of this invention are as follows:
[0025] 1. This invention constructs a hash library of test paper images. During printing, the hash value is used to match and print the correction labels of the test paper images. Teachers can correct all the test papers to be corrected at once, and then print them in a unified manner according to tasks (such as by class). This is not only more flexible, but also ensures the accurate correspondence between correction labels and test papers, and ensures the accuracy of the printing process.
[0026] 2. This invention supports the paper rolls for printing with traces to be placed upright, backwards, or upside down, reducing restrictions on paper feeding during printing, lowering the requirements for operators, and reducing the problem of misaligned traces. Detailed Implementation
[0027] The present invention will be further described below. It should be noted that this embodiment is based on the present technical solution and provides detailed implementation methods and specific operation processes, but the protection scope of the present invention is not limited to this embodiment.
[0028] This embodiment provides a method for optimizing the printing of correction and annotation traces on printed sheets, including the following steps:
[0029] S1. Scan all the paper papers to be graded to obtain a set of paper paper images;
[0030] S2. Perform image preprocessing on each image in the set of printed images to obtain the first set of printed images; the image preprocessing includes scaling all printed images to a preset first size (e.g., 8x8 pixels) to remove the influence of size differences.
[0031] In this embodiment, the image preprocessing further includes converting all roll images into preset level grayscale images.
[0032] S3. Calculate the hash value corresponding to each paper image in the first paper image set, store the hash value corresponding to each paper image in the database and create an index to build a paper image hash library. In the paper image hash library, the corresponding paper image and its correction label can be located and accessed by searching through the hash value.
[0033] S4. The teacher corrects each paper image in the first set of paper images and generates correction marks left by the teacher on the paper images as correction labels for each paper image; the correction labels contain the data of each correction mark and the relative position of each correction mark data on the paper image.
[0034] S5. Print the correction mark data from the correction label of each paper image onto the corresponding side of the paper paper.
[0035] In this embodiment, in step S3, the hash value of each image in the first set of images is calculated using the differential hash (dHash) algorithm, which serves as the fingerprint of each image.
[0036] The process of calculating the hash value using the differential hashing (dHash) algorithm is as follows: calculate the difference between each pixel in the image and its right-side neighboring pixel; then generate a 64-bit binary hash value for the image based on this difference, with the hash value ranging from (0, 1).
[0037] Furthermore, each paper roll includes both a front and a back image. When constructing the image hash library, the front and back images of the same paper roll are further associated. In this way, when marking and printing for verification, only one side (front or back) needs to be identified, avoiding the need to identify both sides of each paper roll before printing.
[0038] In this embodiment, in step S4, the correction trace data includes any one or more of the symbols and / or evaluation statements used by the teacher to correct the test paper, such as checkmarks, crosses, circles, diagonal lines, reasons for errors, and answers.
[0039] In this embodiment, the specific process of step S5 is as follows:
[0040] S5.1 Place the paper roll to be printed (i.e. the paper roll that has been scanned and corrected in step S1) into the paper feed slot of the printer.
[0041] S5.2. Use the printer to capture the image of the current side of each paper roll to be printed, thus obtaining the paper feed image set. The printer's paper feed port can only capture an image of one side of the paper roll, either the front or the back. However, since the front and back images are already associated, when the corresponding image of one side of the paper roll is retrieved using the paper feed image, the image of the other side of the paper roll can be located. Therefore, it is only necessary to capture the image of the current side of the paper roll to be printed as the paper feed image, without needing to capture the image of the other side.
[0042] S5.3. Preprocess each paper image in the paper image set in the same way as step S2, and then calculate the hash value of the preprocessed paper image in the same way as step S3 to obtain the hash value of the paper image.
[0043] S5.4 For each paper image, calculate the Hamming distance between the hash value of the paper image and the hash values corresponding to each paper image in the hash database.
[0044] S5.5 For each paper input image, obtain the correction label of the paper image corresponding to the minimum Hamming distance, and print the correction mark data of each correction mark in the obtained correction label to the corresponding position on the front of the corresponding paper paper according to its relative position on the paper image. At the same time, obtain the correction label of another paper image associated with the paper image corresponding to the minimum Hamming distance, and print the correction mark data of each correction mark in the obtained correction label to the corresponding position on the other side of the corresponding paper paper according to its relative position on the other paper image.
[0045] Furthermore, in step S5.5, before printing, the paper roll is first inspected for inversion based on the paper feed image. If the paper roll is inverted, the correction marks on the correction label are inverted before being printed onto the current surface of the paper roll. For example, if the paper roll is placed upside down, the correction marks on the correction label are rotated 180 degrees before being printed onto the current surface of the paper roll.
[0046] For those skilled in the art, various corresponding changes and modifications can be made based on the above technical solutions and concepts, and all such changes and modifications should be included within the protection scope of the claims of this invention.
Claims
1. A method for optimizing a print of a mark left by a correction on a surface, characterized in that, Includes the following steps: S1. Scan all the paper papers to be graded to obtain a set of paper paper images; S2. Perform image preprocessing on each image in the set of printed images to obtain a first set of printed images; the image preprocessing includes uniformly scaling all printed images to a preset first size. S3. Calculate the hash value corresponding to each paper image in the first paper image set, store the hash value corresponding to each paper image in the database and create an index to build a paper image hash library. In the paper image hash library, the corresponding paper image and its correction label can be located and accessed by searching through the hash value. Each paper paper includes a front image and a back image. When building the paper image hash library, the front image and back image of the same paper paper are further associated. S4. The teacher corrects each paper image in the first set of paper images and generates correction marks left by the teacher on the paper images as correction labels for each paper image; the correction labels contain data of each correction mark and the relative position of each correction mark data on the paper image. S5. Print the correction mark data from the correction label of each sheet image onto the corresponding side of the paper sheet: S5.1 Place the paper roll to be printed into the paper feed slot of the printer; S5.
2. Use the printer to capture the image of the current side of each paper roll to be printed, and obtain the paper feed image set; S5.
3. Preprocess each paper image in the paper image set in the same way as step S2, and then calculate the hash value of the preprocessed paper image in the same way as step S3 to obtain the hash value of the paper image. S5.4 For each paper image, calculate the Hamming distance between the hash value of the paper image and the hash values corresponding to each paper image in the hash database. S5.5 For each paper input image, obtain the correction label of the paper image corresponding to the minimum Hamming distance, and print the correction mark data of each correction mark in the obtained correction label to the corresponding position on the front of the corresponding paper paper according to its relative position on the paper image. At the same time, obtain the correction label of another paper image associated with the paper image corresponding to the minimum Hamming distance, and print the correction mark data of each correction mark in the obtained correction label to the corresponding position on the other side of the corresponding paper paper according to its relative position on the other paper image.
2. The method according to claim 1, characterized in that, In step S2, the image preprocessing also includes converting all roll images into preset level grayscale images.
3. The method according to claim 1, characterized in that, In step S3, the hash value of each image in the first set of images is calculated using the differential hashing algorithm, which serves as the fingerprint of each image.
4. The method according to claim 1, characterized in that, In step S4, the marking trace data includes symbols and / or evaluation statements used by teachers to mark the test papers.
5. The method according to claim 1, characterized in that, In step S5.5, before printing, the paper roll surface is first checked for inversion based on the paper feed image. If the paper roll surface is inverted, the data of each correction mark in the correction label is inverted before being printed on the current surface of the paper roll surface.
6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-5.
7. A computer device, characterized in that, It includes a processor and a memory, the memory being used to store a computer program; the processor being used to execute the computer program to implement the method according to any one of claims 1-5.