CN-121999531-A - Online examination behavior monitoring method and system based on privacy protection

CN121999531ACN 121999531 ACN121999531 ACN 121999531ACN-121999531-A

Abstract

The invention discloses a privacy protection-based online examination behavior monitoring method and system, which belong to the technical field of online education and comprise the following steps of S10, acquiring related data about a target examinee, wherein the related data comprise head images of the target examinee and images of objects related to the online examination of the target examinee, the images comprise images of a computer display screen, acquiring 4 corner points and a central axis of the computer display screen based on the images of the computer display screen, and generating two-dimensional images about the head images, the 4 corner points and the central axis of the target examinee. According to the invention, through combining real-time monitoring of the head image of the examinee and the computer display screen image, abnormal gestures and plagiarism of the examinee are accurately detected, and simultaneously multidimensional data analysis and scientific calculation are utilized, so that the accuracy and reliability of behavior identification are effectively improved, the privacy of the examinee is protected, and meanwhile, the fairness and seriousness of online examination are ensured.

Inventors

ZHU CHENFEI
REN YUNHUI
ZHAO LINLIN
GONG LONG
LI LIYANG

Assignees

扬州大自然网络信息有限公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (9)

1. The online examination behavior monitoring method based on privacy protection is characterized by comprising the following steps of: S10, acquiring relevant data about a target examinee, wherein the relevant data comprise a head image of the target examinee and an image of an object associated with an online examination of the target examinee, the image comprises an image of a computer display screen, 4 corner points and a central axis of the computer display screen are acquired based on the image of the computer display screen, and a two-dimensional image about the head image, the 4 corner points and the central axis of the target examinee is generated; s20, acquiring the height of the central axis, and carrying out real-time monitoring on a target examinee based on the two-dimensional image, wherein the real-time monitoring comprises real-time monitoring on vision and a screen, and in the real-time monitoring on vision, acquiring the height change of the head of the target examinee based on the central axis, wherein the acquisition steps are as follows: Dividing the middle end of the central axis into monitoring height intervals corresponding to the position 10cm upwards from the middle end of the central axis; calculating the height change of the head of the target examinee in the monitoring height interval by taking every 1-2 seconds as an acquisition period, and judging that the target examinee is in an abnormal posture state and giving an early warning if the height of the head of the target examinee moves towards the middle end of the central axis in 3 continuous acquisition periods; S30, when the target examinee is judged to be in an abnormal posture state, acquiring page display changes of a computer display screen in real-time monitoring of the screen, wherein the steps of acquiring are as follows: collecting page display changes of a computer display screen when the target examinee is in an abnormal posture state, wherein the page display changes comprise questions displayed by the computer display screen when the target examinee is in the abnormal posture state, and the questions comprise questions which are not solved; Acquiring a question type corresponding to the question, and collecting page display changes of a computer display screen when the head of a target examinee moves towards the middle end far away from the central axis, wherein the page display changes comprise whether the target examinee solves the question or not, and marking the question type as a reference question type; And S40, if the height of the head of the target examinee is moved towards the middle end far away from the central axis and the question is answered, judging that the target examinee has plagiarism and giving early warning, otherwise, not judging.
2. The method for monitoring on-line examination behaviors based on privacy protection as claimed in claim 1, wherein the height change of the head of the target examinee is calculated in the monitoring height interval by taking every 1-2 seconds as an acquisition period, and is calculated according to the following formula: ; in the formula, Represent the first The height variation of the head of the target examinee in each acquisition period and the last acquisition period; Indicated in the first The height of the head of the target examinee in each acquisition period; Indicated in the first -A height of the head of the target examinee for 1 acquisition period.
3. The method for online examination behavior monitoring based on privacy protection as set forth in claim 2, further comprising the steps of calculating according to the following formula: ; in the formula, Represent the first The height variation of the head of the target examinee in each acquisition period and the last acquisition period; Representing the number of consecutive acquisition periods; Representation of Average height rate of change for each acquisition period.
4. The method for monitoring on-line examination behaviors based on privacy protection according to any one of claims 2 to 3, wherein the rate of head movement of the 1 st acquisition period target examinee towards the middle end of the central axis is obtained in 3 continuous acquisition periods according to the calculation result, and is calculated according to the following formula: ; in the formula, Representing the rate at which the head of the 1 st acquisition period target examinee moves toward the middle end of the center axis; Representing the height of the head of the target examinee from the middle end of the central axis at the beginning of the 1 st acquisition period; The height of the head of the target examinee from the middle end of the central axis when the 1 st acquisition period is finished is represented; the duration of the 1 st acquisition period is indicated.
5. The method for monitoring on-line examination behaviors based on privacy protection as claimed in claim 4, wherein the calculated speed is marked as a reference speed, when a question corresponding to the reference question is acquired on a computer display screen at a future time, the height change of the head of a target examinee is acquired, if the height change is the same as the reference speed, the target examinee is judged to be in an abnormal posture state, and early warning is sent out, otherwise, the method is not judged.
6. The method for online examination behavior monitoring based on privacy protection according to claim 5, wherein 4 corner points are divided into a left corner point and a right corner point based on the central axis, when the target examinee is not judged to be in an abnormal posture state, the change of the head of the target examinee is obtained, the change comprises the steps of calculating the distance between the head of the target examinee and the left corner point and the right corner point, when the distance between the head of the target examinee and the left corner point or the right corner point is in a shortened trend, judging that the head of the target examinee is deviated towards the corner point on the corresponding side, judging that the target examinee is in the abnormal posture state, and giving an early warning, otherwise, judging not.
7. The method for monitoring on-line examination behaviors based on privacy protection as set forth in claim 6, wherein when it is determined that a head of a target examinee is biased toward a corner point on a corresponding side, relevant distance data are collected, the distance data are divided into class I data and class II data, the class I data are collected according to the reference rate, and the step of collecting is as follows: dividing the height of the head of the target examinee from the middle end of the central axis into at least 3 height points by taking every 0.5 second as a time node on the basis of the height of the head of the target examinee from the middle end of the central axis at the beginning of the 1 st acquisition period, and acquiring the distance between the adjacent height points; The step of II type data acquisition is as follows: when the distance between the head of the target examinee and the left corner or the right corner is in a shortening trend, acquiring the maximum movement amplitude of the head of the target examinee towards the left corner or the right corner in the process; When the height change of the head of the target examinee is acquired at a future time, an initial acquisition time point is given, the height change is acquired at the initial acquisition time point, and if the height change is the same as any one of the maximum movement amplitude and the distances between the 1 st height point and the 2 nd height point, the head of the target examinee is judged to be deviated to the corner point on the corresponding side or the head of the target examinee is judged to move to the middle end of the middle axis.
8. The method for online examination behavior monitoring based on privacy protection as claimed in claim 7, wherein when the height change of the head of the target examinee is obtained, if the head of the target examinee is deviated toward the corner point of the corresponding side but the maximum movement amplitude is not reached, calculating the distance that the head of the target examinee is deviated toward the corner point of the corresponding side, and if the distance exceeds two thirds of the maximum movement amplitude, determining that the head of the target examinee is deviated toward the corner point of the corresponding side.
9. A system for applying a privacy-preserving-based on-line examination behavior monitoring method as set forth in claim 1, comprising: The data acquisition module is used for acquiring related data about a target examinee, wherein the related data comprise head images of the target examinee and images of objects associated with the online examination of the target examinee, the images comprise images of a computer display screen, 4 corner points and a central axis of the computer display screen are acquired based on the images of the computer display screen, and two-dimensional images about the head images, the 4 corner points and the central axis of the target examinee are generated; the data processing module comprises an acquisition unit and a calculation unit; The acquisition unit is used for acquiring the height of the central axis and carrying out real-time monitoring on the target examinee based on the two-dimensional image, the real-time monitoring comprises real-time monitoring on vision and a screen, in the real-time monitoring on vision, the height change of the head of the target examinee is acquired based on the central axis, and the acquisition steps are as follows: Dividing the middle end of the central axis into monitoring height intervals corresponding to the position 10cm upwards from the middle end of the central axis; The calculating unit is used for calculating the height change of the head of the target examinee in the monitoring height interval by taking every 1-2 seconds as an acquisition period, and judging that the target examinee is in an abnormal posture state and giving an early warning if the height of the head of the target examinee moves towards the middle end of the central axis in 3 continuous acquisition periods; The display monitoring module is used for acquiring page display changes of the computer display screen in real-time monitoring of the screen when the target examinee is judged to be in an abnormal posture state, and the steps of acquiring are as follows: collecting page display changes of a computer display screen when the target examinee is in an abnormal posture state, wherein the page display changes comprise questions displayed by the computer display screen when the target examinee is in the abnormal posture state, and the questions comprise questions which are not solved; Acquiring a question type corresponding to the question, and collecting page display changes of a computer display screen when the head of a target examinee moves towards the middle end far away from the central axis, wherein the page display changes comprise whether the target examinee solves the question or not, and marking the question type as a reference question type; And the early warning module is used for solving the questions when the height of the head of the target examinee moves towards the middle end far away from the central axis, judging that the target examinee has plagiarism and giving out early warning, and otherwise, not judging.

Description

Online examination behavior monitoring method and system based on privacy protection Technical Field The invention relates to the technical field of online education, in particular to an online examination behavior monitoring method and system based on privacy protection. Background With the popularization of online education, the action behaviors of individual examinees can be accurately acquired through video streams, screen streams, audio streams and the like of online examination equipment, so that whether the examinees have cheating suspicions in examination rooms or not can be detected. For the research in this aspect, the application document with the application number of CN202311610864.0 provides an online examination anti-cheating monitoring method, device, equipment and storage medium, and the technical scheme comprises the steps of judging whether an examinee operation event belongs to a sensitive operation event when the examinee terminal receives the examinee operation event; judging whether the operation event of the examinee belongs to a compliance operation event according to the operation request information of the operation event of the examinee if the operation event of the examinee belongs to a sensitive operation event, responding to the operation event of the examinee if the operation event of the examinee belongs to the compliance operation event, and prohibiting responding to the operation event of the examinee if the operation event of the examinee does not belong to the compliance operation event. According to the technical scheme, the human-computer interaction process of the examinee and the examinee terminal is monitored in real time during the online examination, so that the high-incidence cheating means in the online examination scene are monitored comprehensively, and the safety and fairness of the examination are guaranteed. The other application number is CN202510206240.5, and the technical scheme comprises the steps of determining initial groups of suspected group cheating according to cheating behavior co-occurrence conditions in each period in an examination room, constructing an reachable matrix by taking each target examinee in the initial groups as a row and a column, setting 1a matrix element of any row and any column if the target examinee has visual reachability, determining all elements as empty rows or columns, respectively defining a neighborhood A in the examination room by taking the current target examinee as a center, giving different first weights to each examinee in the neighborhood A according to visual reachability, and giving a second weight to each examinee according to the nearest reachable distance between each examinee in the neighborhood A and other target examinees in the reachable matrix. However, the above technical solution collects data through full screen monitoring, whole-course recording of the camera, and other modes, which is easy to cause sensitive data leakage of examinee biological characteristics (face and voiceprint), and privacy leakage risks exist. Disclosure of Invention The present invention has been made keeping in mind the above problems occurring in the prior art for online education. Therefore, one of the purposes of the invention is to provide an online examination behavior monitoring method and system based on privacy protection, which accurately detects the abnormal posture and plagiarism of an examinee by combining the real-time monitoring of the head image of the examinee and the image of a computer display screen, and simultaneously utilizes multidimensional data analysis and scientific calculation to effectively improve the accuracy and reliability of behavior identification, and ensure the fairness and seriousness of the online examination while protecting the privacy of the examinee. In order to solve the technical problems, the invention provides the following technical scheme: on the one hand, the invention provides an online examination behavior monitoring method based on privacy protection, which comprises the following steps: S10, acquiring relevant data about a target examinee, wherein the relevant data comprise a head image of the target examinee and an image of an object associated with an online examination of the target examinee, the image comprises an image of a computer display screen, 4 corner points and a central axis of the computer display screen are acquired based on the image of the computer display screen, and a two-dimensional image about the head image, the 4 corner points and the central axis of the target examinee is generated; s20, acquiring the height of the central axis, and carrying out real-time monitoring on a target examinee based on the two-dimensional image, wherein the real-time monitoring comprises real-time monitoring on vision and a screen, and in the real-time monitoring on vision, acquiring the height change of the head of the target examinee based on the central axis, wherein t