CN113422975A - Background live video auditing method - Google Patents
Background live video auditing method Download PDFInfo
- Publication number
- CN113422975A CN113422975A CN202110675304.8A CN202110675304A CN113422975A CN 113422975 A CN113422975 A CN 113422975A CN 202110675304 A CN202110675304 A CN 202110675304A CN 113422975 A CN113422975 A CN 113422975A
- Authority
- CN
- China
- Prior art keywords
- user
- video
- auditing
- room
- auditor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012550 audit Methods 0.000 claims abstract description 80
- 238000005516 engineering process Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 18
- 230000006399 behavior Effects 0.000 claims description 10
- 230000003203 everyday effect Effects 0.000 claims description 8
- 230000000903 blocking effect Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000007789 sealing Methods 0.000 claims description 3
- 238000012552 review Methods 0.000 abstract description 29
- 230000000694 effects Effects 0.000 abstract description 11
- 230000009471 action Effects 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 239000013598 vector Substances 0.000 description 5
- 230000000875 corresponding effect Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 241000209140 Triticum Species 0.000 description 2
- 235000021307 Triticum Nutrition 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/21—Server components or server architectures
- H04N21/218—Source of audio or video content, e.g. local disk arrays
- H04N21/2187—Live feed
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/20—Network architectures or network communication protocols for network security for managing network security; network security policies in general
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
- H04N21/2407—Monitoring of transmitted content, e.g. distribution time, number of downloads
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
- H04N21/44204—Monitoring of content usage, e.g. the number of times a movie has been viewed, copied or the amount which has been watched
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
- H04N21/44213—Monitoring of end-user related data
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Security & Cryptography (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Social Psychology (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention discloses a background live video auditing method.A user initiates microphone connection and sends video, audio and video screenshots to a cloud; machine auditing, detecting through a picture recognition technology, and actively handling violation; manual review, binding the video room with a reviewer, and further auditing by the reviewer; group leader checking, wherein the group leader checks the working condition of a checker; and checking the auditing result of the auditor in a specified time period by the system, and checking whether misjudgment and misauditing exist. The invention has the beneficial effects that: the invention combines machine audit and manual audit, on one hand, the machine audit improves the audit efficiency, avoids the untimely manual audit, and reduces the labor intensity of the manual audit; the manual auditing solves the problem that the machine audits the contents which are difficult to audit; the auditing effect of the auditing contents is improved; the group leader review mainly is to review the review result of the manual review, so as to prevent misjudgment of user actions and force the user to be offline due to subjective errors of the manual review.
Description
Technical Field
The invention relates to the technical field of software management, in particular to a background live video auditing method.
Background
With the rapid development of internet technology, live webcasting is also more and more widely applied to daily life of people, and due to the diversity of live content, topics such as terrorism, violence, pornography or political sensitivity may be involved, so that a background needs to judge whether live content violates rules or not when a user directly broadcasts the live webcasting content, so as to prevent illegal content from being rapidly spread on the internet.
In view of the above problems, the prior art generally adopts machine review or manual review; as follows
1. The chinese patent discloses a method and related device for identifying an illegal picture (application number: CN201811157555.1), which are used to improve the efficiency of yellow-identification service. The method provided by the embodiment of the invention comprises the following steps: the live broadcast platform server traverses all live broadcast rooms to obtain live broadcast information of each live broadcast room; the live broadcast platform server intercepts key frame pictures from live broadcast videos of a live broadcast room to be audited according to a video stream address of the live broadcast room to be audited; the live broadcast platform server sends a request message to the yellow-signature server; the live broadcast platform server receives an identification result sent by the yellow identification server, wherein the identification result comprises a score of the key frame picture, and the score is positively correlated with the irregularity; the live broadcast platform server judges whether the score exceeds a preset value or not; and if the number of the key frame pictures exceeds the preset number, the live broadcast platform server confirms that the key frame pictures violate rules and presets a live broadcast room corresponding to the key frame pictures.
2. The chinese patent discloses a method, an apparatus, a server and a storage medium (application number: CN201811628763.5) for live broadcast auditing, the method comprising: acquiring video frames or audio with preset duration in a live video according to a preset time interval; respectively inputting each video frame or audio frequency into different types of convolutional neural network models to obtain corresponding characteristic vectors; combining the feature vectors of the video frames or the feature vectors of the audio to obtain the feature vectors of the live video; and determining the current auditing category of the live video according to the feature vector of the live video. The technical scheme provided by the embodiment of the invention realizes accurate audit of the current live broadcast video, solves the problem that misjudgment and missed judgment are easily caused by determining the audit type according to the detection result of the current single video frame in the existing audit method, and improves the comprehensiveness and accuracy of the live broadcast audit.
Although the prior art can perform live content auditing to a certain extent, the following problems also exist:
1. in the auditing method in the prior art, only one auditing mode is adopted, and manual auditing or machine auditing, such as manual auditing, is adopted, so that the labor intensity of workers is increased, and the auditing efficiency is reduced; if the verification is only machine verification, because many content machines cannot be verified, the verification effect is affected, partial identification results are inaccurate, user actions may be misjudged, and users are forced to be off-line;
2. in the prior art, no supervision mechanism is provided for manual review, and because the manual review inevitably has certain subjective factors, the situation of error break and error break also exists.
Therefore, it is necessary to provide a method for auditing a background live video for the above problems.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a background live video auditing method to solve the problems.
A background live video auditing method comprises the following steps:
s1, the user initiates microphone connection and sends the video, the audio and the video screenshot to the cloud;
s2, machine auditing, detecting through a picture recognition technology, and actively handling violation;
s3, manual auditing, namely binding the video room with an auditor, and further auditing by the auditor;
s4, auditing by the group leader, and checking the working condition of auditors by the group leader; checking the auditing result of the auditor in a specified time period through the system, and checking whether misjudgment and misauditing exist;
the cloud in step S1 stores the video screenshot of the user, where the screenshot rule of the video screenshot is: and (3) capturing X video screenshots in 1 minute when the user just enters the room, and capturing Y video screenshots every 1 minute after 1 minute, and sending the video screenshots to the system.
Wherein the cloud end in step S1 is the server end.
In step S3, a dedicated auditor is allocated to the video connection of the user before manual auditing, the system records the auditor currently on duty, and then allocates the video room to the auditor with the fewest current rooms, and the auditor starts auditing.
The filter condition of the auditor is that when the user turns off the microphone and the video, or the user finishes the video connection, the room is automatically removed from the auditor list.
Wherein the audit work of the auditor comprises the following steps:
and (3) auditing the pictures of the real-time audio and video: the video and the audio on the cloud are pulled through the webpage and displayed on the webpage, at the moment, the real-time picture is synchronously displayed, at the moment, an auditor continuously watches one video room for N minutes, and if no violation condition exists, the next video room is started;
and (3) auditing the speech of the user released in the chat room: the user can publish the speech of the chat room at the live APP,
the cloud end can synchronize the characters and pictures issued by the users to the webpage, and auditors can check whether illegal speech exists or not, and if illegal speech exists, illegal speech processing is carried out;
and (3) auditing the video screenshot of the user: the system records historical video screenshot records of a user, the video screenshots are sorted according to violation levels, the violation is in the front, the suspected violation is in the back, and the suspected violation is in the back.
Wherein the auditing form of manual auditing is as follows: the auditing adopts a circular circulation mode to audit the rooms, after the auditing of one room is completed, the room is placed at the rearmost and then enters the next room, after the rooms are alternately audited, the room returns to the foremost room, and the rooms with the cameras and the microphones closed or closed are automatically filtered in midway.
Wherein the auditing operation of manual auditing comprises the following steps:
room dispersion: when a plurality of users exist in the room, the situation that the processing is not timely can occur, at the moment, the operation of the dismissed room can be used, all the users in the room are kicked off the line, then the record of the room is destroyed, and the operation of an auditor is recorded, so that whether the operation of the auditor is reasonable or not is checked.
Kicking out of the room: when the user in the room has improper speech or improper behavior, if the user has improper behavior, illegal evidence of the user is selected, the illegal evidence is recorded in the cloud and stored permanently, live APP push is sent to the user to put the user out of the room, if the user is a user marked with a white list, the user is removed from the white list user, and in a certain time, other video connection rooms cannot be added.
White list: the system calculates the learning time, learning days and violation conditions of the user in the whole live APP every day, when the condition is that the user has no violation in ten days and is female, the user is added into a white list, the screenshot quantity of the user can be reduced after the user is added into the white list, and the auditing efficiency is improved
Sealing and forbidding: when the auditor judges that the violation behavior of the user is serious, the blocking operation is used, the blocking operation can directly and forcibly disconnect the login of the account number of the user, and then the account number can not be logged in to the live APP, so that the operation of the auditor can be recorded, and the audit is facilitated in the future;
message revocation: the method is used for processing the violation messages of the chat room below the video microphone, the user can watch the real-time video of the user in the chat room, the user can freely release the speech at the moment, and the user can release the bad speech or advertisements in the chat room through withdrawing operation.
Wherein the auditing process of group leader auditing is as follows: the group leader checks the yesterday audit result every day through a special page for carrying out statistical processing on illegal operation, and checks the processing record in the time interval through time.
Wherein the treatment record includes evidence of treatment, reason for treatment, time of treatment, and person of treatment.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention combines machine audit and manual audit, on one hand, the machine audit improves the audit efficiency, avoids the untimely manual audit, and reduces the labor intensity of the manual audit; in addition, manual examination solves the problem that the machine examines the contents which are difficult to examine; the auditing effect of the auditing contents is improved;
2. the group leader audit is added, and the group leader audit mainly audits the audit result of the manual audit, so that the phenomenon that the user action is misjudged and the user is forced to be off-line due to the subjective error of the manual audit is prevented.
3. The method and the device ensure that the live APP conforms to the specification of network safety information, greatly reduce the violation risk of the live APP, create an environment with better experience for users, improve the efficiency of manual auditing, effectively reduce the risk, and maintain the stability of the environment in the live APP through a series of measures.
Drawings
FIG. 1 is a flow diagram of a background live video review method of the present invention;
fig. 2 is a schematic block diagram of a background live video auditing method of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways as defined and covered by the claims.
As shown in fig. 1 in combination with fig. 2, a method for auditing a background live video includes the following steps:
s1, the user initiates microphone connection and sends the video, the audio and the video screenshot to the cloud;
s2, machine auditing, detecting the video screenshot through a picture recognition technology, and actively handling violation;
s3, manual review, namely binding the video room before the auditors, and further auditing the video and the audio by the auditors;
s4, auditing by the group leader, and checking the working condition of auditors by the group leader; checking the auditing result of the auditor in a specified time period through the system, and checking whether misjudgment and misauditing exist;
the version of the system is iterated, the system can update and issue the iteration according to the function of the live APP, the opinions of auditors can be collected, the system is optimized periodically, and continuous optimization is conducted according to the auditing speed, the operation convenience, the display effect on a webpage and the like.
The image recognition technology adopts a neural network recognition algorithm or an image segmentation recognition algorithm for recognition, and the neural network recognition algorithm or the image segmentation recognition algorithm are all the existing common technologies.
The user at step S1 starts a connection (initiates live video broadcast, and can allow others to join in, live video broadcast in real time) at the live broadcast APP, and synchronously pushes video frames and sound to the cloud (server side) through conversion, the cloud stores the video screenshots of the user, the specific video screenshot rules are that the user captures X video screenshots just before entering the room for 1 minute, and after 1 minute, captures Y video screenshots every 1 minute are sent to the system.
The method comprises the steps of acquiring live video, capturing the live video, wherein X is more than 3, Y is more than 1, and the number of the shot is specifically determined, wherein the shot is performed according to the activity of the live video, the higher the activity is, the more the shot is per minute, the lower the activity is, the fewer the shots are per minute, generally X is 12, and Y is 3.
The video screenshot of machine review is processed as follows: the video screenshot processes a plurality of scenes including a plurality of scenes such as concerning yellow, concerning politics, violence and terrorism, bloody smell, advertisements and two-dimensional codes; the identification result is classified into three grades, normal, suspected and illegal, and the advertisement and the two-dimensional code are displayed; and if the other scenes have suspected or more results, the user is forced to be kicked off, the violation reason is sent, the auditor is informed by a nail notice sent in a group manner, and the room is used for confirming whether the user needs to add the penalty again or not. The method can avoid the situation that manual review is not timely, and when suspected violation or violation occurs to the user, the user is firstly processed by the system in advance, so that the violation influence is avoided being enlarged.
The cloud in step S1 is a server side, and is used to store data information such as video, audio, and video screenshots.
In step S3, before manual review, a dedicated auditor is assigned to the video connection of the user, the system records the auditor currently on duty, and then assigns the video room to the auditor with the fewest number of current rooms, and the auditor starts to perform the review.
The filtering condition (removing condition) for manual review is that the room is automatically removed from the list of review workers after the user turns off the microphone and video or finishes the video connection.
Wherein the audit work of the auditors of the manual audit comprises the following steps:
auditing the pictures of real-time audio and video, pulling the video and audio on the cloud through a webpage, displaying the video and audio on the webpage, and synchronously displaying the real-time pictures, wherein an auditor continuously watches a video room for N minutes (N is more than 1 minute, the specific continuous watching time is comprehensively determined according to the activity of the video room and the previous bad record, the activity is high, the watching auditing time needs to be improved if the bad record exists, if the activity is low and the bad record does not exist, the watching auditing time is reduced, generally about 2 minutes, and the next video room can be entered if the violation condition does not exist;
the method comprises the following steps that the speech of a user issued in a chat room is audited, the user can issue the speech of the chat room in a live APP, the cloud end synchronizes characters and pictures issued by the user to a webpage, auditors can check whether illegal speech exists or not, and if illegal speech exists, illegal speech processing is carried out;
and auditing screenshots of the user, recording historical video screenshot records of the user by the system, sorting the video screenshots according to violation levels, wherein the violation is in the front, the suspected violation is in the back and the suspected violation is in the back.
Wherein the auditing form of manual auditing is as follows: the auditing adopts a circular circulation mode to audit the rooms, after the auditing of one room is completed, the room is placed at the rearmost and then enters the next room, after the rooms are alternately audited, the room returns to the foremost room, and the rooms with the cameras and the microphones closed or closed are automatically filtered in midway.
Wherein the auditing operation of manual auditing comprises the following steps:
(1) room dispersion: when a plurality of users exist in the room, the situation that the processing is not timely can occur, at the moment, the operation of the dismissed room can be used, all the users in the room are kicked off the line, then the record of the room is destroyed, and the operation of an auditor is recorded, so that whether the operation of the auditor is reasonable or not is checked.
(2) Kicking out of the room: when the user in the room has improper speech or improper behavior, if the user has improper behavior, illegal evidence of the user is selected, the illegal evidence is recorded in the cloud and stored permanently, live APP push is sent to the user to put the user out of the room, if the user is a user marked with a white list, the user is removed from the white list user, and in a certain time, other video connection rooms cannot be added.
(3) White list: the system calculates the learning time, the learning days and the violation condition of the user in the live APP every day, and adds the user into a white list when the condition is that the user has no violation and is female within ten days, and reduces the screenshot quantity of the user after adding the user into the white list, thereby improving the auditing efficiency;
(4) sealing and forbidding: when the auditor judges that the violation behavior of the user is very serious, the blocking operation can be used, the blocking operation can directly and forcibly disconnect the account number of the user from login, then the user can not log in the APP, the behavior can record the operation of the auditor, and the audit is convenient to examine in the future
(5) Message revocation: the method is used for processing the violation messages of the chat room below the video microphone, the user can watch the real-time video of the user in the chat room, the user can freely release the speech at the moment, and the user can release the bad speech or advertisements in the chat room through withdrawing operation.
The auditing process of the group leader is as follows: the group leader checks the yesterday audit result every day through a special page for carrying out statistical processing on illegal operations (a functional page is arranged in the system and used for recording the manual audit record, the specific record is divided according to time, the audit condition and result of the manual audit are recorded in each period of time, the group leader checks whether the audit results of some auditors are in misoperation or not according to corresponding time), and the group leader checks the processing record in a time interval through time.
Wherein the treatment record includes evidence of treatment, reason for treatment, time of treatment, and person of treatment.
Wherein the system includes: the cloud is used for storing, video, audio and video screenshots;
the machine audit is used for machine audit, a video screenshot sent by the cloud is automatically identified by adopting a picture recognition technology, and a user with risk is automatically processed;
and manual review, which is used for the auditors to pull the information of the videos and the microphones to the system and to review in turn.
The manual spot check module is used for the group leader to check the working condition of the spot check auditors
Compared with the prior art, the invention has the beneficial effects that:
1. according to the invention, machine audit and manual audit are combined, on one hand, machine audit (some contents which can be audited by using a picture recognition technology are all fed back to the machine audit), the audit efficiency is improved, the untimely manual audit is avoided, and the labor intensity of the manual audit is reduced; in addition, manual examination solves the problem that the machine examines the contents which are difficult to examine; the auditing effect of the auditing contents is improved;
2. the group leader audit is added, and the group leader audit mainly audits the audit result of the manual audit, so that the phenomenon that the user action is misjudged and the user is forced to be off-line due to the subjective error of the manual audit is prevented.
3. The method and the device ensure that the live APP conforms to the specification of network safety information, greatly reduce the violation risk of the live APP, create an environment with better experience for users, improve the efficiency of manual auditing, effectively reduce the risk, and maintain the stability of the environment in the live APP through a series of measures.
The machine auditing is mainly to audit the video screenshots of the live videos of the users; the manual review mainly comprises the review of the live video and audio of the user, and the group leader review mainly comprises the review and recheck of the review result of the manual review.
The auditing process has the following conditions:
firstly, when the machine is checked, the machine checking module checks that the suspected violation condition occurs in the video picture, and the system carries out offline processing on the user;
secondly, when the machine is checked, the machine checking module does not check that the video picture is suspected to be illegal, then manual checking is carried out, when the live video and audio of the user are found to be suspected to be illegal through manual checking, checking operation (room dismissal, room kicking, white list blocking and message withdrawing) is carried out on the user, then group leader rechecks the checking result of the manual checking, and if the manual checking is not wrong, the checking result of the manual checking is maintained; and if the manual review has misjudgment, canceling the review operation of the manual review.
And thirdly, when the machine is checked, the machine checking module does not check that the video pictures are suspected to be illegal, then manual checking is carried out, the fact that the video pictures are illegal is not found through manual checking, the system does not interfere live broadcasting of the user, the learning time, the learning days and the illegal conditions of the user in the whole APP are calculated every day, when the fact that the user is illegal within ten days is achieved, the user is added into a white list, screen capture amount of the user is reduced after the user is added into the white list, and checking efficiency is improved.
The following is a detailed description of the three cases:
example 1:
the method comprises the steps that a user initiates microphone connection in a live APP, a system sends video, audio and video screenshots to a server side, the cloud stores the video screenshots of the user, the screenshots of the user are 12 in 1 minute just before the user enters a room, the screenshots are changed into three screenshots in 1 minute after one minute, and the three screenshots are sent to the system; a machine auditing module of the system detects a video section through a picture identification technology, if a suspected situation of one of yellow, administration, riot or bloody smell is detected, the system forces a user to be kicked off the line and sends a violation reason, auditors are informed by a nail notice in a mass mode, and the auditors go to a room to confirm whether the user needs to add penalty again.
Example 2: the method comprises the steps that a user initiates wheat connection in a live broadcast APP, when the machine audit is finished, a suspected violation condition of a video picture is not audited by a machine audit module, then the system enters a manual audit, videos and audios are sent to a server side by the system, when the system enters the manual audit, a video room is bound with an auditor by the manual audit module of the system (a special auditor is distributed to a video connection line of the user, the system records the auditor who is currently on duty, then the video room is distributed to the auditor with the fewest rooms at present), then the auditor audits the videos and audios in the live broadcast APP, audits a picture of real-time audio and video, audits a talk audit issued by the user in a chat room and audits a screenshot of the user respectively, and in the manual audit, if the violation condition occurs, according to the violation degree of the audit and related violation content, the method comprises the following steps of performing one or more than two kinds of auditing operations (room dismissal, room kicking, prohibition and message withdrawal), checking the yesterday auditing result by a group leader on a page of special statistical processing illegal operations the next day, checking the processing record in a time interval by time, and checking the working condition of a group member by the auditing group leader through the information, wherein if the auditing member is mistakenly broken, the auditing operation (room dismissal, room kicking, white list, prohibition and message withdrawal) of the auditing member on a user is cancelled, and the right for use is recovered (if the auditing member is the user dismisses the room before, the room establishment is recovered again, and if the auditing member is the user who is prohibited before, the prohibited and the user are prohibited).
Example 3: the method comprises the steps that a user initiates wheat connection in a live broadcast APP, when the machine audit is finished, a suspected violation condition of a video picture is not audited by a machine audit module, then the system enters a manual audit, videos and audios are sent to a server end by the system, when the system enters the manual audit, a video room is bound with an auditor by the manual audit module of the system (a special auditor is distributed to a video connection line of the user, the system records the auditor who is currently on duty, then the video room is distributed to the auditor with the minimum number of rooms), then the auditor audits the videos and audios in the live broadcast APP, audits a picture of real-time audio and video, audits a talk audit issued by the user in a chat room and audits a screenshot of the user, in the manual audit, the video and the audios of the user are not found to have the violation condition, and the system does not interfere with the live broadcast of the user, the system calculates the learning time, the learning days and the violation condition of the user in the live APP every day, and adds the user into a white list when the violation condition does not exist in ten days, so that the screenshot quantity of the user can be reduced after the user is added into the white list, and the auditing efficiency is improved.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A background live video auditing method is characterized in that: the method comprises the following steps:
s1, the user initiates live broadcast microphone connection through the live broadcast APP, and sends video, audio and video screenshots to the cloud;
s2, machine auditing, detecting the video screenshot through a picture recognition technology, and actively handling violation;
s3, manual auditing, namely binding the video room with an auditor, and further auditing the video and the audio by the auditor;
and S4, auditing by the group leader, wherein the group leader inspects the working condition of the auditors, and checks the auditing result of the auditors by the system in a specified time period to check whether misjudgment and misauditing exist.
2. A background live video auditing method according to claim 1, characterized by: wherein the screenshot rule of the video screenshot is as follows: the user has just entered the video room for 1 minute and has taken X shots, and after 1 minute, has taken Y shots every 1 minute.
3. A background live video auditing method according to claim 1, characterized by: wherein the cloud end in step S1 is the server end.
4. A background live video auditing method according to claim 1, characterized by: in step S3, a dedicated auditor is allocated to the video connection of the user before manual auditing, the system records the auditor currently on duty, and then allocates the video room to the auditor with the fewest current rooms, and the auditor starts auditing.
5. A background live video auditing method according to claim 4, characterized by: the filter condition of the auditor is that when the user turns off the microphone and the video, or the user finishes the video connection, the video room is automatically removed from the auditor list.
6. A background live video auditing method according to claim 4, characterized by: wherein the audit work of the auditor comprises the following steps:
and (3) auditing the pictures of the real-time audio and video: the video and the audio on the cloud are pulled through the webpage, displayed on the webpage, and meanwhile, the real-time pictures of the video rooms are synchronously displayed, an auditor continuously watches one room for N minutes, and if no violation condition exists, the next video room is entered;
and (3) auditing the speech of the user released in the chat room: the user can release the speech of the chat room in the live broadcast APP, the cloud synchronizes the characters and pictures released by the user to the webpage, an auditor can check whether illegal speech exists or not, and if the illegal speech exists, the auditor carries out processing;
and (3) auditing the video screenshot of the user: the system records historical video screenshot records of a user, the video screenshots are sorted according to violation levels, the violation is in the front, the suspected violation is in the back, and the suspected violation is in the back.
7. A background live video auditing method according to claim 1, characterized by: wherein the auditing form of manual auditing is as follows: the auditing adopts a circular circulation mode to audit the rooms, after the auditing of one room is completed, the room is placed at the rearmost and then enters the next room, after the rooms are alternately audited, the room returns to the foremost room, and the rooms with the cameras and the microphones closed or closed are automatically filtered in midway.
8. A background live video auditing method according to claim 1, characterized by: wherein the auditing operation of manual auditing comprises the following steps:
room dispersion: when a plurality of users exist in a room, the situation of untimely processing may occur, at this time, the operation of the dismissal room can be used, all the users in the room are kicked off the line, then the record of the room is destroyed, and the operation of an auditor is recorded, so that whether the operation of the auditor is reasonable or not is checked;
kicking out of the room: when the user in the room has improper speech or improper behavior, if the user is improper behavior, illegal evidence of the user is selected, the illegal evidence is recorded in a cloud end for permanent storage, live broadcast APP push is sent to the user to put the user out of the room, if the user is a user marked with a white list, the user is removed from the white list user, and in a certain time, other video connection rooms cannot be added;
white list: the system calculates the learning time, the learning days and the violation condition of the user in the whole APP every day, and when the violation condition does not exist in ten days, the user is added into a white list, the screenshot quantity of the user is reduced after the user is added into the white list, and the auditing efficiency is improved;
sealing and forbidding: when the auditor judges that the violation behavior of the user is serious, the blocking operation is used, the blocking operation can directly and forcibly disconnect the login of the account number of the user, and then the account number can not be logged in to the live APP, so that the operation of the auditor can be recorded, and the audit is facilitated in the future;
message revocation: the method is used for processing the violation messages of the chat room below the video microphone, the user can watch the real-time video of the user in the chat room, the user can freely publish the speech at the moment, and the contents of bad speech or advertisements published in the chat room by the user can be withdrawn through withdrawal operation.
9. A background live video auditing method according to claim 1, characterized by: wherein the auditing process of group leader auditing is as follows: the group leader checks the yesterday audit result every day through a special page for carrying out statistical processing on illegal operation, and checks the processing record in the time interval through time.
10. A background live video auditing method according to claim 9, characterized by: wherein the treatment record includes evidence of treatment, reason for treatment, time of treatment, and person of treatment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110675304.8A CN113422975A (en) | 2021-06-17 | 2021-06-17 | Background live video auditing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110675304.8A CN113422975A (en) | 2021-06-17 | 2021-06-17 | Background live video auditing method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113422975A true CN113422975A (en) | 2021-09-21 |
Family
ID=77788987
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110675304.8A Pending CN113422975A (en) | 2021-06-17 | 2021-06-17 | Background live video auditing method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113422975A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114125494A (en) * | 2021-09-29 | 2022-03-01 | 阿里巴巴(中国)有限公司 | Content auditing auxiliary processing method and device and electronic equipment |
CN114205645A (en) * | 2021-12-10 | 2022-03-18 | 北京凯视达信息技术有限公司 | Distributed video content auditing method and device |
CN114911557A (en) * | 2022-04-28 | 2022-08-16 | 北京字跳网络技术有限公司 | Information processing method, device, electronic equipment and storage medium |
CN118096093A (en) * | 2024-04-29 | 2024-05-28 | 浙江实在智能科技有限公司 | Novel media platform audio and text content early warning method and system based on AI and RPA |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106412089A (en) * | 2016-10-25 | 2017-02-15 | 华中科技大学 | Bullet screen message processing method with auditing mechanism and keyword filtering |
CN109766807A (en) * | 2018-12-28 | 2019-05-17 | 广州华多网络科技有限公司 | Machine audits processing method, device, electronic equipment and storage medium |
CN109803152A (en) * | 2018-12-28 | 2019-05-24 | 广州华多网络科技有限公司 | Violation checking method, device, electronic equipment and storage medium |
-
2021
- 2021-06-17 CN CN202110675304.8A patent/CN113422975A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106412089A (en) * | 2016-10-25 | 2017-02-15 | 华中科技大学 | Bullet screen message processing method with auditing mechanism and keyword filtering |
CN109766807A (en) * | 2018-12-28 | 2019-05-17 | 广州华多网络科技有限公司 | Machine audits processing method, device, electronic equipment and storage medium |
CN109803152A (en) * | 2018-12-28 | 2019-05-24 | 广州华多网络科技有限公司 | Violation checking method, device, electronic equipment and storage medium |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114125494A (en) * | 2021-09-29 | 2022-03-01 | 阿里巴巴(中国)有限公司 | Content auditing auxiliary processing method and device and electronic equipment |
CN114205645A (en) * | 2021-12-10 | 2022-03-18 | 北京凯视达信息技术有限公司 | Distributed video content auditing method and device |
CN114911557A (en) * | 2022-04-28 | 2022-08-16 | 北京字跳网络技术有限公司 | Information processing method, device, electronic equipment and storage medium |
CN114911557B (en) * | 2022-04-28 | 2024-02-02 | 北京字跳网络技术有限公司 | Information processing method, apparatus, electronic device and storage medium |
CN118096093A (en) * | 2024-04-29 | 2024-05-28 | 浙江实在智能科技有限公司 | Novel media platform audio and text content early warning method and system based on AI and RPA |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113422975A (en) | Background live video auditing method | |
CN109377293B (en) | Advertisement putting effect monitoring method and system | |
CN110708565B (en) | Live broadcast interaction method and device, server and machine-readable storage medium | |
CN108449609B (en) | Live broadcast room event identification method and device, electronic equipment and machine readable medium | |
CN105872773A (en) | Video broadcast monitoring method and device | |
CN105808700A (en) | Game information recommendation method, and server | |
CN107016298B (en) | Webpage tampering monitoring method and device | |
CN105323531A (en) | Video conference hot spot scene detection method and device | |
CN105824805B (en) | Identification method and device | |
CN110047513B (en) | Video monitoring method and device, electronic equipment and storage medium | |
CN105718861A (en) | Method and device for identifying video streaming data category | |
CN111090813A (en) | Content processing method and device and computer readable storage medium | |
CN114245205B (en) | Video data processing method and system based on digital asset management | |
CN111586432B (en) | Method and device for determining air-broadcast live broadcast room, server and storage medium | |
CN107333101B (en) | Court trial personnel behavior patrol method and system based on court announcements | |
CN114786038A (en) | Low-custom live broadcast behavior monitoring method based on deep learning | |
CN110602483B (en) | Video fault determination method, device and computer readable storage medium | |
CN112183406A (en) | Video blockage detection method and detection system | |
CN108965492B (en) | Bullet screen link quality statistical method, storage medium, electronic equipment and system | |
CN113515670A (en) | Method, device and storage medium for identifying state of movie and television resource | |
CN113992984B (en) | Elevator advertisement monitoring and broadcasting method | |
CN106792227A (en) | A kind of live middle interactive method and device | |
US12010358B2 (en) | Live streaming moderating method and apparatus, server, and storage medium | |
CN110321517B (en) | Method and device for detecting playing proportion of browsing resources and readable storage medium | |
CN115955577A (en) | Comment processing method and device, computing device and computer storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |