A method for video conferencing network security management

Abstract

The present invention relates to the technical field of video conferencing, and specifically to a video conferencing network security management method, including: sending a meeting request to the participants of the video conference to obtain meeting-related information; authenticating the identities of the participants of the video conference to determine the object categories for interaction and the meeting configuration; authenticating the identities of the participants of the video conference to obtain identity authentication information; obtaining the video interaction information after identity authentication, and determining the video association authentication score according to the video interaction information and the identity authentication information; verifying the meeting association score associated with the participants of the video conference according to the video association authentication score; performing abnormal detection and analysis on the surrounding environment of the participants of the video conference and setting an abnormal association score; using the video association authentication score, the meeting association score, and the abnormal association score as retrieval conditions to determine the comprehensive evaluation score of the current video conference; improving the efficiency and security of meeting management.

Description

Technical Field

[0001] The present invention relates to the technical field of video conferencing, and specifically to a method for video conferencing network security management. Background Art

[0002] Video conferencing refers to the use of modern communication means to achieve the purpose of holding meetings across regions. With the development of technology, different users in different locations can often exchange information through video conferencing. However, in the existing video conferencing remote interaction system, participants can only participate in a specific meeting, and if they want to participate in different meetings, they need to reload other video conferencing systems. Moreover, the host of the video conferencing system can only understand the progress of the meeting when entering the corresponding meeting system. And the video conferencing system usually realizes authentication through devices, which cannot prevent unauthorized devices from accessing the communication network, resulting in a large number of illegal devices possibly participating in the meeting during the video conferencing, not only affecting the user experience, but also possibly causing losses to the video conferencing organizer.

[0003] For example, Chinese Patent Publication No. CN114615461A discloses a video conferencing remote interaction system with multiple meetings coexisting, including several meeting terminals, an identification module, a central control module, and a cloud database. The identification module includes face acquisition devices respectively arranged on corresponding interactive devices. The central control module is used to determine the actual permissions of the participants using the interactive device according to the initial permissions of the interactive device and the participation permissions of the participants using the interactive device, and unlock the corresponding interactive operations according to the actual permissions of the participants during the meeting.

[0004] For example, Chinese Patent Publication No. CN113132675A discloses a data management method and system for realizing interactive video conferencing. The present invention can determine and record the meeting resource configuration information of the interactive video conferencing scenario according to the video conferencing association information and the video conferencing participant category corresponding to the object identity information of each video conferencing participant obtained. The meeting resource configuration information is used to represent the interaction status and meeting resource allocation situation between different video conferencing participants, and the meeting resource configuration information is constantly changing.

[0005] When the prior art manages video conferencing, it tends to focus on how to authenticate the connection of video conferencing and complete video conferencing management based on this information authentication. However, this management method will ignore the association problems existing during the connection of the video conferencing itself and the problems existing in the meeting itself during the video conferencing, resulting in a reduction in the security and user experience of the meeting itself. Summary of the Invention

[0006] To solve the above technical problems, the technical solution adopted by the present invention is as follows: A video conferencing network security management method, comprising: S1, sending a meeting request to the participants of the video conference, and obtaining the meeting association information corresponding to the video conference participation. The meeting association information includes the link information and identity information of the video conference participants.

[0007] S2, performing identity authentication on the video conference participants according to the identity information and link information of the video conference participants, and determining the object category and meeting configuration situation of the video conference participants during the meeting interaction.

[0008] S3, performing identity authentication on the video conference participants according to the meeting configuration situation and object category to obtain identity authentication information; obtaining the video interaction information of the video conference participants after identity authentication, and determining the video association authentication score according to the video interaction information and identity authentication information.

[0009] S4, controlling the resource permissions of the video conference participants according to the video association authentication score, determining the speaking situation of the video conference participants in the corresponding meeting, and verifying the meeting association score of the video conference participants in the participation association.

[0010] S5, performing abnormal detection and analysis on the surrounding environment of the video conference participants, determining the abnormal conditions occurring in the video conference, and setting the abnormal association score.

[0011] S6, using the video association authentication score, meeting association score and abnormal association score as retrieval conditions to determine the comprehensive evaluation score of the current video conference.

[0012] The beneficial effects of the present invention are as follows: First, the present invention performs multi-level identity verification by combining link information and identity information to ensure that only legitimate users can join the meeting; introducing the comparison of initial device permissions enhances the dual verification of users and their devices, improving the security of authentication.

[0013] Second, the present invention dynamically adjusts the resource permission level of users according to the video association authentication score to ensure that each participant only has the necessary operation permissions; by analyzing the time interval, length and completion degree of speaking behaviors, accurately evaluating the interaction of users, and adjusting permissions accordingly to ensure the order of the meeting.

[0014] Third, the present invention conducts comprehensive monitoring from multiple dimensions such as video interruption, delay to background change to ensure that any abnormality can be captured in time; based on the threshold response rate and the lowest response permission, automatically select the most appropriate response measures, reduce manual intervention, and improve the response speed.

[0015] IV. By integrating the video correlation authentication score, the conference correlation score, and the anomaly correlation score into a comprehensive evaluation score, the present invention provides a comprehensive and quantitative evaluation criterion, improving the efficiency and quality of conference processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below with reference to the drawings and embodiments.

[0017] Figure 1 is a schematic flowchart of a video conference network security management method.

[0018] Figure 2 is a schematic flowchart of step S2 of a video conference network security management method.

[0019] Figure 3 is a schematic flowchart of step S3 of a video conference network security management method.

[0020] Figure 4 is a schematic flowchart of step S4 of a video conference network security management method.

[0021] Figure 5 is a schematic flowchart of step S5 of a video conference network security management method. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The embodiments of the present invention will be described in detail below. The following described embodiments are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention. For those not specified in the embodiments regarding specific technologies or conditions, they shall be carried out according to the technologies or conditions described in the literature in the field or according to the product specifications.

[0023] Refer to Figure 1 , a video conference network security management method, including: S1, sending a conference request to the participants of the video conference, and obtaining the conference correlation information corresponding to the participation in the video conference, where the conference correlation information includes the link information and identity information of the participants in the video conference.

[0024] S2, authenticating the identity of the participants in the video conference according to the identity information and link information of the participants in the video conference, and determining the object category and conference configuration situation of the participants in the video conference for interaction.

[0025] S3, authenticating the identity of the participants in the video conference according to the conference configuration situation and object category to obtain the identity authentication information; obtaining the video interaction information of the participants in the video conference after identity authentication, and determining the video correlation authentication score according to the video interaction information and the identity authentication information.

[0026] S4. Control the resource permissions of the video conference participants according to the video association authentication score, determine the speaking status of the video conference participants in the corresponding meeting, and verify the meeting association score of the video conference participants under the participation association.

[0027] S5. Perform anomaly detection and analysis on the surrounding environment of the video conference participants, determine the anomalies that occur in the video conference, and set an anomaly association score.

[0028] S6. Use the video association authentication score, meeting association score, and anomaly association score as retrieval conditions to determine the comprehensive evaluation score of the current video conference.

[0029] In step S1, the identity information represents the permissions of the current user to participate, and the specific identity corresponding to the corresponding permissions; the link information represents information such as the connection port through which the current video conference participant joins the meeting.

[0030] In step S1, first create a new meeting instance. After setting the basic parameters of the meeting such as time and theme, generate a meeting identifier, select the people who need to participate in the meeting, and set an access token for each video conference participant to verify their identity. After passing the preliminary identity verification, connect to the video conference to obtain information such as the images and voices of the video conference participants when participating in the video conference. At this time, the faces, pronunciations, and the environments where the people are located will be recognized to determine whether the current people are participating in the video conference normally. At the same time, whenever a new user joins, a new record will be generated to establish the association record of the current meeting, continuously update the association record, record the speaking status and actions of the people to determine whether the current video conference is proceeding in the expected form, and finally adjust the communication form of the video conference according to these association information to reduce the interaction delay that occurs in the meeting and the situation where the corresponding person cannot be accurately recognized during the interaction.

[0031] In an embodiment of the present invention, when performing identity authentication, the identity information of the video conference participants in each video conference will be obtained, and it will be judged what the object category of the current video conference participant is, so as to configure the current meeting.

[0032] As Figure 2 shown, the processing method of step S2 further includes: S21. Obtain the initial device permissions of the video conference participants. The device permissions include multiple levels. The first-level device permission indicates that the current video conference participant has the permission to initiate a meeting and manages the process of the video conference as the meeting organizer. The second-level device permission indicates that the current video conference participant has the permission to speak in the video conference as a participant in the video conference. The third-level device permission indicates the permission of the video conference participant as an observer of the video conference, and only has the permission to watch the current video conference.

[0033] S22. Compare the initial device permissions of the video conference participants with their identity information, read the user passwords and session keys of the current video conference participants, and classify the video conference participants according to the user passwords and session keys to obtain the object categories.

[0034] S23. Configure the video conference participants according to the similarity between the initial device permissions and the identity information to obtain the conference configuration. The similarity between the initial device permissions and the identity information can be calculated in the form of cosine similarity, which is obtained by converting the content of the initial device permissions and the identity information into vector form, so as to determine the position where the current video conference participants need to be configured in the video conference and how to set the corresponding conference.

[0035] In an embodiment of the present invention, step S3 mainly authenticates the identity of the video conference participants according to the conference configuration and the object categories, and obtains the video interaction information of the video conference participants after identity authentication. The video interaction information is the corresponding video data of the user during the video conference, and this video will include the face, environment of the participants, and the voice of the participants during the conference, etc.

[0036] For the identity authentication information, it includes: according to the conference configuration and the object categories, determine whether the current video conference participant is the entry object of the current video conference, and whether the permissions of the current video conference participant correspond to the current video conference. If they correspond to each other, regard the video conference participant as the entry object of the current video conference, and configure according to the permissions of the video conference participant, and regard the configured video conference participant as the identity authentication information.

[0037] For the video interaction information, it represents the video data of the video conference participants collected by the camera when the current user is in the video conference. This video data includes the face, background and voice of the current video conference participant, and calibrate these contents to determine the specific situation of the current person participating in the video conference and whether there are problems related to the background and voice; generally in a specific video conference, the face in the current video should meet the same situation when compared with the preset face, and at the same time, the background also needs to have specific restrictions, such as no background transfer or large changes, no movement of people, etc., and the voice also needs to correspond to the lip shape or other situations in the video to prevent situations such as voice interruption and delay.

[0038] Such as Figure 3As shown, when determining the video association authentication score based on video interaction information and identity authentication information, step S3 further includes: S31, using the meeting configuration and object category, obtaining the preset face image of the current video conference participant and the real-time face image in the video interaction information; the preset face image is the basic avatar representing the current meeting participants, and the real-time face needs to be embedded into this basic avatar to compare whether the combined face still corresponds to the current face. When the face difference is large, it indicates that the identity of the current participant is different from the actual entry information, and there may be a risk of information leakage.

[0039] S32, perform image embedding of the real-time face image on the preset face image, calculate the embedded preset face image and the real-time face image, and obtain the first similarity.

[0040] At this time, the method of performing image embedding of the real-time face image on the preset face image is to extract feature points from the real-time face image and the preset face image, set a hash value for these feature points, and then correspond the feature points of the real-time face image and the preset face image one by one. After that, calculate the average value of the hash values of the feature points as the hash value on the embedded preset face image. Finally, calculate the similarity between the hash value on the embedded preset face image and the hash value of the real-time face image, and output this similarity value as the first similarity; according to the calculated similarity value at this time, it can be known whether the real-time face image and the preset face image are roughly the same. If the similarity is low after embedding, it means that the current person participating in the video conference is not the person corresponding to this permission, and an alarm needs to be issued and the user needs to be controlled.

[0041] When calculating the first similarity, compare the hash values of the embedded preset face image and the real-time face image at the corresponding feature points. If the values of both are the same, return 1, otherwise return 0. Compare all the feature points, and take the average value of the values returned by the feature points as the first similarity at this time.

[0042] S33, compare the real-time face images at adjacent time points, calculate the image difference ratio of the real-time face image with respect to the scene light, combine the image difference ratios according to the length of the time points to obtain an image difference sequence, calculate the trend quantity of the values in the image difference sequence, and output the calculated trend quantity and trend direction as the second similarity.

[0043] For the second similarity, it is determined whether there are instabilities or other problems in the current video conference by comparing the real-time face images at adjacent time points. It not only considers the similarity of the images themselves but also adds an analysis of the change trend in the time series, which can better reflect the impact of scene light changes on face recognition.

[0044] The above image difference ratio can be expressed as follows: for each adjacent time point t, the Euclidean distance between the real-time face images at adjacent time points on the image frame is calculated as the image difference ratio.

[0045] ; where represents the image difference ratio at time point t, represents the image frame of the real-time face image at time point t, represents the image frame of the real-time face image at time point t + 1, represents the Euclidean distance metric between the image frame of the real-time face image at time point t + 1 and the image frame of the real-time face image at time point t. This Euclidean distance metric is the sum of the distance values between all pixel points in these two image frames; represents the norm of the image frame of the real-time face image at time point t, which is used to standardize the difference. At this time, the norm is the sum of the absolute values of all pixel points in the image frame.

[0046] After that, the trend quantity is expressed as: ; where represents the trend quantity, represents the length value of the time point, represents the average value of the time point, represents the average value of the image difference ratio, and the value range of t is from 1 to T.

[0047] The trend direction is set according to the magnitude of the trend quantity. For example, if the trend quantity is greater than 0, it means that the image difference increases over time, indicating that the scene light may become brighter or more complex; if the trend quantity is less than 0, it means that the image difference decreases over time, indicating that the scene light may become darker or more stable; if the trend quantity is approximately equal to 0, it means that there is no obvious change trend and the scene light remains relatively constant.

[0048] At this time, the calculation method of the second similarity is as follows: ; where represents the second similarity, represents the adjustment coefficient, which is used to adjust the conversion speed. For example, the adjustment coefficient can be set to values from 1 to 5, represents the exponential constant.

[0049] S34. According to the obtained first similarity and second similarity, evaluate the current video conference participants to obtain the video association authentication score.

[0050] The video association authentication score can be set as follows: obtain the average value, standard deviation, and covariance of the pixel points in the real-time face image and the preset face image, and calculate the video association authentication score.

[0051] ; wherein, represents the video association authentication score, represents the average value of pixel points in the real-time face image, represents the standard deviation of pixel points in the real-time face image. When calculating the standard deviation and average value of the real-time face image, the images within the image difference sequence will be used for calculation to obtain a relative average value and standard deviation; represents the covariance of pixel points between the real-time face image and the preset face image. This covariance will be calculated after the real-time face image and the preset face image are in one-to-one correspondence. For example, an image obtained by weighted averaging all the real-time face images in the image difference sequence is used to calculate with the preset face image to obtain the covariance value at this time; represents the average value of pixel points in the preset face image, represents the standard deviation of pixel points in the preset face image, and represent constants used to prevent the denominator from being zero. represents the first similarity. At this time, the first similarity and the second similarity are used as the adjustment parts of the pixel point values in the current real-time face image and the preset face image to adjust the comprehensive values of these numerical values under the corresponding similarities, so as to represent the association situation between the current video interaction information and the originally preset information. 、 、 represent weight coefficients, and the values of the weight coefficients are set to 0.3, 0.3, and 0.4 in sequence.

[0052] In an embodiment of the present invention, step S4 mainly controls the resource permissions of the participants in the video conference according to the video association authentication score, and at the same time verifies the corresponding speech situation in the video conference. The score obtained by combining these situations is used as the conference association score.

[0053] The conference association score is essentially to obtain the operation status information under these scores, and judge whether the participants in the video conference with the operation status information are speaking normally, as well as the information situation of the corresponding audience, so as to complete the recognition of the corresponding situations in the current conference.

[0054] Such as Figure 4 shown, the implementation manner of step S4 further includes: S41, performing permission mapping according to the video association authentication score to determine the resource permission level of the participants in the video conference. At this time, the resource permission level represents the content that the user can operate specifically, defining different levels of permissions such as the host, speaker, listener, etc., and clarifying the operable resources corresponding to each permission, such as speaking, screen sharing, file uploading, etc., and mapping the obtained video association authentication score to these contents.

[0055] S42. Determine the speaking behavior of the participants in the video conference corresponding to the resource permission level, extract the speaking time interval, speaking time length, and number of completed speeches for the speaking behavior, and determine the completion degree of the current speaking behavior. The completion degree of the speech is used to evaluate whether the current person has spoken normally and whether the speech is determined to be an effective behavior; at the same time, the speech also needs to be checked whether it conforms to the preset communication mode to determine whether the current permission can correspond.

[0056] S43. According to the obtained completion degree of the speech, determine the security policy of the completion degree of the speech relative to the video conference, and calculate the conference correlation score.

[0057] The completion degree of the speech is expressed as: ; where represents the completion degree of the speech, represents the speaking time interval of the i-th speech, represents the number of completed speeches, and the value range of i is from 1 to n; represents the speaking time length of the i-th speech, represents the number of effective speeches. Here, the number of effective speeches means that there are no network fluctuations, interruptions, or voice delays in the video conference during the current speaking process, and these normal speaking parts are regarded as effective speeches; represents the total preset speaking interval duration, which represents the time interval that normally exists during a normal speech. If there are corresponding network fluctuations and abnormalities in the current video conference, this preset time will be affected, resulting in problems with the finally calculated time; represents the adjustment coefficient, which is used to adjust the influence of the speaking interval. At this time, the adjustment coefficient is set to a value range of 0 to 1; represents the initial value of the speaking time length, represents the initial value of the speaking time interval; at this time, the initial value of the speaking length can be the average value of the corresponding situation in the historical data, and the speaking time interval is also set using the average value of the corresponding historical data. This completion degree of the speech tends to describe the process of the overall video conference under normal procedures. After completing all port or video certifications, whether the data transmission and reception capabilities of the current video conference can meet the corresponding speaking situation requirements while maintaining the corresponding security requirements, so as to improve the efficiency of the corresponding security policy.

[0058] According to the obtained completion degree of the speech, determine the security policy of the completion degree of the speech relative to the video conference, and calculate the conference correlation score.

[0059] In step S43, the implementation method of the security policy is as follows: when both the resource permission level and the speech completion degree are matched, the audio code stream of the current video conference is parsed to determine the hash message authentication value in the audio code stream; the hash message authentication value is an authentication code set by the audio code stream for the transmitted video interaction information. At this time, it is necessary to verify whether the current process and the transmitted data are the corresponding hash message authentication values, so as to determine whether the participants in the video conference are normal conference participants; according to the hash message authentication value of the audio code stream, verify whether the resource permission level of the participants in the video conference changes during the conversion of the audio code stream, and select a security policy according to the resource permission level and the speech completion degree. The method of selecting a security policy is to select the security policy corresponding to the maximum support degree and confidence degree according to the support degree and confidence degree of the resource permission level and the speech completion degree appearing in the historical data.

[0060] At this time, after obtaining the security policy, the conference association score will be comprehensively set according to the appearance probability of the security policy and the co-occurrence probability of the speech completion degree under the corresponding security policy to obtain the conference association score.

[0061] The conference association score is expressed as follows: obtain the appearance probability of the security policy, the appearance probability of the speech completion degree, and the co-occurrence probability of the security policy and the speech completion degree, and calculate to obtain the conference association score.

[0062] ; where represents the conference association score, represents the security policy, represents the appearance probability of the security policy, represents the appearance probability of the speech completion degree, represents the co-occurrence probability of the security policy and the speech completion degree, represents the standard value of the appearance probability of the speech completion degree; this standard value will be set according to the average value of the appearance probability of the speech completion degree in the historical data.

[0063] In an embodiment of the present invention, step S5 performs abnormal detection and analysis on the surrounding environment of the participants in the video conference to determine the abnormal conditions that occur in the video conference. The abnormal conditions can be set as video interruption, video delay, the voice not matching the lip movement at all, a large difference in the background of the participants, the behavior of the participants not matching the permissions, etc. These situations will be regarded as abnormal situations in the video conference. At this time, it is necessary to perform abnormal detection on the conference association and, according to the results of the abnormal detection, achieve the level of security management of the video conference; set the abnormal association score.

[0064] Such as Figure 5As shown, step S5 also includes the following processing methods when identifying abnormal situations: S51, obtain the security policies of the current video conference participants, configure the connection endpoints of the video conference participants based on the protocol communication mechanism of the security policies, transmit abnormal detection information according to the connection endpoints, and determine the abnormal situations occurring in the video conference according to the abnormal detection situations uploaded within a preset time period.

[0065] S52, according to the occurrence positions of the corresponding abnormal detection information at the corresponding connection endpoints, set the distribution positions of the corresponding abnormal detection information in the order of time points.

[0066] S53, compare and analyze the abnormal detection information with the security policies, judge the threshold response rate and the minimum response authority corresponding to the current video conference participants when the abnormal detection information appears, and calculate the abnormal association score according to the distribution probabilities of the threshold response rate and the minimum response authority. The threshold response rate represents the probability that the system immediately responds when a specific abnormality occurs; for example, when a video interruption is detected, the system should have an 80% probability of immediately responding.

[0067] The minimum response authority is the minimum level of response measures that the system needs to take when an abnormality is detected; for example, when a slight video delay occurs, the system will send a response warning message to indicate that there may be fluctuations in the current network, but if there are multiple background changes of a certain user during a normal video conference, the speaking authority may need to be restricted.

[0068] The method of comparing and analyzing the abnormal detection information with the security policies is to compare the specific content of the abnormal detection information with the content in the security policies, determine the specific abnormal situation occurring at this time, and identify the minimum response authority and the threshold response rate for this abnormal situation to determine whether the current abnormal situation needs to be processed.

[0069] Therefore, the calculation method of the abnormal association score is ; where represents the abnormal association score, represents the distribution probability of the threshold response rate, represents the distribution probability of the minimum response authority; represents the weight coefficient of the threshold response rate, represents the weight coefficient of the minimum response authority. The weight coefficients are set to 0.5 and 0.5 in the order of the threshold response rate and the minimum response authority.

[0070] In an embodiment of the present invention, the comprehensive evaluation score is expressed as: ; where represents the comprehensive evaluation score, represents the weight of the video association authentication score, Represents the standard value of the anomaly correlation score, Represents the weight of the anomaly correlation score, Represents the standard value of the meeting correlation score, Represents the weight of the meeting correlation score. The weights of the video correlation authentication score, the meeting correlation score, and the anomaly correlation score are set to 0.3, 0.4, and 0.3 in sequence; the standard value of the anomaly correlation score is represented as the standard deviation of the anomaly correlation score in historical data, and the standard value of the meeting correlation score is represented as the standard deviation of the meeting correlation score in historical data.

[0071] After obtaining the comprehensive evaluation score, according to this comprehensive evaluation score, the specific situation under the current video conference can be understood, and whether there are corresponding problems with the video conference corresponding to this situation, so as to ensure that the video conference can be carried out normally without being affected by external devices.

[0072] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention, and still be covered by the protection scope of the present invention.

Claims

1. A video conferencing network security management method, characterized in that, Including: S1. Send a meeting request to the participants of the video conference, and obtain the meeting association information corresponding to the participation in the video conference. The meeting association information includes the link information and identity information of the participants in the video conference; S2. Authenticate the participants in the video conference according to their identity information and link information, and determine the object category and meeting configuration for the participants in the video conference to interact in the meeting; S3. Authenticate the participants in the video conference according to the meeting configuration and object category to obtain the identity authentication information; obtain the video interaction information of the participants in the video conference after identity authentication, and determine the video association authentication score according to the video interaction information and identity authentication information; S4. Control the resource permissions of the participants in the video conference according to the video association authentication score, determine the speaking situation of the participants in the video conference under the corresponding meeting, and verify the meeting association score of the participants in the video conference in relation to the participation in the meeting; S5. Conduct anomaly detection and analysis on the surrounding environment of the participants in the video conference, determine the anomalies occurring in the video conference, and set the anomaly association score; S6. Use the video association authentication score, meeting association score, and anomaly association score as retrieval conditions to determine the comprehensive evaluation score of the current video conference; The video interaction information is the corresponding video data when the user participates in the video conference. This video data includes the face, background, and voice of the current participant in the video conference; The anomalies are set as video interruption, video delay, complete mismatch between voice and lip movement, significant difference in the background of the participants, and inconsistent behavior and permissions of the participants.

2. The video conferencing network security management method according to claim 1, characterized in that, Step S2 further includes: S21. Obtain the initial device permissions of the participants in the video conference; S22. Compare the initial device permissions of the participants in the video conference with their identity information, read the user password and session key of the current participant in the video conference, and classify the participants in the video conference according to the user password and session key to obtain the object category; S23. Configure the participants in the video conference according to the similarity between the initial device permissions and identity information to obtain the meeting configuration.

3. A video conferencing network security management method according to claim 1, characterized in that, Step S3 further includes: S31. Use the meeting configuration and object category to obtain the preset face image of the current participant in the video conference and the real-time face image in the video interaction information; S32. Embed the real-time face image into the preset face image, calculate the embedded preset face image and the real-time face image to obtain the first similarity; S33. Compare the real-time face images at adjacent time points, calculate the image difference ratio of the real-time face image with respect to the scene light, combine the image difference ratios according to the length of the time points to obtain an image difference sequence, calculate the trend quantity of the values in the image difference sequence, and output the calculated trend quantity and trend direction as the second similarity; S34. Evaluate the current participant in the video conference according to the obtained first similarity and second similarity to obtain the video association authentication score.

4. A method for video conferencing network security management according to claim 3, characterized in that, The video correlation authentication score is expressed as follows: obtain the average value, standard deviation, and covariance of the pixel points in the real-time face image and the preset face image, and calculate the video correlation authentication score; ； Among them, represents the video association authentication score, represents the average value of pixel points in the real-time face image, represents the standard deviation of pixel points in the real-time face image, represents the covariance of pixel points in the real-time face image and the preset face image; represents the average value of pixel points in the preset face image, represents the standard deviation of pixel points in the preset face image, and represents a constant, represents the first similarity, represents the second similarity, represents an exponential constant; , , represent weight coefficients.

5. A video conferencing network security management method according to claim 1, characterized in that, The implementation method of step S4 also includes: S41, perform permission mapping according to the video correlation authentication score to determine the resource permission level of the video conference participants; S42, determine the speaking behavior of the video conference participants corresponding to the resource permission level, extract the speaking time interval, speaking time length, and number of completed speeches for the speaking behavior, and determine the completion degree of the current speaking behavior; S43, according to the obtained completion degree of the speech, determine the security policy of the speech completion degree relative to the video conference, and calculate the conference correlation score.

6. A video conferencing network security management method according to claim 5, characterized in that, The completion degree of the speech is expressed as: ； Among them, represents the speech completion degree, represents the speech time interval of the i-th speech, represents the number of completed speeches, and the value range of i is from 1 to n; represents the speech time length of the i-th speech, represents the number of valid speeches; represents the total preset speech interval duration; represents the adjustment coefficient; represents the initial value of the speech time length, represents the initial value of the speech time interval.

7. A video conferencing network security management method according to claim 6, characterized in that The conference correlation score is expressed as follows: obtain the occurrence probability of the security policy, the occurrence probability of the speech completion degree, and the co-occurrence probability of the security policy and the speech completion degree, and calculate the conference correlation score; ； Among them, represents the conference association score, represents the security policy, represents the occurrence probability of the security policy, represents the occurrence probability of the speech completion degree, represents the co-occurrence probability of the security policy and the speech completion degree, represents the standard value of the occurrence probability of the speech completion degree.

8. A video conferencing network security management method according to claim 1, characterized in that Step S5 also includes the following processing methods: S51, obtain the security policy of the current video conference participants, configure the connection endpoints of the video conference participants based on the protocol communication mechanism of the security policy, transmit the abnormal detection information according to the connection endpoints, and determine the abnormal conditions that occur in the video conference according to the abnormal detection status uploaded within the preset time period; S52, according to the occurrence position of the abnormal detection information at the corresponding connection endpoint, set the distribution position of the abnormal detection information in the order of time points for the corresponding abnormal detection information; S53, compare and analyze the abnormal detection information with the security policy, judge the threshold response rate and the lowest response permission corresponding to the current video conference participants when the abnormal detection information appears, and calculate the abnormal correlation score according to the distribution probability of the threshold response rate and the lowest response permission.

9. The video conferencing network security management method according to claim 8, characterized in that The calculation method of the abnormal correlation score is: ； Among them, represents the abnormal association score, represents the distribution probability of the threshold response rate, represents the distribution probability of the lowest response permission; represents the weight coefficient of the threshold response rate, represents the weight coefficient of the lowest response permission.

10. A video conferencing network security management method according to claim 1, characterized in that, The comprehensive evaluation score is expressed as: ； Among them, represents the comprehensive evaluation score, represents the weight of the video association authentication score, represents the standard value of the abnormal association score, represents the weight of the abnormal association score, represents the standard value of the meeting association score, represents the weight of the meeting association score, represents the abnormal association score, represents the meeting association score, represents the video association authentication score.

Images