Online written test video recording and proctoring methods, electronic devices and storage media

Abstract

The present application provides a video recording and invigilation method, electronic device and storage medium for an online written test, wherein the video recording method for an online written test comprises: establishing an association between a main recording device and at least one auxiliary recording device; the main recording device and at least one auxiliary recording device record the test video of the examinee in real time; encoding the test video; after the main recording device triggers a recording object conversion instruction, the recording object is switched from the examinee to the examinee's answer sheet, the main recording device captures the answer sheet image and records a freeze-frame video, and sends the answer sheet image to a third server; after the main recording device triggers a recording end instruction, the main recording device stops recording and sends the recording end instruction to the virtual examination room of the second server, and if at least one auxiliary recording device polls the recording end instruction of the virtual examination room from the second server, at least one auxiliary recording device stops recording. The present application can improve the invigilation effect of online written tests.

Description

Technical Field

[0001] The present application relates to the field of monitoring technology, and in particular to a video recording and monitoring method for an online written test, an electronic device, and a storage medium. Background Art

[0002] Remote online examinations mainly include online interviews and online written examinations. The current online written examinations can be answered on the computer by answering objective questions, text questions and simple drawing questions, and supervised by the examination computer equipped with a camera and microphone. However, for art examinations, writing and drawing can only be done on paper, and then the candidates upload the paper version of the photo for scoring. However, this method lacks monitoring of the candidates' answer sheets, which easily causes candidates to continue answering questions outside of the time of answering questions. Summary of the invention

[0003] In view of this, it is necessary to provide an online written test proctoring method, electronic equipment and storage medium, which can improve the validity of the test results by monitoring the abnormal behavior of candidates.

[0004] In a first aspect, the present application discloses a method for invigilating an online written test, which is applied to a plurality of recording devices, wherein the plurality of recording devices include a main recording device and at least one auxiliary recording device, and the recording method includes:

[0005] The primary recording device and the at least one auxiliary recording device are associated with each other through a virtual examination room created on the second server;

[0006] The main recording device and the at least one auxiliary recording device record the test video of the examinee in real time;

[0007] Encoding the test video includes:

[0008] The main recording device encodes the recorded test video to generate a first invigilation video, and pushes the first invigilation video to the first server; the at least one auxiliary recording device encodes the recorded test video to generate a second invigilation video, and pushes the second invigilation video to the first server;

[0009] After the main recording device triggers the recording object switching instruction, the recording object is switched from the examinee to the examinee's answer sheet, the main recording device captures the answer sheet image and records a freeze-frame video, and sends the answer sheet image to the third server;

[0010] After the main recording device triggers the recording end instruction, it stops recording and sends the recording end instruction to the virtual examination room of the second server. If the at least one auxiliary recording device polls the recording end instruction of the virtual examination room from the second server, the at least one auxiliary recording device stops recording.

[0011] In some optional implementations, the main recording device records a freeze-frame video after capturing the answer sheet image, including:

[0012] Obtaining a freeze-frame screenshot of the freeze-frame video;

[0013] Comparing the answer sheet image with the frozen screenshot to determine whether the examinee has any abnormal behavior;

[0014] If the answer sheet image does not match the frozen screenshot, it is determined that the examinee has abnormal behavior;

[0015] If the answer sheet image matches the frozen screenshot, it is determined that the examinee has no abnormal behavior.

[0016] In some optional implementations, establishing association between the primary recording device and the at least one auxiliary recording device includes:

[0017] Based on the first identification code of the master recording device, the master recording device generates a QR code;

[0018] After scanning the QR code, the at least one auxiliary recording device obtains the first identification code and stores it as a second identification code, starts recording, and polls the second server with the second identification code;

[0019] The main recording device performs identity authentication on the examinee. After the authentication is passed, the main recording device triggers a recording start instruction, and the main recording device sends the first identification code to the second server;

[0020] After the at least one auxiliary recording device obtains the first identification code matching the second identification code through polling from the second server, an association is established between the at least one auxiliary recording device and the main recording device.

[0021] In some optional implementations, the main recording device and the at least one auxiliary recording device recording the test video of the examinee in real time include:

[0022] The at least one auxiliary recording device sends a virtual examination room creation instruction to the second server, and after receiving the virtual examination room creation instruction, the second server establishes a virtual examination room associated with the at least one auxiliary recording device;

[0023] The at least one auxiliary recording device sends a virtual room creation instruction to the first server, and after receiving the virtual room creation instruction, the first server establishes a virtual room associated with the at least one auxiliary recording device;

[0024] After establishing an association between the at least one auxiliary recording device and the main recording device, the main recording device establishes an association with the virtual room and the virtual examination room associated with the at least one auxiliary recording device, and the main recording device triggers a recording start instruction to send the first monitoring video to the virtual room of the first server. After responding to the recording start instruction, the first server executes cloud storage and stores it in the virtual room.

[0025] The third aspect of the present application discloses a proctoring method based on a proctoring video obtained by using a video recording method for an online written test, which is applied to a proctoring device, wherein the proctoring device is communicatively connected with a first server, a second server, and a third server, and the method comprises:

[0026] Receiving the invigilation video sent from the first server;

[0027] Acquire a video frame from the invigilation video as a reference image;

[0028] Dividing the reference image into blocks, and determining an area not blocked by the examinee as a first stable block;

[0029] Obtaining video screenshots from the invigilation video at preset time intervals;

[0030] Dividing the video screenshot into blocks, and selecting a block of the video screenshot having the same position as the first stable block as a second stable block;

[0031] Calculating a characteristic difference between a first characteristic value of the first stable block and a second characteristic value of the second stable block, and judging whether the examinee has abnormal behavior according to the characteristic difference;

[0032] If the invigilation device polls the second server for an end-of-shooting instruction, it obtains the examinee's answer sheet image from the third server. The end-of-shooting instruction is sent to the second server by the main recording device in the examination room after shooting the answer sheet image. Based on the review of the answer sheet image, it is determined whether the answer sheet image is qualified.

[0033] In some optional implementations, determining whether the answer sheet image is qualified based on the review of the answer sheet image includes:

[0034] Comparing the answer sheet image with a standard answer sheet image to determine whether the answer sheet image is qualified;

[0035] If the answer sheet image is determined to be unqualified, an unqualified instruction is sent to the second server, and the second server controls the main recording device to re-shoot the answer sheet image within a preset time period, obtains the re-shot answer sheet image and sends it to the invigilation device; or

[0036] If the answer sheet image is determined to be qualified, a qualified instruction is sent to the second server, and the second server controls the main recording device and the at least one auxiliary recording device to stop recording the candidate's video.

[0037] In some optional implementations, the calculating the first eigenvalue of the first stable block and the second eigenvalue of the second stable block includes:

[0038] Reducing the image sizes of the first stable block and the second stable block respectively according to the same proportion to obtain proportional images;

[0039] Calculate the grayscale value of each pixel in the proportional image and the grayscale average value of all pixels in the proportional image;

[0040] Compare the grayscale value of each pixel with the grayscale average value to obtain a binary quantized value of each pixel;

[0041] Combining the binary quantization values ​​in a preset order to generate an image fingerprint;

[0042] The image fingerprint of the first stable block is generated as the first feature value, and the image fingerprint of the second stable block is generated as the second feature value.

[0043] In some optional implementations, judging whether the examinee has abnormal behavior according to the feature difference includes:

[0044] If the characteristic difference is greater than a preset threshold, a warning message is output and the examinee is marked as having abnormal behavior;

[0045] If the feature difference is less than or equal to a preset threshold, the reference image of the invigilation video is updated, and the video screenshot is used as the updated reference image.

[0046] The third aspect of the present application discloses an electronic device, which includes: a processor; and a memory, wherein the memory stores a plurality of program modules, and the plurality of program modules are loaded by the processor and execute the above-mentioned online written test video recording method or invigilation method.

[0047] The fourth aspect of the present application discloses a computer-readable storage medium having program instructions stored thereon, wherein the instructions are loaded by a processor and execute the above-mentioned online written test video recording method or invigilation method.

[0048] It should be understood that the electronic device of the third aspect provided above and the computer-readable storage medium described in the fourth aspect both correspond to the methods of the first aspect and the second aspect mentioned above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above and will not be repeated here.

[0049] Compared with the prior art, the video recording and invigilation method, electronic device and storage medium for online written examinations proposed in this application can monitor the examination process of candidates by associating a main recording device and at least one auxiliary recording device. During the examination, the main recording device and at least one auxiliary recording device are used to record the candidate's video, and the monitoring of the abnormal behavior of the candidate by the invigilation device is combined to improve the efficiency of invigilation. When the examination is just over, the main recording device monitors the candidate's answer sheet by freeze-frame video, which improves the validity of the examination results, and the answer sheet is reviewed in combination with the invigilation device, further strengthening the control of the answer sheet quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0051] Figure 1 It is an architectural diagram of the application environment of the online written examination proctoring method provided by the preferred embodiment of the present application.

[0052] Figure 2 It is a flow chart of the video recording method of the online written test provided by this application.

[0053] Figure 3 A flow chart for establishing an association between a primary recording device and at least one secondary recording device.

[0054] Figure 4 The process of judging whether the examinee has abnormal behavior based on the answer sheet image.

[0055] Figure 5 A flowchart of a method for invigilation based on invigilation video obtained by using a video recording method for online written examinations provided in this application.

[0056] Figure 6 A flowchart for determining whether the examinee's behavior is abnormal.

[0057] Figure 7 A schematic diagram of the structure of an electronic device provided for a preferred embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the present application is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0059] In the following description, many specific details are set forth to facilitate a full understanding of the present application. The embodiments described are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0061] See also Figure 1 , Figure 1 An architectural diagram of the application environment of the online written examination proctoring method provided for the preferred embodiment of the present application.

[0062] In this embodiment, the online written test video recording and proctoring method is applied to multiple electronic devices, for example, a main recording device 2, one or more auxiliary recording devices 3, and a proctoring device 4. The main recording device 2 and the auxiliary recording device 3 can be a computer that can run WINDOWS TM / LINUX TM / ANDROID TM The main recording device 2 is used to record the video of the examinee during the examination. The auxiliary recording device 3 is used to record the video of the examinee during the examination. The invigilation device 4 can be a computer that can run a desktop operating system, which is installed with a monitoring video recording program and an invigilation program, and is connected to a camera and a speaker. The invigilation device 4 is used to monitor whether the examinee has abnormal behavior during the examination.

[0063] In this embodiment, the main recording device 2, the auxiliary recording device 3, and the invigilation device 4 respectively establish network connections with the first server 5, the second server 6, and the third server 7. The network can be a wired network or a wireless network, such as radio, wireless fidelity (Wi-Fi), cellular, satellite, broadcast, etc. The cellular network can be a 4G network or a 5G network.

[0064] The first server 5, also called a streaming media service center, may be a single server, a cloud server or a cluster server, and is used to store, forward and encode received audio and video.

[0065] The second server 6, also called a service control center, may be a single server, a cloud server or a cluster server, and is used to receive polls and respond to and push information related to the polls.

[0066] The third server 7, also called a storage center, may be a single server, a cloud server or a cluster server, and is used to store data.

[0067] See also Figure 2 As shown, it is a flowchart of the video recording method of the online written test provided by the present application, which is applied to multiple recording devices, wherein the multiple devices include a main recording device and at least one auxiliary recording device. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.

[0068] 21. The primary recording device and the at least one auxiliary recording device are associated with each other through a virtual examination room created on the second server.

[0069] In one embodiment, the main recording device is used for identity authentication, question issuance, answer sheet photography and control of the examination process, and to shoot a first invigilation video at a close distance near the examinee. The main recording device may be an electronic device with a camera function, and the main recording device controls the examination process as the main device for information exchange, including receiving examinee abnormal prompt instructions, issuing recording start instructions, and recording end instructions.

[0070] During the examination, in addition to the main recording device installed as the recording host, at least one auxiliary recording device is installed in the examination room as an auxiliary device to further assist in monitoring the examination room. Multiple auxiliary recording devices can achieve all-round shooting from different angles.

[0071] For a detailed description of the association process, refer to Figure 3 As shown, it is a flowchart of establishing an association between a main recording device and at least one auxiliary recording device.

[0072] 31. Based on the first identification code of the master recording device, the master recording device generates a QR code.

[0073] The first identification code may be a device ID of a primary recording device, and the first identification code may be used to identify the device identity.

[0074] A login interface is provided in the main recording device. When the examinee logs into the login interface, a QR code is displayed on the login interface of the main recording device.

[0075] 32. After scanning the QR code, the at least one auxiliary recording device obtains the first identification code and stores it as a second identification code, starts recording, and polls the second server with the second identification code.

[0076] When the examinee logs in to at least one auxiliary recording device, at least one auxiliary recording device scans the QR code to obtain the first identification code of the main recording device. It can be understood that there is no limit to the number of auxiliary recording devices, as long as the auxiliary recording device scans the QR code of the main recording device, the first identification code can be obtained. The auxiliary recording device stores the first identification code of the main recording device in the local end of the auxiliary recording device. For the sake of distinction, the first identification code stored in the local end of the auxiliary recording device is called the second identification code, which is used for subsequent matching and association with the main recording device.

[0077] 33. The main recording device performs identity authentication on the examinee. After the authentication is passed, the main recording device triggers a recording start instruction and sends the first identification code to the second server.

[0078] The examinee logs in to the main recording device, and the method of identity authentication is not limited to face recognition. After the authentication is passed, the recording start instruction is triggered and the first identification code is sent to the second server.

[0079] 34. After at least one auxiliary recording device polls the first identification code matching the second identification code from the second server, establish an association between the at least one auxiliary recording device and the main recording device.

[0080] In one embodiment, after at least one auxiliary recording device obtains the second identification code, the at least one auxiliary recording device carries the second identification code to poll the second server at preset time intervals. It can be understood that in order to improve the efficiency of polling to obtain relevant information, the auxiliary recording device carries the second identification code during the polling process.

[0081] After at least one auxiliary recording device polls the first identification code matching the second identification code from the second server, an association is established between the at least one auxiliary recording device and the main recording device.

[0082] In one embodiment, while at least one auxiliary recording device carries a second identification code to poll the second server, at least one auxiliary recording device sends a virtual examination room creation instruction to the second server. After receiving the virtual examination room creation instruction, the second server establishes a virtual examination room associated with the at least one auxiliary recording device.

[0083] At the same time, at least one auxiliary recording device sends a virtual room creation instruction to the first server. After receiving the virtual room creation instruction, the first server establishes a virtual room associated with the at least one auxiliary recording device.

[0084] After at least one auxiliary recording device polls the first identification code matching the second identification code from the second server, an association is established between the at least one auxiliary recording device and the main recording device.

[0085] After establishing an association between at least one auxiliary recording device and the main recording device, the main recording device is associated with the virtual room associated with the at least one auxiliary recording device and the virtual examination room, and the main recording device triggers a recording start instruction to send the first invigilation video to the virtual room of the first server. After responding to the recording start instruction, the first server executes cloud storage and stores it in the virtual room.

[0086] In the above embodiment, the main recording device and at least one auxiliary recording device are associated. On the one hand, after successfully comparing the first identification code and the second identification code, the same virtual room and virtual examination room can be associated by creating a virtual room and a virtual examination room after the auxiliary recording device starts recording. At this time, the auxiliary recording device also carries the second identification code to poll the second server when requesting to create the virtual room and the virtual examination room. On the other hand, the virtual room and the virtual examination room can be associated directly in the second server by comparing the first identification code and the second identification code without creating the virtual room and the virtual examination room.

[0087] In this embodiment, after the primary recording device and at least one secondary recording device are associated, they sense each other through a heartbeat mechanism. The heartbeat mechanism means that the sender sends a fixed-format message to the receiver according to certain rules, and the receiver replies with a fixed-format message after receiving the message. If no reply message is received for a long time, it is considered that the current connection has failed. When the secondary recording device is offline, the primary recording device will issue a voice warning or text reminder so that the examinee can rescan the QR code displayed by the primary recording device and reestablish the connection.

[0088] 22. The main recording device and at least one auxiliary recording device record the test video of the examinee in real time.

[0089] The examination video includes a first invigilation video generated by encoding the examination video by a main recording device, and a second invigilation video generated by encoding by at least one auxiliary recording device.

[0090] In one embodiment, after the main recording device establishes an association with at least one auxiliary recording device, the main recording device triggers a recording start command and sends a first monitoring video to the virtual room of the first server. After responding to the recording start command, the first server executes cloud storage and stores any received monitoring video in the corresponding virtual room.

[0091] Specifically, after requesting the first server to create a virtual room, at least one auxiliary recording device starts streaming the second proctoring video to the virtual room. After the main recording device establishes an association with at least one auxiliary recording device, the main recording device streams the first proctoring video to the virtual room when the recording start instruction is triggered. It can be understood that, according to the above-established association, the virtual room streamed by the main recording device is the same as the virtual room streamed by at least one auxiliary recording device.

[0092] 23. The main recording device encodes the recorded test video to generate a first invigilation video, and streams the first invigilation video to the first server.

[0093] The main recording device encodes the recorded test video to generate a first invigilation video and a first recorded video, wherein the resolution of the first invigilation video is smaller than the resolution of the first recorded video, and the bit rate of the first invigilation video is smaller than the bit rate of the first recorded video.

[0094] The first invigilation video is used for streaming to the first server, and the first recorded video is used for storage in the main recording device. It is understandable that the resolution and bit rate of the first invigilation video and the first recorded video may be the same or different, and are not limited in practical applications. For example, the network requirements for the first invigilation video for streaming are relatively high. Video encoding uses H.264 or H.265 encoding depending on the model, and the bit rate is controlled within the range of 200kbps to 1000kbps, the resolution is 180P to 1080P, and the frame rate is 5fps to 30fps.

[0095] 24, the at least one auxiliary recording device encodes the recorded test video to generate a second invigilation video, and streams the second invigilation video to the first server. It should be noted that 23 and 24 are encoded synchronously by the main recording device and at least one auxiliary recording device, respectively, without any time sequence restriction.

[0096] The auxiliary recording device encodes the recorded examination video to generate a second invigilation video and a second recorded video, wherein the resolution of the second invigilation video is smaller than the resolution of the second recorded video, and the bit rate of the second invigilation video is smaller than the bit rate of the second recorded video.

[0097] The second invigilation video is used for streaming to the first server, and the second recorded video is used for storage in the auxiliary recording device. It is understandable that the resolution and bit rate of the second invigilation video and the second recorded video can be the same. In this case, the network requirements for the second invigilation video to be streamed are relatively high. Video encoding uses H.264 or H.265 encoding depending on the model, and the bit rate is controlled within the range of 200kbps to 1000kbps, the resolution is 180P to 1080P, and the frame rate is 5fps to 30fps.

[0098] 25. After the main recording device triggers the recording object conversion instruction, the recording object is switched from the examinee to the examinee's answer sheet, the main recording device captures the answer sheet image and records a freeze-frame video, and sends the answer sheet image to a third server.

[0099] After triggering the recording object switching instruction, the main recording device stops recording the first invigilation video and switches the recording object from the examinee to the examinee's answer sheet. In practical applications, a small interval time can be allowed during this switching recording process, and the interval time can be 30 seconds.

[0100] After the main recording device completes shooting the answer sheet image, it sends a shooting end instruction to the second server and enters the stage of recording the freeze-frame video. After the main recording device enters the freeze-frame video recording, the main recording device freezes the freeze-frame video every certain period of time, for example, 5 seconds, so as to obtain a freeze-frame screenshot of the entire answer area in the test answer sheet, wherein the clarity of the freeze-frame screenshot is lower than the clarity of the answer sheet image. It can be understood that the first invigilation video and the freeze-frame video are recorded separately, and each is an independent video.

[0101] The resolution of the first invigilation video and the second invigilation video may be the same, which is different from the resolution of the freeze-frame video, and the resolution of the first invigilation video and the second invigilation video is smaller than the resolution of the freeze-frame video. It is understandable that the first invigilation video and the second invigilation video are videos recorded by two recording devices respectively, and the resolution of the first invigilation video and the second invigilation video may be different. The resolution of the freeze-frame video is above 1280*720 resolution, and can be a high-definition video or an ultra-high-definition video.

[0102] After acquiring the freeze-frame screenshot and the answer sheet image, the main recording device analyzes the freeze-frame screenshot and the answer sheet image.

[0103] Specifically, the main recording device compares the examinee's answer sheet image with the freeze-frame screenshot to determine, for example, whether there is an occluded area in the freeze-frame screenshot. The occluded area may be a black area caused by the examinee's hand when the photo is taken at the end of the exam while the examinee is still answering questions, or the freeze-frame screenshot contains answer traces that do not exist on the answer sheet image during the matching process of the answer sheet image.

[0104] The examinee's answer sheet image is compared with the freeze-frame screenshot to obtain a judgment result, which is whether abnormal behavior exists or not.

[0105] When the matching rate is greater than or equal to 90%, it is considered that the answer sheet image matches the freeze-frame screenshot. It can be understood that a certain error is allowed in the matching process, and the candidate does not have any abnormal behavior.

[0106] When the matching rate is less than 90%, it is considered that the answer sheet image and the freeze-frame screenshot do not match. The test paper may be obscured, causing shadows in the answer sheet screenshot. It is understandable that if the answer sheet is obscured, the answer sheet image and the freeze-frame screenshot will be very different, and the candidate may have abnormal behavior. The main recording device issues a prompt instruction to remind the candidate to deal with the obscured area. If the matching rate is still less than 90% within the preset number of prompts, it is determined that the candidate has abnormal behavior. The abnormal behavior may be that there are traces of answers in the answer sheet screenshot that are not in the answer sheet image.

[0107] When it is determined that the examinee has no abnormal behavior, the first invigilation video, answer sheet image, and freeze-frame screenshot are stored in the third server. After the test recording is completed, the auxiliary recording device also stores the second invigilation video in the third server for backup. Three copies can be backed up at the same time.

[0108] When the main recording device receives the recording object switching instruction, it obtains the image of the examinee's answer sheet. At this time, the answering time has ended, but the auxiliary recording device is still recording the examinee's second invigilation video. If the examinee continues to answer, it will be considered abnormal. Therefore, when the main recording device receives the recording object switching instruction, it switches the recording of the test to the recording of the answer sheet. Before starting the freeze-frame recording, it takes a photo of the examinee's answer sheet and uses the answer sheet image at this time as the examinee's final answer sheet style. At the same time, in order to track in real time whether the examinee continues to answer, the main recording device stops recording the first invigilation video and starts recording the freeze-frame video of the answer sheet, and obtains the freeze-frame screenshot of the examinee's answer sheet accordingly.

[0109] In this real-time system, the behavior of candidates is monitored in real time within a short period of time after the exam. By analyzing the answer sheets, further control of abnormal behavior is achieved, thereby improving the validity of the scores.

[0110] like Figure 4 As shown, it is a flow chart for judging whether a examinee has abnormal behavior based on the answer sheet image.

[0111] 41. Get a freeze-frame screenshot of the freeze-frame video.

[0112] After the main recording device enters the freeze-frame video recording, the main recording device takes freeze-frame screenshots of the freeze-frame video at regular intervals, for example, every 5 seconds, thereby obtaining freeze-frame screenshots of the entire answer area in the test answer sheet.

[0113] 42. Compare the answer sheet image and the freeze-frame screenshot to determine whether the candidate has any abnormal behavior.

[0114] 43. If the answer sheet image does not match the freeze-frame screenshot, it is determined that the candidate has behaved abnormally.

[0115] The candidate's answer sheet image is compared with the freeze-frame screenshot. When the matching rate is less than 90%, it is considered that the candidate's answer sheet image does not match the freeze-frame screenshot. For example, in the process of comparing the freeze-frame screenshot with the answer sheet photo, there are different answers corresponding to the same test question or the number of words does not match. It is considered that the candidate continues to answer after the test is over, and the candidate's behavior is considered abnormal.

[0116] Exemplarily, when the matching rate is less than 90%, it is considered that the answer sheet image and the freeze-frame screenshot do not match. The test paper may be obscured, causing shadows in the answer sheet screenshot. It is understandable that if the answer sheet is obscured, the answer sheet image and the freeze-frame screenshot will be quite different, and the candidate may have abnormal behavior. The main recording device issues a prompt instruction to remind the candidate to deal with the obscured area. If the matching rate is still less than 90% within the preset number of prompts, it is determined that the candidate has abnormal behavior. The abnormal behavior may be that traces of answers appear in the answer sheet screenshot that are not in the answer sheet image.

[0117] 44. If the answer sheet image matches the freeze-frame screenshot, it is determined that the candidate has not behaved abnormally.

[0118] When the matching rate is greater than or equal to 90%, the freeze-frame screenshot corresponding to the candidate's answer sheet image matches, and it is considered that the candidate did not continue to answer questions on the answer sheet after the exam. It can be understood that a certain error is allowed in the matching process and there is no abnormal behavior.

[0119] In this embodiment, since it is determined that the examinee cannot continue to answer questions at the end of the test, many examinees will continue to answer questions beyond the test time, but the prior art lacks control over this time period, so that the examinee continues to answer questions or modifies the answer sheet after the test. Therefore, this application strengthens the control of examinees' abnormal behavior after the test by monitoring the test paper after the test, which can improve the effectiveness of the test and achieve fairness and justice in the test.

[0120] 26. After the main recording device triggers the recording end instruction, it sends the recording end instruction to the virtual examination room of the second server. If the at least one auxiliary recording device polls the recording end instruction of the virtual examination room from the second server, the at least one auxiliary recording device stops recording.

[0121] Polling refers to a method by which the CPU decides how to provide services to peripheral devices, also known as programmed I / O. The concept of polling is that the CPU periodically sends out inquiries to each peripheral device in order to see if it needs its service. If so, it will provide service. After the service is completed, it will ask the next peripheral device, and polling will be repeated at intervals.

[0122] After the association between the primary recording device and the at least one auxiliary recording device is established, when the at least one auxiliary recording device polls a recording end instruction, the video recording is stopped.

[0123] Figure 5 , Figure 5 The invention discloses an invigilation method based on an invigilation video obtained by using a video recording method for an online written test, and is applied to an invigilation device, wherein the invigilation device is communicatively connected with a first server, a second server and a third server.

[0124] 51. Receive the invigilation video sent from the first server.

[0125] The monitoring video includes a first monitoring video of a primary recording device and a second monitoring video of a secondary recording device.

[0126] The invigilation device establishes a communication connection with the main recording device and the auxiliary recording device via the first server, the second server and the third server, wherein the invigilation device includes an invigilation module and an audit module.

[0127] The invigilation module mainly obtains the first invigilation video and the second invigilation video in the first server, and monitors the first invigilation video and the second invigilation video for abnormal behavior of the examinee. The invigilation module and the audit module can communicate with the second server via instant messages respectively, and the invigilation module and the audit module can also exchange messages by periodically polling the second server.

[0128] The invigilation module obtains the virtual examination room by polling the second server and monitors the virtual examination room in real time. The virtual examination room is a virtual examination room associated with both the main recording device and the auxiliary recording device. Through the virtual examination room, the invigilator can interact with the examinee in real time.

[0129] 52. Acquire a video frame from the invigilation video as a reference image.

[0130] For example, when the invigilation device receives the invigilation video, a screenshot is taken of the initial video frame of the invigilation video, and the initial video frame is used as a reference image. In practical applications, other video frames can also be selected as reference images according to requirements.

[0131] 53. Divide the reference image into blocks and determine the area not blocked by the examinee as the first stable block.

[0132] After setting the reference images of the first invigilation video and the second invigilation video, the reference images are divided into blocks, and the reference images are divided into 16 or 32 rectangular blocks of equal size, and the surrounding blocks that are not blocked by the examinees in the reference images of the first invigilation video and the second invigilation video are selected as the first stable blocks.

[0133] 54. Obtain video screenshots from the invigilation video at every preset time period.

[0134] In the process of real-time monitoring of the examinee by the invigilation module, the first invigilation video and the second invigilation video are obtained in real time, and video frames of the first invigilation video and the second invigilation video are captured as video screenshots every N (1 to 60) seconds in the invigilation module.

[0135] 55. Divide the video screenshot into blocks, and select the video screenshot block with the same position as the first stable block as the second stable block.

[0136] Similarly, the obtained video screenshot is divided into 16 or 32 rectangular blocks of the same size as the reference image, and the block at the same position as the first stable block is selected as the second stable block.

[0137] 56. Calculate a characteristic difference between a first characteristic value of the first stable block and a second characteristic value of the second stable block, and determine whether the examinee has abnormal behavior according to the characteristic difference.

[0138] After obtaining the first stable block and the second stable block, the first eigenvalue of the first stable block and the second eigenvalue of the second stable block are calculated respectively, and whether there is abnormal behavior is determined based on the comparison between the characteristic difference between the calculated first eigenvalue and the second eigenvalue and a preset threshold.

[0139] This embodiment uses a perceptual hash algorithm (PHA) to calculate the feature difference between the first eigenvalue and the second eigenvalue. The specific calculation process is as follows:

[0140] 1) Reducing the image sizes of the first stable block and the second stable block respectively according to the same proportion to obtain proportional images.

[0141] Specifically, the images of the first stable block and the second stable block are reduced to N*N, where N can be 8 as the scale image, with a total of 64 pixels. The purpose of the preprocessing of reducing the image is to remove the details of the image, retain only basic image information such as structure and brightness, and reduce the calculation error of the image.

[0142] 2) Calculate the grayscale value of each pixel in the proportional image and the grayscale average of all pixels.

[0143] Obtain 64 pixels in the proportional images of the first stable block and the second stable block respectively, and generate a grayscale value of each pixel;

[0144] Calculate the grayscale average of 64 pixels in the first region ratio image.

[0145] Calculate the grayscale average of 64 pixels in the second region ratio image.

[0146] 3) Compare the grayscale value of each pixel with the grayscale average value to obtain a binary quantization value of each pixel.

[0147] Since each pixel has a grayscale value, the grayscale value of each pixel is compared with the grayscale average value of all pixels. It can be understood that all pixels refer to 64 pixels in the proportional image of the first stable block or 64 pixels in the proportional image of the second stable block.

[0148] The grayscale value of each pixel is compared with the grayscale average value to obtain the binary quantization value of each pixel. Specifically, if the grayscale value is greater than or equal to the grayscale average value, the corresponding pixel is recorded as 1, otherwise it is recorded as 0.

[0149] 4) Combining the binary quantization values ​​in a preset order to generate an image fingerprint.

[0150] The binary quantization values ​​obtained according to 3) above are combined in order as the fingerprint of the image. There is no specific setting for the combination order of the binary quantization values, and it is only necessary to ensure that the order of the binary quantization values ​​of the combined pixels in the first stable block and the second stable block is consistent.

[0151] 5) The fingerprint of the first stable block is generated as the first feature value, and the image fingerprint of the second stable block is generated as the second feature value.

[0152] The image fingerprints corresponding to the first stable block and the second stable block are used as corresponding feature values, so as to subsequently calculate feature difference values ​​to compare the image fingerprints corresponding to the first stable block and the second stable block.

[0153] See also Figure 6 As shown, this is a flow chart for determining whether there is abnormality in the examinee's behavior.

[0154] 61. Calculate the feature difference between the first feature and the second feature, and determine whether the feature difference is greater than a preset threshold.

[0155] The image fingerprint of the first stable block and the image fingerprint of the second stable block are calculated, the image fingerprints of different stable blocks are compared, and the difference between the image fingerprints of different stable blocks is used as a feature difference.

[0156] 62. If the feature difference is greater than the preset threshold, a warning message is output and the examinee is marked as having abnormal behavior.

[0157] During the examination, if the examinee's movements are too large while answering questions, the difference between the preset reference image and the real-time video screenshot will be too large, that is, the feature difference is greater than the preset threshold, which will be judged as suspicious behavior. When suspicious behavior is determined, a warning message will be issued in the invigilation module to remind the invigilator to check. If no abnormality is found, the reference image will be reset to prevent misjudgment again. If the invigilator determines that there is indeed an abnormality after checking, the invigilation module will issue a stop test instruction to the second server and mark the examinee for abnormal behavior. The second server will then send the stop test instruction to the main recording device to notify the examinee to stop answering. Among them, marking the examinee for abnormal behavior can be recording the examinee's seat number.

[0158] 63. If the feature difference is less than or equal to a preset threshold, update the reference image of the first invigilation video and the second invigilation video, and use the video screenshot as the updated reference image.

[0159] When the feature difference is less than or equal to the preset threshold, it indicates that the examinee has no abnormal behavior. The absence of abnormal behavior only means that the examinee's movements are relatively small during the test, but small body movements may also exist. Therefore, when no abnormal behavior of the examinee is found, the reference image is reset.

[0160] 64. Based on the updated reference image, calculate the feature difference between the updated reference image and the real-time video screenshot, and compare it with the preset threshold again.

[0161] In order to prevent misjudgment, the monitoring module uses the obtained video screenshots as updated reference images as the basis for judging the abnormal behavior of the examinee, and again uses the above-mentioned calculation method for comparative analysis to achieve real-time monitoring of the examinee's behavior.

[0162] The present application is not limited to the perceptual hash algorithm when calculating the characteristic value of the stable block, and can also use the Hamming distance algorithm and the histogram comparison algorithm.

[0163] The similarity calculation of the first stable block and the second stable block is performed using the Hamming distance, including: respectively calculating the number of different characters in the corresponding positions of the length string in the first stable block and the second stable block, for example, the string of the first stable block is "1011101", and the string of the second stable block is "1001001", then the Hamming distance between the string of the first stable block and the string of the second stable block is 2. If the preset threshold is 1, then the first stable block and the second stable block do not match. At this time, the proctoring module will issue a warning and mark the examinee for abnormal behavior.

[0164] A histogram comparison algorithm is used to calculate the similarity between the first stable block and the second stable block, including: obtaining image pixels corresponding to the first stable block and the second stable block, generating respective histogram data according to the image pixels, and using a Bhattacharyya coefficient algorithm to calculate the histogram data respectively to calculate the similarity between the first stable block and the second stable block.

[0165] During the examination, the examinee's abnormal behavior is analyzed through the invigilation module. If the examinee has no abnormal behavior before the end of the examination, the main recording device starts to shoot the examinee's answer sheet after the examination and further monitors whether there is abnormal behavior. If the examinee stops answering within the specified time, it is determined that the examinee has no abnormal behavior. After it is determined that the examinee has no abnormal behavior, the first invigilation video, the second invigilation video, the freeze-frame video, the freeze-frame screenshot, and the answer sheet image are all stored in the third server for subsequent scoring of the examinee.

[0166] 57. If the invigilation device polls the second server for the end-of-shooting instruction, it obtains the examinee's answer sheet image from the third server. The end-of-shooting instruction is sent to the second server by the main recording device in the examination room after shooting the answer sheet image. Based on the review of the answer sheet image, it is determined whether the answer sheet image is qualified.

[0167] In one embodiment, the answer sheet image is compared with the standard answer sheet image to determine whether the answer sheet image is qualified; if the answer sheet image is determined to be unqualified, an unqualified instruction is sent to the second server, and the second server controls the main recording device to re-shoot the answer sheet image within a preset time period, obtain the re-shot answer sheet image and send it to the invigilation device; or if the answer sheet image is determined to be qualified, a qualified instruction is sent to the second server, and the second server controls the main recording device and at least one auxiliary recording device to stop recording the candidate's video.

[0168] Specifically, after the invigilator receives the shooting end instruction, it determines whether the answer sheet image meets the shooting requirements based on the review of the answer sheet image, wherein the review content may be to determine whether there are black blocks on the answer sheet caused by the occlusion area, and the occlusion area may also be the shooting shadow caused by placing the answer tool on the answer sheet, and the answer tool is not limited to a pen, an eraser, etc. It can be understood that in the process of reviewing the answer sheet image, if there is an answer sheet area that does not meet the requirements, a reminder is issued in the form of voice or mark, so that the invigilator can check and judge in time, and input the judgment result in the invigilator device and mark the block that does not meet the requirements, which speeds up the review efficiency and ensures the effectiveness of the review. In addition, in the process of reviewing the answer sheet image, the first invigilator video or the second invigilator video of the auxiliary recording device of the examinee is displayed in real time, which facilitates the invigilator and the examinee to exchange messages in a timely manner, and improves the efficiency of the review. For example, the examinee can be directly told which parts of the answer sheet image do not meet the requirements and how to re-photograph the answer sheet image. The efficiency of the examinee in obtaining the answer sheet review results is improved, and the time for re-shooting is saved.

[0169] Before the exam begins, a standard answer sheet image is preset in the review module as a reference and compared with the answer sheet image to see if any answer area is blocked.

[0170] If the answer sheet image is determined to be unqualified, an unqualified instruction is sent to the second server, and the second server controls the main recording device to re-shoot the answer sheet image within a preset time period, obtains the re-shot answer sheet image and sends it to the invigilation device for review again.

[0171] The present application uses freeze-frame video to determine whether there is abnormal behavior after the test, thereby ensuring the validity of the answer sheet image and avoiding abnormal behavior of candidates during the first shooting of the answer sheet image and the review period, wherein the abnormal behavior of candidates includes continuing to answer questions or modifying existing answers after the test.

[0172] If the answer sheet image is determined to be qualified, a qualified instruction is sent to the second server, and the second server controls the main recording device and at least one auxiliary recording device to stop recording the candidate's video.

[0173] If the answer sheet image is judged to be qualified during the review process, for example, there is no obstruction in the answer area of ​​the answer sheet image, the answer sheet image is directly stored in the third server as a basis for subsequent scoring.

[0174] In this embodiment, by reviewing the answer sheet image, the quality of obtaining the answer sheet image is improved, and effective scoring of the answer sheet image is facilitated.

[0175] This application monitors the abnormal behavior of candidates during the examination in real time through the main recording device and the auxiliary recording device, thereby realizing real-time monitoring of candidates during the examination. Furthermore, by obtaining the answer sheet image and comparing it with the still-frame video, it is possible to monitor whether the candidates answer within the specified time, and mark the candidates who answer beyond the specified time as having abnormal behavior, thereby ensuring the fairness and justice of the examination. Compared with the prior art, in terms of examination security and prevention of abnormalities, this application uses multiple recording devices to identify abnormal behavior and other methods to issue abnormal warnings in a timely manner, uses the method of still-frame recording of answer sheets to strengthen the control of candidates' abnormal behavior, and combines manual review of answer sheet images to further ensure the quality of answer sheet images and improve the validity of examination results.

[0176] In the above embodiment, a monitoring video recording and invigilation method is realized through the coordinated operation of the main recording device, the auxiliary recording device, the first server, the second server, the third server and the invigilation device. It should be noted that in other embodiments, the above first server, the second server, the third server and the invigilation device can also be integrated into the same device or two devices for implementation, and the actual application is not limited to this. For example, the first server, the second server and the third server can be integrated into a single server to perform the functions of the above two servers. For another example, the first server, the second server and the third server and the invigilation device can also be integrated into a single device to perform the corresponding functions.

[0177] See also Figure 7 As shown, it is a schematic diagram of the structure of the preferred implementation of various electronic devices of the present application.

[0178] In the embodiment of the present application, the electronic device may be the main recording device 2, the proctoring device 4 and the auxiliary recording device 3, and the main recording device 2, the proctoring device 4 and the auxiliary recording device 3 include, but are not limited to, a processor 10, a memory 20 and a computer program 30 stored in the memory 20 and executable on the processor 10, such as a proctoring program for an online written test. When the processor 10 executes the computer program 30, the steps in the proctoring method for an online written test are implemented.

[0179] Those skilled in the art will appreciate that the schematic diagram is merely an example of the main recording device 2, the invigilation device 4 and the auxiliary recording device 3, and does not constitute a limitation on the main recording device 2, the invigilation device 4 and the auxiliary recording device 3, and may include more or less components than those shown in the diagram, or a combination of certain components, or different components. For example, the main recording device 2, the invigilation device 4 and the auxiliary recording device 3 may also include input and output devices, network access devices, buses, etc.

[0180] The processor 10 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor 10 may be any conventional processor, etc. The processor 10 is the control center of the main recording device 2, the invigilation device 4 and the auxiliary recording device 3, and uses various interfaces and lines to connect the various parts of the entire main recording device 2, the entire invigilation device 4 and the entire auxiliary recording device 3.

[0181] The memory 20 can be used to store the computer program 30 and / or modules / units. The processor 10 implements various functions of the main recording device 2, the invigilation device 4 and the auxiliary recording device 3 by running or executing the computer program and / or modules / units stored in the memory 20 and calling the data stored in the memory 20. The memory 20 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data (such as audio data, a phone book, etc.) created according to the use of the main recording device 2, the invigilation device 4 and the auxiliary recording device 3, etc. In addition, the memory 20 can include volatile and non-volatile memories, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other storage devices. The display screen 40 is a liquid crystal display (LCD) or an organic light-emitting diode (OLED) display screen.

[0182] If the modules / units integrated by the main recording device 2, the invigilation device 4 and the auxiliary recording device 3 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the present application 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 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, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory).

[0183] The video recording and supervision method, electronic device and storage medium for online written examinations provided in this application can shoot and upload videos at different angles through multiple commonly used electronic devices, and compare and analyze the videos to achieve real-time monitoring of online examinations, effectively ensuring the validity of the results of online examinations.

[0184] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the spirit or basic features of the present application. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the present application is limited by the attached claims rather than the above description, so it is intended to include all changes that fall within the meaning and scope of the equivalent elements of the claims in the present application. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the device claim can also be implemented by the same unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0185] The above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person skilled in the art should understand that the technical solution of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present application.

Claims

1. A video recording method for an online written test, applied to multiple recording devices, characterized in that: The plurality of recording devices include a main recording device and at least one auxiliary recording device, and the recording method includes: The primary recording device and the at least one auxiliary recording device are associated with each other through a virtual examination room created on the second server; The main recording device and the at least one auxiliary recording device record the test video of the examinee in real time; Encoding the test video includes: The main recording device encodes the recorded test video to generate a first invigilation video, and pushes the first invigilation video to the first server; the at least one auxiliary recording device encodes the recorded test video to generate a second invigilation video, and pushes the second invigilation video to the first server; After the main recording device triggers the recording object switching instruction, the recording object is switched from the examinee to the examinee's answer sheet, the main recording device captures the answer sheet image and records a freeze-frame video, and sends the answer sheet image to the third server; After the main recording device triggers the recording end instruction, it stops recording and sends the recording end instruction to the virtual examination room of the second server. If the at least one auxiliary recording device polls the recording end instruction of the virtual examination room from the second server, the at least one auxiliary recording device stops recording.

2. The video recording method for online written examination as claimed in claim 1, characterized in that: The main recording device records a freeze-frame video after taking the answer sheet image, including: Obtaining a freeze-frame screenshot of the freeze-frame video; Comparing the answer sheet image with the frozen screenshot to determine whether the examinee has any abnormal behavior; If the answer sheet image does not match the frozen screenshot, it is determined that the examinee has abnormal behavior; If the answer sheet image matches the frozen screenshot, it is determined that the examinee has no abnormal behavior.

3. The video recording method for online written examination as claimed in claim 1, characterized in that: The primary recording device establishing an association with the at least one auxiliary recording device comprises: Based on the first identification code of the master recording device, the master recording device generates a QR code; After scanning the QR code, the at least one auxiliary recording device obtains the first identification code and stores it as a second identification code, starts recording, and polls the second server with the second identification code; The main recording device performs identity authentication on the examinee. After the authentication is passed, the main recording device triggers a recording start instruction, and the main recording device sends the first identification code to the second server; After the at least one auxiliary recording device obtains the first identification code matching the second identification code through polling from the second server, an association is established between the at least one auxiliary recording device and the main recording device.

4. The video recording method for online written examination as claimed in claim 3, characterized in that: The main recording device and the at least one auxiliary recording device record the test video of the examinee in real time, including: The at least one auxiliary recording device sends a virtual examination room creation instruction to the second server, and after receiving the virtual examination room creation instruction, the second server establishes a virtual examination room associated with the at least one auxiliary recording device; The at least one auxiliary recording device sends a virtual room creation instruction to the first server, and after receiving the virtual room creation instruction, the first server establishes a virtual room associated with the at least one auxiliary recording device; After establishing an association between the at least one auxiliary recording device and the main recording device, the main recording device establishes an association with the virtual room and the virtual examination room associated with the at least one auxiliary recording device, and the main recording device triggers a recording start instruction to send the first monitoring video to the virtual room of the first server. After responding to the recording start instruction, the first server executes cloud storage and stores it in the virtual room.

5. A proctoring method based on a proctoring video obtained by using the video recording method for an online written test according to any one of claims 1 to 4, applied to a proctoring device, characterized in that: The invigilation device is connected to the first server, the second server and the third server in communication, and the method includes: Receiving the invigilation video sent from the first server; Acquire a video frame from the invigilation video as a reference image; Dividing the reference image into blocks, and determining an area not blocked by the examinee as a first stable block; Obtaining video screenshots from the invigilation video at preset time intervals; Dividing the video screenshot into blocks, and selecting a block of the video screenshot having the same position as the first stable block as a second stable block; Calculating a characteristic difference between a first characteristic value of the first stable block and a second characteristic value of the second stable block, and judging whether the examinee has abnormal behavior according to the characteristic difference; If the invigilation device polls the second server for an end-of-shooting instruction, it obtains the examinee's answer sheet image from the third server. The end-of-shooting instruction is sent to the second server by the main recording device in the examination room after shooting the answer sheet image. Based on the review of the answer sheet image, it is determined whether the answer sheet image is qualified.

6. The invigilation method according to claim 5, characterized in that: The step of determining whether the answer sheet image is qualified based on the review of the answer sheet image includes: Comparing the answer sheet image with a standard answer sheet image to determine whether the answer sheet image is qualified; If the answer sheet image is determined to be unqualified, an unqualified instruction is sent to the second server, and the second server controls the main recording device to re-shoot the answer sheet image within a preset time period, obtains the re-shot answer sheet image and sends it to the invigilation device; or If the answer sheet image is determined to be qualified, a qualified instruction is sent to the second server, and the second server controls the main recording device and the at least one auxiliary recording device to stop recording the candidate's video.

7. The invigilation method according to claim 5, characterized in that: The method further comprises: Reducing the image sizes of the first stable block and the second stable block respectively according to the same proportion to obtain proportional images; Calculate the grayscale value of each pixel in the proportional image and the grayscale average value of all pixels in the proportional image; Compare the grayscale value of each pixel with the grayscale average value to obtain a binary quantized value of each pixel; Combining the binary quantization values ​​in a preset order to generate an image fingerprint; The image fingerprint of the first stable block is generated as the first feature value, and the image fingerprint of the second stable block is generated as the second feature value.

8. The invigilation method according to claim 5, characterized in that: The determining, based on the characteristic difference, whether the examinee has abnormal behavior includes: If the characteristic difference is greater than a preset threshold, a warning message is output and the examinee is marked as having abnormal behavior; If the feature difference is less than or equal to a preset threshold, the reference image of the invigilation video is updated, and the video screenshot is used as the updated reference image.

9. An electronic device, characterized in that: The electronic device comprises a processor and a memory: a processor, which stores program instructions; A memory for reading the program instructions to implement the video recording method for an online written test as described in any one of claims 1 to 4 or the invigilation method as described in any one of claims 5 to 8.

10. A computer-readable storage medium having program instructions stored thereon, characterized in that: The program instructions are loaded by the processor and execute the video recording method for online written examinations as described in any one of claims 1 to 4 or the invigilation method as described in any one of claims 5 to 8.

Images