Sorting method for self-proposition test papers of students in examination

By using machine vision technology to automatically identify and print test paper labels, the problem of low efficiency and easy errors in the sorting process of postgraduate self-designed examination papers has been solved, an efficient and accurate test paper sorting process has been achieved, and the level of examination management has been improved.

CN120733987APending Publication Date: 2025-10-03HUBEI UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510815423.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-18
Publication Date
2025-10-03

AI Technical Summary

Technical Problem

The process of sorting test papers for postgraduate self-designed examinations is inefficient and prone to errors. In particular, the sorting work before and after the examination is complicated and requires high professionalism from the personnel, which affects the fairness and efficiency of the examination.

Method used

Machine vision technology is used to take photos of the labels on the test paper packaging, extract text information, compare it with the data of the postgraduate examination authority, print complete information labels and re-affix them to achieve automated sorting.

Benefits of technology

It improves the efficiency and accuracy of test paper sorting, reduces the tedious steps of manual operation, reduces the error rate, and ensures the fairness of the examination and management efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120733987A_ABST
    Figure CN120733987A_ABST
Patent Text Reader

Abstract

The invention provides a sorting method for self-proposition test papers of student enrollment examinations, and belongs to the technical field of artificial intelligence. Comprising the following steps: A) photographing labels on respective proposition test paper packaging objects through a machine vision technology, and extracting character information on the respective proposition test paper packaging objects; b) comparing the extracted character information on the self-proposition test paper with the data of the examination administrative department of the postgraduate; c) after the comparison is correct, printing complete information labels of respective proposition test papers on the basis of data provided by the examination administrative department of the graduate students; d) the complete information label is pasted back to the corresponding self-proposition test paper packaging object; and E) sorting the self-proposition test papers before examination and after examination under different requirements by scanning complete information labels. The method has the advantages of high efficiency and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of artificial intelligence and relates to a method for sorting self-written test papers for postgraduate entrance examinations. Background Art

[0002] There are 2,132 graduate admissions institutions nationwide, each independently producing 29,000 different self-designed exams, with up to five sessions. A specific testing center receives different self-designed exams from different institutions for different subjects. Smaller centers typically handle over 3,000 self-designed exams, while larger centers can handle tens of thousands. These self-designed exams undergo two sorting processes at each testing center. The first is after the venues are finalized. The center sorts all received self-designed exams and places them in bags corresponding to the candidate's session and venue. After the exams are concluded, the center organizes and sorts all the returned self-designed exams and mails them back to each institution. Due to confidential mailing requirements, some institutions with larger volumes of exams need to split the returned papers into several thousand separate parts. Since the postgraduate entrance examination is an important part of the country's selection of high-level talents, it is related to the vital interests of the majority of candidates, educational equity, and national economic and social development. It must be ensured to be "absolutely safe". For this reason, conventional manual sorting has strict requirements on the work attitude and business familiarity of the participants.

[0003] Generally, the conventional sorting process of each test center is as follows: (1) The test center receives the self-designed test papers from each enrollment unit, and counts and verifies the papers received from each enrollment unit with the candidate and subject information in the registration information; (2) All candidates of this test center are arranged into examination rooms according to the rules; (3) For all test papers of each enrollment unit, labels are made in advance according to the subject, session, examination room number, and seat number information. The label information includes the candidate's name, test number, ID card, registration subject code, registration subject name, registration unit code, registration unit name, session, examination room number, seat number, and return confidential paper number (how to arrange the paper number?); (4) For each test paper, according to the candidate and subject information, find the corresponding label from thousands to tens of thousands of labels and paste it; (5) Put the corresponding test paper bag according to the session and examination room number on the label; (6) After the test, after the test papers are collected, they are divided and packaged according to the confidential number; (7) The confidential paper is returned to each enrollment unit.

[0004] The pre-exam sorting and post-exam confidential return of self-designed exam papers is a specialized step in the graduate entrance examination organization process, involving many personnel, carrying out the heaviest workload, and being prone to errors, with the most severe consequences. This purely manual process is complex, inefficient, and prone to errors. Furthermore, the high-intensity, long-term, and stressful nature of the sorting process creates significant challenges. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for sorting self-designed examination papers for postgraduate entrance examinations in response to the above-mentioned problems existing in the existing technology. The technical problem to be solved by the present invention is how to improve the sorting efficiency and sorting accuracy of self-designed examination papers for postgraduate entrance examinations.

[0006] The objectives of the present invention can be achieved through the following technical solutions: A method for sorting self-designed test papers for postgraduate entrance examinations, characterized in that it includes the following steps: A) taking photos of the labels on the packages of the respective test papers through machine vision technology, and extracting the text information on the packages of the respective test papers; B) comparing the extracted text information on the self-designed test papers with the data of the administrative department in charge of the postgraduate examination; C) after the comparison is correct, printing out the complete information labels of the respective test papers based on the data provided by the administrative department in charge of the postgraduate examination; D) pasting the complete information labels back to the corresponding self-designed test paper packages; E) completing the pre-exam sorting and post-exam sorting of the self-designed test papers according to different needs by scanning the complete information labels.

[0007] Furthermore, the text information on the packaging of the self-designed test paper includes the candidate's name, candidate's ID card, registration subject code, registration subject name, registration unit code, and registration unit name; the complete information label includes the candidate's name, test number, candidate's ID card, registration subject code, registration subject name, registration unit code, registration unit name, session number, examination room number, seat number, and return confidential paper number.

[0008] Furthermore, in step A), when photographing the package of the self-designed test paper, fill-in light processing is performed as needed.

[0009] Furthermore, in step B), when the text information extracted from the self-designed test paper does not correspond to the data of the administrative department in charge of the postgraduate examination, it is checked and corrected with the admissions unit and the administrative department in charge of the postgraduate examination, and the corrected information is used as the printing information of the complete information label.

[0010] Furthermore, the complete information tag is a QR code or text information or a combination of the QR code and text information.

[0011] Furthermore, in step D), the complete information labels are affixed back to the locations of the respective test paper packages for standardization to facilitate manual classification, inspection, return of the envelopes, and online scanning and collection of information.

[0012] Furthermore, the pre-examination sorting of self-designed examination papers includes the pre-examination examination paper packaging and sorting according to the session information and examination room number information, and the sequential sorting, packaging and packaging according to the seat number; the post-examination sorting of self-designed examination papers includes the sorting and packaging of returned examination papers according to the examination unit code and examination unit name, the sorting and packaging of all examination papers of the same examination unit according to the return confidential paper number, and the sorting of examination papers before and after the grading of the corresponding examination unit.

[0013] Furthermore, image information formed by photographing the labels on the packaging of the respective test papers is stored.

[0014] Furthermore, the self-designed test papers are sorted before and after the test by manual sorting, robot sorting or online automatic sorting.

[0015] Using AI-based machine vision technology, the system uses real-time video images to identify the envelopes of self-designed test papers. It then automatically prints labels containing the session number, exam room number, confidential paper number, and candidate and subject information, allowing for seamless sorting. Sorting and organizing test papers requires no experience or collaboration from staff, and can be reduced from multiple rounds of inspections by six or seven people to just two or three. There's no need to print labels in advance, and there's no need to hold the test paper to find the corresponding label. It only takes 3-5 seconds for one person to sort each test paper, increasing sorting efficiency by at least dozens of times. According to actual trials at this test center (Hubei University of Science and Technology) in 2024, the sorting accuracy rate was 100% under standardized operation and normal equipment operation.

[0016] The examination papers for postgraduate examinations are divided into self-written examination papers and unified examination papers. Self-written examination papers are independently written by the admissions unit, and the candidates who refer to the self-written examination papers are distributed in different examination centers across the country. Therefore, different candidates in the same examination center and the same examination room refer to different examination papers of different admissions units. For the corresponding examination centers, all the examination papers received need to be sorted before the examination to accurately match each examination paper with the corresponding candidate information, examination room information, session information, etc. The admissions unit cannot obtain the examination room information, session information, etc. When sending out the examination papers, the examination paper packaging of the admissions unit only contains part of the required information, namely the candidate's name, candidate's ID card, registration subject code, registration subject name, registration unit code, etc. , the name of the applying unit, and when the test center is arranged before the test, it is also necessary to sort the test papers before the test according to the session number, examination room number, seat number, confidential paper number and other information arranged by the National Examination Institute. Therefore, the above scheme is set based on the complicated processing of self-designed test papers for postgraduate students. It has very practical and important value for the test center staff, enrollment units, and clear responsibilities. It improves the efficiency and accuracy of pre-test sorting by test center staff, the efficiency and accuracy of post-test sorting by enrollment units, and the clarity of responsibilities of various connecting units in the event of omissions. At the same time, it also greatly improves management efficiency, reduces the work cost of self-designed examinations for postgraduate students, effectively controls the risks of such large-scale examinations, and is conducive to the fairness and justice of the examinations. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the sorting process of the existing self-designed test papers for the postgraduate entrance examination.

[0018] Figure 2 This is a schematic diagram of the sorting process of self-designed test papers for the postgraduate entrance examination in this plan.

[0019] Figure 3 This is the logic diagram for sorting self-designed test papers for the postgraduate entrance examination in this scheme.

[0020] Figure 4 It is a schematic diagram of the structure of the equipment that completes photography and label printing in this solution.

[0021] Figure 5 It is a software and hardware architecture diagram.

[0022] Figure 6 It is a text information style diagram for printing labels.

[0023] Figure 7 This is the interface diagram of the display screen in the initial startup state. DETAILED DESCRIPTION

[0024] The following are specific embodiments of the present invention and the accompanying drawings to further describe the technical solutions of the present invention, but the present invention is not limited to these embodiments.

[0025] like Figure 1 As shown, at present, the test center receives the self-designed test paper machine components from each enrollment unit, and counts and verifies the candidate and subject information in the test papers of each enrollment unit and the registration information; each test paper package has the candidate information, registration subject information, registration unit information, etc., but no examination room information, and the test center staff needs to arrange all the candidates in the test center into the examination room according to the rules; therefore, the test center staff needs to make labels in advance for all the test papers of each enrollment unit according to the examination arrangement provided by the Examination Institute, according to the subject, session, examination room number, and seat number information. The label information includes the candidate's name, test number, ID card, examination subject code, examination subject name, examination unit code, examination unit name, session, examination room number, seat number, confidential paper number for return; the prepared labels need to correspond to the corresponding examination papers, and according to the candidate and subject information, the corresponding labels are found from thousands to tens of thousands of labels and pasted; then, according to the session and examination room number of the labels, the corresponding examination paper bags are placed in them for distribution during the examination; after the examination, after the examination papers are collected, all the examination papers of the same enrollment unit are divided into multiple packages according to the confidential number, paper packaging requirements, and the packaging capacity of a single package, and sent back confidentially to each enrollment unit.

[0026] In current practice, labels are printed centrally, and the correspondence between labels and test papers needs to be completed manually. In other words, it is necessary to find the label corresponding to the corresponding test paper among a large number of labels, or find the test paper corresponding to the corresponding label among a large number of test papers. This is tedious and error-prone. If the accuracy of the information on the test paper packaging needs to be verified, then there will be additional work of verifying the information one by one. Human errors and errors in the original label information on the test paper are almost impossible to detect. Therefore, the rigor and professionalism of the entire process are required to be high, the operators need professional training, and the determination of responsibility after errors is difficult to implement.

[0027] like Figure 2 and Figure 3 As shown, in this solution, technologies such as artificial intelligence and machine vision are used, and the process is simplified as follows: the test center receives self-designed test paper components from each enrollment unit. Machine vision takes a photo, identifies and extracts the text information on the test paper packaging, and compares the information with the enrollment test information provided by the Examination Institute. If the comparison is correct, a new label corresponding to the test paper is immediately printed. The new label contains all the information required before and after the test, such as the test center, examination room, test number, seat number, and return confidential paper number. The self-designed test paper information is checked and the new labels are affixed one by one. Before the test, the test paper can be sorted according to the information on the new label for the examination room layout, corresponding to the seat number and other information, so that the test papers can be distributed and collected during the test. After the test, the test papers can be packaged and returned according to the return confidential paper number and enrollment unit information. Generally speaking, the same enrollment unit has multiple test papers at a test center, which need to be packaged in multiple packages. This is the purpose of the return confidential paper number.

[0028] It is not difficult to see that this solution only requires taking photos of the test papers sent to the test site by the admissions unit one by one. After taking the photo, a new label will be generated and pasted back to the packaging of the test paper, which greatly reduces the complicated and error-prone processes of "finding labels" and "finding test papers" caused by traditional methods.

[0029] The label printed instantly by the machine can be a text information label, a QR code label, or a label in other forms or a combination of multiple forms of labels.

[0030] After the new label is affixed to the test paper envelope, in the subsequent sorting process, the test paper is in the envelope, which carries the basic information of the examinee: test number, name, registered unit, etc. The teacher needs to place the envelope under the camera for photo recognition. After recognition, a new label is generated, which carries the examinee's seat number, confidential number and other information. Figure 6 The labels are affixed to the envelopes and then sorted manually.

[0031] like Figure 4 As shown, a postgraduate examination self-designed test paper sorting auxiliary system is used to complete machine vision photography, information extraction and label printing. It includes a central processing module, a camera is connected to the central processing unit, and the image formed after the camera takes the picture is transmitted to the image processing module through the central processing module. After the image processing module corrects and processes the clarity, the OCR optical character recognition module extracts the text character information in the image. At the same time, the data import unit imports the candidate examination room information provided by the examination institute into the central processing module. The central processing module compares the text character information extracted from the image with the database imported by the data import unit. Since the information extracted from the image is not a single information, it can be compared after comparison. To confirm whether there is information corresponding to the candidate that matches the image information in the database. At this time, the original label information on the test paper of the default image is correct, and the printing module prints a new label covering all the information. If the corresponding candidate information is not found after comparison, then either the image extraction is incorrect, or the original label information of the corresponding test paper is incorrect, or the information provided by the examination institute is incorrect, or there are omissions in the database provided by the examination institute. Through communication, verification and correction are achieved. A very small number of "abnormal" test papers are handled separately, and other test papers can complete the pasting of new labels at one time according to the above process; for possible artificial intelligence OCR recognition errors, manual entry of the test number will be required to print the label.

[0032] In order to better complete the above work, a display screen is set up, abnormal information can be displayed on the display screen, and a mouse and keyboard control module is set up. When an external mouse and keyboard are connected, the virtual buttons on the display screen can be controlled by the mouse and keyboard, that is, the software can be controlled. In addition, a light intensity detection module is provided to detect the shooting environment. If the shooting environment is not conducive to clear shooting, the fill light is used to fill in the light to complete the shooting.

[0033] The image information is stored in a storage unit for subsequent verification.

[0034] If we separate the various parts of this plan, its technical means are all relatively mature ones at present, such as photography, OCR image recognition, the work performed by the central processing unit, etc. However, at present, there is no artificial intelligence solution that is suitable for the actual situation of postgraduate self-designed examinations. Therefore, this plan mainly proposes a method that is significantly better than the existing practices in terms of application scenarios and application environment, and has important economic and social value in practical applications.

[0035] like Figure 5 As shown, the system architecture is roughly divided into three parts. The first part is the camera, which is used to collect pictures; the second part is the software part, which is used to identify, analyze, extract and process pictures; the third part is the printing part, which uses a marking machine to print out the results.

[0036] like Figure 7 As shown in the figure, this is the initial screen interface after powering on the computer. A live preview of the camera is displayed at the top of the screen. With an external mouse and keyboard connected, the software can operate the system using both the mouse and keyboard. When the user places the test paper file under the camera, they can click the [Recognize] button. The software will display the recognition results in the [Recognition Results] text field on the right. After confirming the data is correct, they can click the [Print] button to print the results. Alternatively, they can click the [Auto Recognize] button. The software will automatically recognize the text placed under the camera, and any matching results will be displayed in the [Recognition Results] text field. If the user selects the [Auto Print After Recognition] checkbox in the lower right corner, the results will be automatically printed after successful recognition. If recognition fails, the user can enter the corresponding test number in the [Test Number] text field in the upper right corner and select the corresponding session in the [Session]. The input will automatically complete and the recognition results will be displayed below. Click the [Print] button to print the recognition results. Please note that manual entry requires stopping [Auto Recognition] first.

[0037] The procurement of major components is as follows:

[0038] Device Name Manufacturer model CPU Nvidia JetsonTX2 Switching Power Supply Sunbang 24V6A Display Punctual Atom RGBLCD Fill light camera WN-L2004K325L-NB1080P

[0039] The specific embodiments described herein are merely illustrative of the spirit of the present invention. Persons skilled in the art may make various modifications, additions, or substitutions to the described specific embodiments without departing from the spirit of the present invention or exceeding the scope of the appended claims.

Claims

1. A method for sorting self-written examination papers for postgraduate entrance examinations, characterized in that: The method includes the following steps: A) taking photos of the labels on the packaging of each test paper through machine vision technology, and extracting the text information on the packaging of each test paper; B) comparing the extracted text information on the self-written test paper with the data of the administrative department in charge of the postgraduate examination; C) after the comparison is correct, printing out the complete information label of each test paper based on the data provided by the administrative department in charge of the postgraduate examination; D) pasting the complete information label back to the corresponding self-written test paper packaging; E) completing the pre-examination and post-examination sorting of the self-written test papers according to different needs by scanning the complete information label.

2. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: The text information on the packaging of the self-designed test paper includes the candidate's name, candidate's ID card, registration subject code, registration subject name, registration unit code, and registration unit name; the complete information label includes the candidate's name, test number, candidate's ID card, registration subject code, registration subject name, registration unit code, registration unit name, session number, examination room number, seat number, and return confidential paper number.

3. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: In step A), when taking a photo of the package of the self-designed test paper, fill-in light processing is performed as needed.

4. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: In step B), when the text information extracted from the self-designed test paper does not correspond to the data of the administrative department in charge of the postgraduate examination, it is checked and corrected with the admissions unit and the administrative department in charge of the postgraduate examination, and the corrected information is used as the printing information of the complete information label.

5. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: The complete information label is a QR code or text information or a combination of the QR code and text information.

6. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: In step D), the complete information labels are affixed back to the locations of the respective test paper packages for standardization to facilitate manual classification, inspection, return of the envelopes, and online scanning and collection of information.

7. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: The pre-examination sorting of self-designed examination papers includes the packaging and sorting of examination papers according to the session information and examination room number information, and the sorting, packaging and packaging in sequence according to the seat number; the post-examination sorting of self-designed examination papers includes the sorting and packaging of returned examination papers according to the examination unit code and examination unit name, the sorting and packaging of all examination papers of the same examination unit according to the confidential paper number for return, and the sorting of examination papers before and after the grading of the corresponding examination unit.

8. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: Store the image information formed by taking pictures of the labels on the packaging of the respective test papers.

9. A method for sorting self-designed examination papers for postgraduate entrance examination according to claim 1, characterized in that: The methods of sorting self-designed test papers before and after the exam are manual sorting, robot sorting or online automatic sorting.