Anti-cheating method and system for online examination

Abstract

The invention relates to the technical field of online education, and provides an anti-cheating method and system for an online examination, and the method comprises the steps: carrying out the confirmation of the identity and registration information of an examinee through face recognition before the examination; questions and setting options are randomly extracted according to a video of an online examination, wherein the video cannot be dragged and fast forward; behavior characteristics of the examinee in the examination process are monitored, and abnormal behaviors are identified by analyzing the stability of the behavior characteristics; cheating behaviors are identified according to behavior characteristics of examinees in the examination process, and an alarm is given in time; and storing the registration information and the behavior characteristics of the examinees in the examination process. Identity verification is carried out on examinees through face recognition, it is ensured that reference personnel are consistent with registration information, and the behavior of replacing examinees is prevented. Behavior characteristics of examinees are monitored in real time, abnormal behaviors of the examinees are recognized, and cheating is effectively prevented; and when an abnormal behavior is detected, an alarm is immediately given to invigilators, so that the fairness and integrity of an examination are ensured.

Description

Technical Field

[0001] The present invention relates to the field of online education technology, and more specifically, to a method and system for preventing cheating in online examinations. Background Art

[0002] With the popularity of online education and remote work, traditional anti-cheating measures such as invigilator supervision and examination rule constraints are difficult to implement in an online environment. With the rapid development of Internet technology, online exams have become more and more popular due to their convenience and accessibility. However, this examination method also brings new challenges, especially in ensuring the fairness and integrity of the examination. Traditional offline exams can prevent cheating through direct supervision by invigilators, but the virtuality and decentralization of the online examination environment make this method no longer applicable. Summary of the invention

[0003] The present invention aims to provide a method and system for preventing cheating in online examinations, which can authenticate the identity of examinees through face recognition, ensure that the examinees are consistent with the registration information, and prevent the occurrence of substitute examinations; monitor the behavioral characteristics of examinees in real time, identify the abnormal behavior of examinees, and thus effectively prevent cheating; once abnormal behavior is detected, an alarm is immediately sent to the invigilator, thereby ensuring the fairness and integrity of the examination.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a method for preventing cheating in online examinations, comprising the following steps: Use facial recognition to confirm the candidate's identity and registration information before the exam begins; Randomly select questions and set options based on the online exam video, in which the video cannot be dragged or fast-forwarded; Monitor the behavioral characteristics of candidates during the exam and identify abnormal behaviors by analyzing the stability of behavioral characteristics; Identify cheating behaviors based on the behavioral characteristics of candidates during the exam and issue an alert in a timely manner; The registration information and the behavioral characteristics of the examinees during the examination are stored for easy post-examination verification.

[0005] Preferably, the confirmation of the candidate's identity and registration information by face recognition before the start of the exam includes: Through live body collection, the examinee's face is dynamically recognized to generate several face images; By building a CNN model to extract the feature vectors of face images, the candidate's identity is determined by calculating the Euclidean distance between the feature vectors; The calculation results are fed back to the invigilator in real time. If the verification is passed, the candidate is allowed to enter the examination room; if the verification fails, the candidate is prohibited from taking the exam.

[0006] Preferably, the dynamic recognition of the examinee's face through live body collection is specifically: Through the three-dimensional imaging principle of 3D structured light, the three-dimensional features of the face are compared to determine whether the examinee is a real living person; Determine the location of the eye area and use a computer vision processor to analyze the open and closed state of the eyes; Determine the location of the mouth area and use a computer vision processor to analyze the opening and closing state of the mouth; Identify shaking and nodding movements by detecting the distance changes of facial key points.

[0007] Preferably, the method of determining the examinee's identity by calculating the Euclidean distance between the feature vectors is specifically as follows: The Euclidean distance between feature vectors is calculated using the Euclidean distance algorithm; The Euclidean distance is compared with a predefined threshold. When the Euclidean distance is greater than the predefined threshold, it is determined that the feature vectors do not match, that is, the candidate's identity is inconsistent with the registration information; When the Euclidean distance is less than a predefined threshold, the feature vectors are considered to match, that is, the candidate's identity is consistent with the registration information.

[0008] Preferably, the randomly selecting questions and setting options based on the online test video includes: A pseudo-random number generator is used to pause the video at the time of answering questions, and randomly draw questions and set options. The video will continue to play after the questions are answered and submitted, or if the questions are not answered within the time limit.

[0009] Preferably, the video cannot be dragged and fast-forwarded specifically because: The video content is encrypted, and the video progress bar cannot be dragged or fast-forwarded without correct decryption.

[0010] Preferably, monitoring the behavioral characteristics of examinees during the examination and identifying abnormal behaviors by analyzing the stability of the behavioral characteristics includes: Real-time recording of behavioral characteristics during the exam, including mouse movement trajectory and keyboard tapping frequency; The stability data of examinees’ behavioral characteristics were obtained through variance analysis, including the average speed and acceleration of mouse movement and the average and peak frequency of keyboard tapping; Compare the stability data of the examinee's behavioral characteristics with the normal behavior data to identify abnormal behavior.

[0011] Preferably, the identifying cheating behavior according to the behavioral characteristics of the examinee during the examination and issuing an alarm in time includes: Cluster analysis is used to cluster the behavioral characteristics of examinees and identify outliers, i.e., possible cheating behaviors, including switching screens and opening search engines; Calculate the distance between the examinee's behavior characteristics and the cluster center to determine whether it is abnormal behavior. If the distance exceeds the preset threshold, it is considered abnormal. Once abnormal behavior is determined, an alarm will be immediately sent to the invigilator so that the invigilator can take timely measures.

[0012] Preferably, the storing of registration information and behavioral characteristics of examinees during the examination process for post-examination verification includes: Store candidate information and behavior data; Provide a data query interface to facilitate subsequent data retrieval and viewing; Back up candidate information and behavior data and upload them to the cloud.

[0013] Preferably, a system for preventing cheating in online examinations comprises: Identity verification module, used to confirm the candidate's identity and registration information before the exam begins through facial recognition; The random question module is used to randomly select questions and set options based on the online test video, where the video cannot be dragged or fast-forwarded; Behavior analysis module, which is used to monitor the behavioral characteristics of candidates during the examination and identify abnormal behaviors by analyzing the stability of behavioral characteristics; The cheating detection module is used to identify cheating behaviors based on the behavioral characteristics of candidates during the exam and issue an alarm in a timely manner; The data storage module is used to store registration information and the behavioral characteristics of candidates during the examination process for easy post-examination verification.

[0014] The beneficial effects of the present invention are: The present application provides a method and system for preventing cheating in online examinations, which authenticates the identity of examinees through face recognition to ensure that the examinees are consistent with the registration information and prevent the occurrence of substitute examinations; monitors the behavioral characteristics of examinees in real time and identifies abnormal behaviors of examinees, thereby effectively preventing cheating; once abnormal behavior is detected, an alarm is immediately sent to the invigilator, ensuring the fairness and integrity of the examination. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a method for preventing cheating in online examinations according to the present invention; Figure 2 The present invention is a module schematic diagram of a system for preventing cheating in online examinations.

[0016] The drawings are only used for illustrative purposes and are not to be construed as limitations of the present invention. To better illustrate the embodiments, some parts of the drawings may be omitted, enlarged, or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions may be omitted in the drawings. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, a method for preventing cheating in online examinations of the present invention comprises the following steps: S1: Confirm the candidate’s identity and registration information before the exam through facial recognition; S2: Randomly select questions and set options based on the online test video, in which the video cannot be dragged or fast-forwarded; S3: Monitor the behavioral characteristics of candidates during the exam and identify abnormal behaviors by analyzing the stability of behavioral characteristics; S4: Identify cheating behaviors based on the behavioral characteristics of candidates during the exam and issue an alarm in time; S5: Store the registration information and the behavioral characteristics of the examinees during the examination to facilitate post-examination verification.

[0019] In the above solution, a comprehensive online examination monitoring method is constructed by comprehensively using facial recognition, video surveillance, behavioral analysis and data storage.

[0020] Before the exam begins, the identity of the candidate is verified through high-precision facial recognition and matched with the registration information to ensure the authenticity of the candidate's identity. At the same time, liveness detection is combined to prevent photo, video or other forms of fraud, effectively prevent the phenomenon of proxy examinations, and ensure that every candidate takes the exam in person.

[0021] During the exam, candidates are prevented from manipulating the video stream by dragging or fast-forwarding, thus ensuring the real-time and fairness of the exam. At the same time, questions and options can be randomly selected, and the order of questions and options for each candidate is unique, which increases the randomness of the exam and reduces the possibility of cheating.

[0022] By monitoring the behavioral characteristics of candidates during the exam, such as mouse movement trajectory and keyboard typing frequency, the stability of behavior can be analyzed.

[0023] Once abnormal behavior of candidates is identified, such as switching screens and opening search engines, an alarm will be issued in time to notify the invigilator for further investigation. This real-time monitoring and response mechanism can effectively prevent cheating and ensure the fairness of the examination.

[0024] All candidates' registration information and behavioral characteristics during the examination process will be stored in the system. This data is not only used for monitoring during the examination, but also for post-examination inspection and analysis to facilitate review of the examination process and ensure the accuracy and reliability of the examination results.

[0025] Preferably, in step S1, confirming the candidate's identity and registration information by face recognition before the start of the test includes: Through live body collection, the examinee's face is dynamically recognized to generate several face images; By building a CNN model to extract the feature vectors of face images, the candidate's identity is determined by calculating the Euclidean distance between the feature vectors; The calculation results are fed back to the invigilator in real time. If the verification is passed, the candidate is allowed to enter the examination room; if the verification fails, the candidate is prohibited from taking the exam.

[0026] In the above scheme, dynamic face recognition has a high recognition rate and can accurately identify personal identity even under different lighting, angles and expressions; the face recognition process is fast and can complete identity verification in a short time, which is suitable for application in examination environments that require real-time monitoring.

[0027] Preferably, the dynamic recognition of the examinee's face through live body collection is specifically: Through the three-dimensional imaging principle of 3D structured light, the three-dimensional features of the face are compared to determine whether the examinee is a real living person; Determine the location of the eye area and use a computer vision processor to analyze the open and closed state of the eyes; Determine the location of the mouth area and use a computer vision processor to analyze the opening and closing state of the mouth; Identify shaking and nodding movements by detecting the distance changes of facial key points.

[0028] In the above scheme, the verification examinee is prevented from cheating by using non-living materials such as photos, videos or 3D models. Liveness detection can be performed in a variety of ways, including silent video liveness detection, digital video liveness detection, action liveness detection and online picture liveness detection. These detection methods can identify PS, face fusion, video face swapping and other means; extract key features in the face image through face recognition technology, such as the size, position, distance and other attributes of the facial features such as the iris, nose wing, and mouth corner, and calculate their geometric feature quantities to form a feature vector describing the face image; compare the feature vector of the collected face image with the faceprint code stored in the database to confirm the identity of the examinee; this process usually requires a judgment time of less than 1 second to ensure the smoothness of the examination process.

[0029] Preferably, the method of determining the examinee's identity by calculating the Euclidean distance between the feature vectors is specifically as follows: The Euclidean distance between feature vectors is calculated using the Euclidean distance algorithm; The Euclidean distance is compared with a predefined threshold. When the Euclidean distance is greater than the predefined threshold, it is determined that the feature vectors do not match, that is, the candidate's identity is inconsistent with the registration information; When the Euclidean distance is less than a predefined threshold, the feature vectors are considered to match, that is, the candidate's identity is consistent with the registration information.

[0030] In the above scheme, in the field of identity authentication and information matching, the Euclidean distance algorithm is a commonly used method to measure the similarity between feature vectors; extract feature vectors from the candidate's registration information, including facial features, fingerprints or other biometric data. At the same time, extract the corresponding feature vectors from the candidate information collected in real time; compare the calculated Euclidean distance with a predefined threshold, which is determined based on historical data and statistical analysis and represents the maximum acceptable difference between feature vectors; if the calculated Euclidean distance is greater than the predefined threshold, the system will determine that the feature vectors do not match, that is, the candidate's identity is inconsistent with the registration information, and there may be a risk of identity fraud; if the Euclidean distance is less than the predefined threshold, the system will determine that the feature vectors match, that is, the candidate's identity is consistent with the registration information, and the verification is passed.

[0031] Among them, the threshold can be dynamically adjusted according to different environmental conditions and candidate characteristics. For example, under different lighting conditions, the threshold of facial recognition may need to be adjusted to adapt to different recognition accuracy; in order to improve the accuracy of verification, multiple biometric features can be combined for identity authentication. For example, facial recognition and fingerprint recognition are used at the same time to improve the robustness of recognition by fusing these two feature vectors.

[0032] Preferably, in step S2, randomly selecting questions and setting options based on the online test video includes: A pseudo-random number generator is used to pause the video at the time of answering questions, and randomly draw questions and set options. The video will continue to play after the questions are answered and submitted, or if the questions are not answered within the time limit.

[0033] In the above scheme, during the test, a pseudo-random number generator is used at the preset answering time point to decide whether to pause the video playback. This randomness can prevent candidates from predicting the time when the questions will appear, increasing the fairness of the test; when the video is paused, a pseudo-random number generator is used to randomly extract questions from the question bank, which can ensure that the questions received by each candidate are random and reduce the possibility of cheating; for objective questions, such as multiple-choice questions, a pseudo-random number generator is also used to randomly set the order of options, which can prevent candidates from cheating by memorizing the position of the options; after the candidate completes the answer within the specified time, the system will accept the answer and continue to play the video. If the candidate does not complete the answer within the specified time, the system will automatically submit the unfinished answer and continue to play the video to ensure the continuity of the test process.

[0034] Preferably, the video cannot be dragged and fast-forwarded specifically because: The video content is encrypted, and the video progress bar cannot be dragged or fast-forwarded without correct decryption.

[0035] Preferably, in step S3, monitoring the behavioral characteristics of the examinee during the examination and identifying abnormal behavior by analyzing the stability of the behavioral characteristics includes: Real-time recording of behavioral characteristics during the exam, including mouse movement trajectory and keyboard tapping frequency; The stability data of examinees’ behavioral characteristics were obtained through variance analysis, including the average speed and acceleration of mouse movement and the average and peak frequency of keyboard tapping; Compare the stability data of the examinee's behavioral characteristics with the normal behavior data to identify abnormal behavior.

[0036] In the above scheme, the movement path, speed, acceleration, etc. of the mouse can be recorded by tracking the changes in the plane coordinates of the mouse; the frequency of the examinee's keyboard tapping, including the average tapping frequency and the peak tapping frequency, can be recorded by monitoring keyboard events; through variance analysis, the average speed and acceleration of the examinee's mouse movement, as well as the average frequency and peak frequency of keyboard tapping are obtained; the data baseline of normal behavior is pre-set or obtained through learning, which serves as a reference standard for judging whether the examinee's behavior is normal; if the examinee's behavioral characteristic stability data is significantly different from the normal behavior data baseline, abnormal behavior will be identified.

[0037] Preferably, in step S4, identifying cheating behavior according to the behavioral characteristics of the examinee during the examination and issuing an alarm in a timely manner includes: Cluster analysis is used to cluster the behavioral characteristics of examinees and identify outliers, i.e., possible cheating behaviors, including switching screens and opening search engines; Calculate the distance between the examinee's behavior characteristics and the cluster center to determine whether it is abnormal behavior. If the distance exceeds the preset threshold, it is considered abnormal. Once abnormal behavior is determined, an alarm will be immediately sent to the invigilator so that the invigilator can take timely measures.

[0038] Preferably, in step S5, storing the registration information and the behavioral characteristics of the examinee during the examination process for post-examination verification includes: Store candidate information and behavior data; Provide a data query interface to facilitate subsequent data retrieval and viewing; Back up candidate information and behavior data and upload them to the cloud.

[0039] In the above solution, the personal information of the examinees and their behavior data during the test, such as mouse movement trajectory, keyboard tapping frequency, etc., need to be securely stored in the database. These data are crucial for subsequent data analysis, abnormal behavior identification and verification of test results; the stored data must be encrypted, and a strong encryption algorithm must be used to ensure the security of the data during storage and transmission; the backup data must be uploaded to the cloud, and the storage solutions of the cloud service provider, such as AWS S3 object storage, must be used to ensure high data availability and disaster recovery capabilities.

[0040] Preferably, Figure 2 As shown, a system for preventing cheating in online examinations comprises: Identity verification module, used to confirm the candidate's identity and registration information before the exam begins through facial recognition; The random question module is used to randomly select questions and set options based on the online test video, where the video cannot be dragged or fast-forwarded; Behavior analysis module, which is used to monitor the behavioral characteristics of candidates during the examination and identify abnormal behaviors by analyzing the stability of behavioral characteristics; The cheating detection module is used to identify cheating behaviors based on the behavioral characteristics of candidates during the exam and issue an alarm in a timely manner; The data storage module is used to store registration information and the behavioral characteristics of candidates during the examination process for easy post-examination verification.

[0041] In the above scheme, the identity authentication module verifies the identity of the examinee before the exam by using facial recognition technology, and compares it with the registration information to prevent identity fraud. Secondly, the random question module randomly selects questions and sets options based on the video content to ensure the randomness and fairness of the questions. At the same time, the video playback cannot be controlled by the examinee to prevent cheating. The behavior analysis module monitors the examinee's real-time behavior, such as mouse movement and keyboard strokes, and identifies abnormal behavior by analyzing the stability of the behavior. The cheating detection module further determines cheating behavior based on behavioral characteristics and issues an alarm in time so that the invigilator can take action. Finally, the data storage module is responsible for saving examinee information and behavior data, providing a basis for post-exam inspection and ensuring the security and fairness of the exam.

[0042] The present invention also provides a device, which is a structural schematic diagram of a preferred embodiment of the device provided by the present invention, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, a method for preventing cheating in online examinations as described in any of the above embodiments is implemented.

[0043] It should be noted that, in some specific embodiments, the device may further include multiple memories and / or multiple processors, and their specific number and connection method may be set and adaptively adjusted according to actual needs.

[0044] The present invention also provides a computer-readable storage medium, specifically including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for preventing cheating in online examinations as described in any of the above embodiments.

[0045] It should be noted that the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be further noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0046] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for preventing cheating in online examinations, characterized in that: The following steps are involved: Use facial recognition to confirm the candidate's identity and registration information before the exam begins; Randomly select questions and set options based on the online exam video, in which the video cannot be dragged or fast-forwarded; Monitor the behavioral characteristics of candidates during the exam and identify abnormal behaviors by analyzing the stability of behavioral characteristics; Identify cheating behaviors based on the behavioral characteristics of candidates during the exam and issue an alert in a timely manner; The registration information and the behavioral characteristics of the examinees during the examination are stored for easy post-examination verification.

2. A method for preventing cheating in online examinations according to claim 1, characterized in that: The confirmation of the candidate's identity and registration information by face recognition before the start of the exam includes: Through live body collection, the examinee's face is dynamically recognized to generate several face images; By building a CNN model to extract the feature vectors of face images, the candidate's identity is determined by calculating the Euclidean distance between the feature vectors; The calculation results are fed back to the invigilator in real time. If the verification is passed, the candidate is allowed to enter the examination room; if the verification fails, the candidate is prohibited from taking the exam.

3. A method for preventing cheating in online examinations according to claim 2, characterized in that: The dynamic recognition of the examinee's face through live body collection is specifically as follows: Through the three-dimensional imaging principle of 3D structured light, the three-dimensional features of the face are compared to determine whether the examinee is a real living person; Determine the location of the eye area and use a computer vision processor to analyze the open and closed state of the eyes; Determine the location of the mouth area and use a computer vision processor to analyze the opening and closing state of the mouth; Identify shaking and nodding movements by detecting the distance changes of facial key points.

4. A method for preventing cheating in online examinations according to claim 3, characterized in that: The method of determining the candidate's identity by calculating the Euclidean distance between the feature vectors is specifically as follows: The Euclidean distance between feature vectors is calculated using the Euclidean distance algorithm; The Euclidean distance is compared with a predefined threshold. When the Euclidean distance is greater than the predefined threshold, it is determined that the feature vectors do not match, that is, the candidate's identity is inconsistent with the registration information; When the Euclidean distance is less than a predefined threshold, the feature vectors are considered to match, that is, the candidate's identity is consistent with the registration information.

5. A method for preventing cheating in online examinations according to claim 4, characterized in that: The random selection of questions and setting options based on the online exam video include: A pseudo-random number generator is used to pause the video at the time of answering questions, and randomly draw questions and set options. The video will continue to play after the questions are answered and submitted, or if the questions are not answered within the time limit.

6. A method for preventing cheating in online examinations according to claim 5, characterized in that: The video cannot be dragged and fast-forwarded specifically because: The video content is encrypted, and the video progress bar cannot be dragged or fast-forwarded without correct decryption.

7. A method for preventing cheating in online examinations according to claim 6, characterized in that: The monitoring of the examinee's behavioral characteristics during the examination and identifying abnormal behavior by analyzing the stability of the behavioral characteristics include: Real-time recording of behavioral characteristics during the exam, including mouse movement trajectory and keyboard tapping frequency; The stability data of examinees’ behavioral characteristics were obtained through variance analysis, including the average speed and acceleration of mouse movement and the average and peak frequency of keyboard tapping; Compare the stability data of the examinee's behavioral characteristics with the normal behavior data to identify abnormal behavior.

8. A method for preventing cheating in online examinations according to claim 7, characterized in that: The method of identifying cheating behavior based on the behavioral characteristics of examinees during the examination and issuing an alarm in a timely manner includes: Cluster analysis is used to cluster the behavioral characteristics of examinees and identify outliers, i.e., possible cheating behaviors, including switching screens and opening search engines; Calculate the distance between the examinee's behavior characteristics and the cluster center to determine whether it is abnormal behavior. If the distance exceeds the preset threshold, it is considered abnormal. Once abnormal behavior is determined, an alarm will be immediately sent to the invigilator so that the invigilator can take timely measures.

9. A method for preventing cheating in online examinations according to claim 8, characterized in that: The storage of registration information and behavioral characteristics of examinees during the examination process for post-examination verification includes: Store candidate information and behavior data; Provide a data query interface to facilitate subsequent data retrieval and viewing; Back up candidate information and behavior data and upload them to the cloud.

10. A system for preventing cheating in online examinations, characterized in that: include: Identity verification module, used to confirm the candidate's identity and registration information before the exam begins through facial recognition; The random question module is used to randomly select questions and set options based on the online test video, where the video cannot be dragged or fast-forwarded; Behavior analysis module, which is used to monitor the behavioral characteristics of candidates during the examination and identify abnormal behaviors by analyzing the stability of behavioral characteristics; The cheating detection module is used to identify cheating behaviors based on the behavioral characteristics of candidates during the exam and issue an alarm in a timely manner; The data storage module is used to store registration information and the behavioral characteristics of candidates during the examination process for easy post-examination verification.

Images