[go: up one dir, main page]

CN107517391B - Method and equipment for identifying abnormal live broadcast information in video live broadcast - Google Patents

Method and equipment for identifying abnormal live broadcast information in video live broadcast Download PDF

Info

Publication number
CN107517391B
CN107517391B CN201610430465.XA CN201610430465A CN107517391B CN 107517391 B CN107517391 B CN 107517391B CN 201610430465 A CN201610430465 A CN 201610430465A CN 107517391 B CN107517391 B CN 107517391B
Authority
CN
China
Prior art keywords
live
video
live broadcast
identification
mode
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.)
Active
Application number
CN201610430465.XA
Other languages
Chinese (zh)
Other versions
CN107517391A (en
Inventor
宋朝阳
闵庆欢
向西西
祁海
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to CN201610430465.XA priority Critical patent/CN107517391B/en
Publication of CN107517391A publication Critical patent/CN107517391A/en
Application granted granted Critical
Publication of CN107517391B publication Critical patent/CN107517391B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/232—Content retrieval operation locally within server, e.g. reading video streams from disk arrays
    • 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
    • H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
    • H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
    • 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
    • H04N21/238—Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
    • H04N21/2385—Channel allocation; Bandwidth allocation
    • 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]
    • H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client 
    • H04N21/61—Network physical structure; Signal processing
    • 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]
    • H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client 
    • H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
    • H04N21/633—Control signals issued by server directed to the network components or client
    • 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]
    • H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client 
    • H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
    • H04N21/637—Control signals issued by the client directed to the server or network components
    • H04N21/6375—Control signals issued by the client directed to the server or network components for requesting retransmission, e.g. of data packets lost or corrupted during transmission from server
    • 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]
    • H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client 
    • H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
    • H04N21/647—Control signaling between network components and server or clients; Network processes for video distribution between server and clients, e.g. controlling the quality of the video stream, by dropping packets, protecting content from unauthorised alteration within the network, monitoring of network load, bridging between two different networks, e.g. between IP and wireless
    • 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]
    • H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client 
    • H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
    • H04N21/647—Control signaling between network components and server or clients; Network processes for video distribution between server and clients, e.g. controlling the quality of the video stream, by dropping packets, protecting content from unauthorised alteration within the network, monitoring of network load, bridging between two different networks, e.g. between IP and wireless
    • H04N21/64723—Monitoring of network processes or resources, e.g. monitoring of network load
    • H04N21/6473—Monitoring network processes errors
    • 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]
    • H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client 
    • H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
    • H04N21/647—Control signaling between network components and server or clients; Network processes for video distribution between server and clients, e.g. controlling the quality of the video stream, by dropping packets, protecting content from unauthorised alteration within the network, monitoring of network load, bridging between two different networks, e.g. between IP and wireless
    • H04N21/64784—Data processing by the network
    • H04N21/64792—Controlling the complexity of the content stream, e.g. by dropping packets

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computer Security & Cryptography (AREA)
  • Databases & Information Systems (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • Image Analysis (AREA)

Abstract

The method comprises the steps of detecting whether the video live broadcast in a first identification mode meets a trigger condition for switching to a second identification mode; when the triggering condition is met, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode; and determining whether the live video contains abnormal live broadcast information or not according to the live broadcast image frame, so that switching of different identification modes of the live video is realized, and meanwhile, manual participation is reduced, thereby reducing the cost of manual participation and effectively improving the real-time performance and the lasting efficiency of identifying the abnormal live broadcast information in the live video.

Description

Method and equipment for identifying abnormal live broadcast information in video live broadcast
Technical Field
The application relates to the field of computers, in particular to a technology for identifying abnormal live broadcast information in video live broadcast.
Background
With the rapid development of internet technology and multi-screen intelligent terminals, the video live broadcast industry is rapidly developed, and people can contact the content of the video live broadcast no matter whether the television, the PC or the mobile phone. At the back of the rapid development of the video live broadcast industry, live broadcast contents which challenge law and social ethical bottom lines, such as 'live broadcast dolls' and 'live broadcast coasters' appear occasionally, so that the identification of illegal live broadcast contents in the video live broadcast is very important.
In the prior art, a manager manages the illegal live broadcast content of a live broadcast platform and performs human patrol on a live broadcast room, wherein firstly, the manager enters the reported live broadcast room for patrol according to whether the manager receives a report from a user or not, and then randomly selects one live broadcast room for patrol if the manager does not receive the report; and then, corresponding processing is carried out according to the inspection result, if the inspection result has illegal live broadcast content, the live broadcast room is warned or closed, then inspection is continuously carried out according to the report of the user or one live broadcast room is randomly selected for inspection, and if the inspection result has no illegal live broadcast content, inspection is carried out according to the report of the user. Because live video is patrolled by managers, the detection is almost general survey of all live content, difference treatment is lacked, and huge manpower investment is needed; meanwhile, the identification of live video can be only carried out in a spot check mode, so that the processing efficiency of the live video is low.
Therefore, in the prior art, the illegal live broadcast content of the live broadcast platform is patrolled by the administrator, so that the real-time performance is poor, the efficiency is low, the cost is high, and the expansibility is low.
Disclosure of Invention
The application aims to provide a method and equipment for identifying abnormal live broadcast information in live video broadcast, and the method and equipment are used for solving the problems of poor real-time performance, low efficiency, high cost and low expansibility caused by the fact that a person patrols illegal live broadcast contents of a live broadcast platform through an administrator in the prior art.
According to one aspect of the application, a method for identifying abnormal live broadcast information in video live broadcast is provided, and the method comprises the following steps:
detecting whether the live video in the first authentication mode meets a trigger condition for switching to a second authentication mode;
when the triggering condition is met, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode;
and determining whether the video live broadcast contains abnormal live broadcast information or not according to the live broadcast image frame.
Further, when the trigger condition is met, switching the live video to the second authentication mode, and capturing live video frames of the live video according to an image capturing frequency corresponding to the second authentication mode includes:
when the trigger condition is met, switching the live video to the second authentication mode;
determining image capture frequency corresponding to the second identification mode;
and capturing live broadcast image frames of the video live broadcast according to the image capturing frequency corresponding to the second identification mode.
According to another aspect of the present application, there is also provided an apparatus for identifying abnormal live broadcast information in a live video broadcast, including:
the switching detection device is used for detecting whether the live video in the first authentication mode meets the triggering condition for switching to the second authentication mode;
the switching device is used for switching the live video broadcast to the second identification mode when the trigger condition is met, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode;
and the abnormity determining device is used for determining whether the video live broadcast contains abnormal live broadcast information according to the live broadcast image frame.
Further, the switching device includes:
the switching unit is used for switching the live video to the second authentication mode when the triggering condition is met;
the frequency determining unit is used for determining the image capturing frequency corresponding to the second identification mode;
and the capturing unit is used for capturing the live broadcast image frames of the video live broadcast according to the image capturing frequency corresponding to the second identification mode.
Compared with the prior art, the method and the equipment for identifying the abnormal live broadcast information in the live video broadcast provided by the application detect whether the live video broadcast in the first identification mode meets the triggering condition for switching to the second identification mode; when the triggering condition is met, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode; determining whether the live video broadcast contains abnormal live broadcast information or not according to the live broadcast image frame, realizing switching of different identification modes of the live video broadcast, simultaneously capturing the live broadcast image frame of the live video broadcast according to the image capture frequency corresponding to the second identification mode, and determining whether the live video broadcast contains the abnormal live broadcast information or not according to the live broadcast image frame so as to treat the abnormal live broadcast information in the live video broadcast, thereby reducing manual participation, reducing the cost of manual participation and increasing the lasting efficiency of the identification of the abnormal live broadcast information in the live video broadcast; further, when the trigger condition is met, switching the live video to the second authentication mode; determining image capture frequency corresponding to the second identification mode; and capturing the live broadcast image frames of the live video according to the image capturing frequency corresponding to the second identification mode, so that the live broadcast image frames of the live video in the second identification mode are captured, manual patrol is avoided, and the real-time performance of capturing and processing the live broadcast image frames of the live video is effectively improved.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 illustrates a flow diagram of a method for identifying anomalous live information in a live video, in accordance with an aspect of the subject application;
fig. 2 is a flow chart of a switching method of an authentication mode for authenticating abnormal live information in a live video according to a preferred embodiment of an aspect of the present application;
FIG. 3 illustrates a flow diagram of a capture method for identifying live video frames of anomalous live information in a live video according to a preferred embodiment of an aspect of the present application;
FIG. 4 shows a flow diagram of a picture authentication method for authenticating abnormal live information in a live video according to a preferred embodiment of an aspect of the present application;
FIG. 5 illustrates a flow diagram of a method for identifying anomalous live information in a live video in accordance with a preferred embodiment of an aspect of the subject application;
FIG. 6 illustrates a block diagram of an apparatus for identifying anomalous live information in a live video feed, in accordance with an aspect of the subject application;
fig. 7 is a schematic structural diagram of an apparatus for discriminating abnormal live information in a video live according to a preferred embodiment of an aspect of the present application.
The same or similar reference numbers in the drawings identify the same or similar elements.
Detailed Description
The present application is described in further detail below with reference to the attached figures.
Fig. 1 shows a flow diagram of a method for identifying abnormal live information in a live video according to an aspect of the present application. The method comprises a step S11, a step S12 and a step S13, wherein the step S11 detects whether the live video in the first authentication mode meets a trigger condition for switching to the second authentication mode; when the triggering condition is met, the step S12 switches the live video broadcast to the second authentication mode, and captures a live video frame of the live video broadcast according to an image capture frequency corresponding to the second authentication mode; the step S13 determines whether the live video includes abnormal live information from the live image frame.
It should be noted that the first authentication mode may include, but is not limited to, a conventional authentication mode. In a preferred embodiment of the present application, it is preferred that the regular authentication mode is the first authentication mode. It should be understood by those skilled in the art that the conventional authentication mode is only a preferred embodiment of the first authentication mode, and other existing or future possible first authentication modes may be applicable to the present application and are included within the scope of the present application and are hereby incorporated by reference.
It should be noted that the second authentication mode may include, but is not limited to, a super authentication mode. In a preferred embodiment of the present application, it is preferable that the regular authentication mode is the second authentication mode. It will be understood by those skilled in the art that the super authentication mode is only a preferred embodiment of the second authentication mode, and other existing or future second authentication modes may be applicable to the present application and are included within the scope of the present application and are hereby incorporated by reference.
It should be noted that the abnormal live broadcast information may include, but is not limited to, erroneous live broadcast information, illegal live broadcast information, abnormal video live broadcast information, and the like, and the illegal live broadcast information includes illegal speech live broadcast information, illegal edge live broadcast information, illegal video live broadcast information, and the like. In a preferred embodiment of the present application, it is preferable that the abnormal live broadcast information is illegal live broadcast information, and it should be understood by those skilled in the art that the illegal live broadcast information is only a preferred embodiment of the abnormal live broadcast information, and other existing or future abnormal live broadcast information, if applicable, should be included in the scope of the present application, and is included herein by reference.
In an embodiment of the present application, the step S11 detects whether the live video in the first authentication mode satisfies a trigger condition for switching to the second authentication mode; when the triggering condition is met, the step S12 switches the live video broadcast to the second authentication mode, and captures a live video frame of the live video broadcast according to an image capture frequency corresponding to the second authentication mode; step S13 determines whether the live video broadcast includes abnormal live broadcast information according to the live broadcast image frames, so as to switch different discrimination modes of the live video broadcast, and at the same time, captures live broadcast image frames of the live video broadcast according to the image capture frequency corresponding to the second discrimination mode, and determines whether the live video broadcast includes abnormal live broadcast information according to the live broadcast image frames, so as to treat the abnormal live broadcast information in the live video broadcast, thereby reducing manual participation, reducing the cost of manual participation, and increasing the durability and efficiency of the abnormal live broadcast information discrimination in the live video broadcast.
In a preferred embodiment of the present application, before the step S11, a video live broadcast of a live broadcast room needs to be subjected to data processing, and first, an identification ID is set for each live broadcast room, so as to uniquely identify the video live broadcast of the live broadcast room in a video live broadcast authentication mode; then, recording the live broadcast flow change condition of the live broadcast room under each identification ID, wherein the live broadcast flow is in direct proportion to the number of viewers, so that the change of the number of viewers can be represented by the change of the live broadcast flow, and when illegal live broadcast information is subjected to live broadcast, the number of viewers in various mood states is increased, and the change of the number of viewers in the live broadcast room is represented by the change condition; then, each identification ID corresponds to a flag bit N, and the flag bit is used for indicating the condition that the identification ID is in a white list state, a gray list state and a black list state, wherein the white list is used for identifying the live videos which are reported to be live, such as the live videos of the evening party or the sports event, the gray list is used for identifying the live videos which are warned to be live or the live videos which walk around the edge of the illegal live videos, and the black list is used for marking the live videos which have been recorded in an illegal way; finally, when the default of the flag bit is 0, the current live broadcast room is not in any list; when the flag bit is 1, indicating that the current live broadcast room is in a white list; when the flag bit is 2, the current live broadcast room is in a grey list; when the flag bit is 3, it indicates that the current live broadcast room is in the blacklist.
It should be noted that the direct broadcasting is realizedBefore switching the authentication mode of live video, it is necessary to determine whether to switch live video by acquiring the change of live video traffic corresponding to the live video, so that it is necessary to calculate the live video traffic of each live video, and Q is selected in a preferred embodiment of the present applicationt,iTo indicate the live traffic at time t of the live room with ID i, although it will be understood by those skilled in the art that Q ist,iOther existing or future algorithms for representing live traffic at any time in a live room, such as may be applicable to the present application, are also included within the scope of the present application and are hereby incorporated by reference.
Next, in the preferred embodiment of the present application, the video live broadcast platforms corresponding to the live broadcast rooms all have a specific flow monitoring system for recording historical live broadcast flow and monitoring current live broadcast flow, and Q is usedt,iThe live broadcast flow of the live broadcast room with ID i at the time t is represented, and due to the fact that the normal live broadcast flow change situation of the live broadcast room does not have too large difference with the historical live broadcast flow change, when the live broadcast flow of one live broadcast room is greatly increased and has change different from the historical increase situation, illegal live broadcast information signals can possibly occur in the live broadcast room. Here, an upper limit of a live traffic growth rate under a normal condition may be calculated according to historical data of live traffic, and in a preferred embodiment of the present application, the live traffic growth rate is calculated as follows:
Figure GDA0002836952130000061
wherein p ist,iIndicating the live traffic growth rate at time t of the live room with identification ID i,
Figure GDA0002836952130000071
representing the sum of the live traffic in the time period between time T-T and time T of the live room with identity ID i,
Figure GDA0002836952130000072
the method comprises the steps that the sum of live broadcast flow of a live broadcast room with an identification ID of i in a time period from time T-2T to time T-T is represented, wherein T is a configuration value of a flow monitoring system in a video live broadcast platform corresponding to the live broadcast room, and the configuration value T can be correspondingly adjusted according to specific requirements of an actual live broadcast room; by Qmax,iIndicating a live traffic threshold, p, for a live room with an ID of imax,i denotes a live traffic growth rate threshold identifying the live room with ID i.
Further, the step S11 detects whether the live video in the first authentication mode satisfies the trigger condition for switching to the second authentication mode, where the trigger condition includes at least any one of:
live broadcast flow of the video live broadcast exceeds a preset live broadcast flow threshold;
the flow rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow rate threshold value;
live broadcast flow of the live video exceeds a preset live broadcast flow upper limit threshold of the first authentication mode;
the flow increase rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow increase rate lower limit threshold of the first identification mode;
live streaming of the live video exceeds a predetermined live streaming threshold of the second authentication mode;
the rate of flow increase of live traffic of the live video exceeds a predetermined live traffic increase rate threshold of the second authentication mode.
Following the above embodiments of the present application, the triggering condition in step S12 may be at least any one of the following: live broadcast flow Q of live videot,iExceeding a predetermined live traffic threshold Qmax,i(ii) a If the live broadcast flow of the video live broadcast has the flow growth rate pt,iExceeding a predetermined live traffic growth rate threshold pmax,i; live broadcast flow Q of live videot,iExceeding a predetermined upper live traffic threshold Q of the regular authentication modeUpper limit of max,i(ii) a The live broadcast flow rate p of the live broadcast flow of the videot,iExceeding a predetermined lower threshold value P of the rate of increase of live traffic of said regular authentication modeLower limit of max,i(ii) a Live broadcast flow Q of live videot,iExceeding a predetermined live traffic threshold Q of the super authentication modeSuper grade max,i(ii) a The live broadcast flow rate p of the live broadcast flow of the videot,iExceeding a predetermined live traffic growth rate threshold P of the super authentication modeSuper grade max,i。
Further, the step S12 includes: when the triggering condition is met, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode; specifically, the step S12 includes: and when the triggering condition is met and the state of the target video is a grey list state, a blacklist state or an undetermined state, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode.
Next, in the preferred embodiment of the present application, switching of the authentication mode of live video will be performed by taking the following trigger condition as an example; preferably the trigger condition is: live traffic growth rate p of live broadcast room when identification ID is it,iIs less than pmax,iIf the identification ID is in the normal condition, identifying the live video in the live broadcast room with the identification ID being i by adopting a conventional identification mode; live traffic growth rate p of live broadcast room when identification ID is it,iGreater than pmax,When the identification ID is i, the live video of the live broadcast room with the identification ID being i is in an abnormal mode, and the live video is identified by adopting a super identification mode; namely, if the live broadcast flow of the live broadcast room with the ID i is in a normal state, starting a conventional authentication mode, and if the ID is in a normal state, starting the conventional authentication modeAnd when the live broadcast flow of the live broadcast room with the ID of i is in an abnormal state, starting a super authentication mode, wherein the conventional authentication mode is different from the super authentication mode in that the conventional authentication mode and the super authentication mode have different image capture frequencies.
For example, fig. 2 is a flowchart illustrating a switching method of a discrimination mode for discriminating abnormal live information in a live video according to a preferred embodiment of an aspect of the present application, the switching method including step S21, step S22, step S23, step S24, and step S25. Wherein, in the step S21, a normal authentication mode is started; step S22, determining whether the live broadcast traffic increase rate reaches an upper threshold, if yes, executing step S23, otherwise, executing step S24; the step S23, determining whether the target video is in the white list, if yes, performing step S24, otherwise, performing step S25; said step S24, continuing the normal authentication mode; said step S25, starting super authentication mode; further, in the step S23, it is determined whether the target video is in a white list, and if not, the live video is switched to the super authentication mode, where a condition that the state of the target video is not in the white list state at least includes any one of: a blacklist status, or a pending status; further, in the step S25, capturing live broadcast image frames of the live video broadcast based on the image capture frequency corresponding to the super-authentication mode, and if there is no abnormal live broadcast information after the super-authentication mode is T1, automatically switching to the normal mode, where T1 is the super-authentication time of the video live broadcast platform configured to run the super-authentication mode, and the super-authentication time T1 may be adaptively configured according to the specific requirement of the actual live broadcast room.
Further, the step S12 includes: when the triggering condition is met, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode; specifically, the step S12 further includes: and when the trigger condition is met and the state of the target video is a white list state, keeping the live video in the first identification mode.
Following the above preferred embodiment of the present application, when the trigger condition is satisfied: live traffic growth rate pt,i has reached the upper threshold pmax,i, and the status of the target video is a white list status, continuing to keep the live video in the normal authentication mode, as shown in fig. 2.
Further, the step S12 includes: when the triggering condition is met, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode; specifically, the step S12 includes: when the trigger condition is met, switching the live video to the second authentication mode; determining image capture frequency corresponding to the second identification mode; and capturing live broadcast image frames of the video live broadcast according to the image capturing frequency corresponding to the second identification mode.
Next to the above preferred embodiment of the present application, in the step S12, if the preferred triggering condition is satisfied, the preferred triggering condition is: if the live broadcast flow increase rate in fig. 2 reaches the upper limit threshold and the live video is not in the white list state, switching the live video from the conventional authentication mode to the super authentication mode; wherein the regular authentication mode is different from the super authentication mode in having different image capture frequencies. When the live video is in a conventional authentication mode, capturing live video frames from the live video with an image capture period T2 corresponding to the conventional mode in a live video platform, wherein T2 is the inverse number of the image capture frequency of the live video frames in the conventional authentication mode configured by the live video platform of a live video room corresponding to the live video, and T2 can configure different values of the conventional image capture period according to different live video platforms; when the preferable trigger condition is met, the live video is switched from the conventional authentication mode to the super authentication mode; then, an image capture period T3 corresponding to the super authentication mode needs to be determined, where T3 dynamically adjusts the reciprocal of the image capture frequency for capturing live broadcast image frames according to the change condition of live broadcast traffic in the super authentication mode and the condition that the live video is in different list states; then, the step S12 captures a live video frame of the video according to the image capture cycle T3 corresponding to the super-authentication mode.
For example, fig. 3 shows a flowchart of a capture method for identifying live video frames of abnormal live information in a video live according to a preferred embodiment of an aspect of the present application, wherein the capture method includes step S31, step S32, step S33 and step S34. In step S31, capturing a live video frame of a live video; the step S32, determining whether a trigger condition corresponding to switching to the super authentication mode is satisfied, if not, performing step S33, and if so, performing step S34; the step S33, capturing live video frames at a set T2 time interval; in step S34, the capture time interval is adjusted to the image capture period T3 corresponding to the super-authentication mode.
Further, the determining of the image capture frequency corresponding to the second authentication mode in the step S12 includes:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode.
Following the above preferred embodiment of the present application, the image capture period corresponding to the normal authentication mode is T2, and the image capture period corresponding to the second authentication mode is T3, wherein the image capture period T3 corresponding to the super authentication mode is calculated as follows:
Figure GDA0002836952130000101
n is a zone bit corresponding to each live broadcast room of the ID, N is a positive integer greater than or equal to 2 in the super discrimination mode, and pt,iThe live broadcast traffic increase rate of the live broadcast traffic of the live broadcast currently corresponding to the video broadcast in the live broadcast room is shown, therefore, if the image capture period corresponding to the second authentication mode is T3 and is smaller than the image capture period corresponding to the conventional authentication mode is T2, the image capture frequency corresponding to the second authentication mode is greater than the image capture frequency corresponding to the first authentication mode.
Further, the determining of the image capture frequency corresponding to the second authentication mode in the step S12 includes:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode and the live video state or current flow growth information, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode.
It should be noted that the status of the live video may include, but is not limited to, a white list status, a grey list status, a black list status, that is, a pending status, and the like. Of course, other existing or future situations where live video may be available are also encompassed within the scope of the present application and are hereby incorporated by reference.
It should be noted that the current traffic increase information of the live video may include, but is not limited to, a current traffic increase rate, a current traffic increase amount, and the like. Of course, other existing or future current traffic growth information for the live video, as applicable to the present application, is also included within the scope of the present application and is hereby incorporated by reference.
For example, in the step S12, if the video live broadcast has a corresponding image capture period T2 of 7 seconds in the normal mode, and the video live broadcast is in the grey list state, the flag N corresponding to the live broadcast room of the identifier ID is 2, and the live broadcast room currently corresponds to the traffic increase rate p of the live broadcast traffic of the video live broadcastt,i1.5, based on the calculation formula of the image capture period T3 corresponding to the super authentication mode:
Figure GDA0002836952130000111
and obtaining the image capture period T3 corresponding to the super authentication mode as 2 seconds.
Further, the step S13 includes: determining whether the video live broadcast contains abnormal live broadcast information or not according to the live broadcast image frame; specifically, the step S13 includes: acquiring a first authentication result and a second authentication result corresponding to the live broadcast image frame; and determining whether the video live broadcast contains abnormal live broadcast information or not according to the first identification result and the second identification result.
It should be noted that the image authentication algorithm for the live image frame of the present application may include, but is not limited to, the following: color image based discrimination algorithms and contour image based discrimination algorithms, etc. Of course, other image discrimination algorithms for live video frames that are currently or later become available, such as may be applicable to the present application, are also intended to be included within the scope of the present application and are hereby incorporated by reference.
Next, in the above preferred embodiment of the present application, in order to reduce the false positive probability, in step S13, picture identification is performed on the live broadcast image frame through two-stage identification, where the two-stage identification is one-stage identification and two-stage identification, and the one-stage identification and the two-stage identification use different picture identification techniques, so as to reduce the false positive probability, and the two-stage identification may be implemented through manual identification, and after a second identification result obtained after the picture identification technique is used, the two-stage identification using a manual identification mode can also effectively reduce manual input, thereby improving the picture processing efficiency. In step S13, first, a first authentication result and a second authentication result corresponding to the live video frame after two-stage authentication are obtained; and then determining whether the video live broadcast contains illegal live broadcast information according to the first identification result and the second identification result, wherein in a preferred embodiment of the application, the abnormal live broadcast information is preferably the illegal live broadcast information.
Further, the acquiring the first authentication result and the second authentication result corresponding to the live image frame in step S13 includes:
acquiring a first identification result corresponding to the live broadcast image frame;
and when the first identification result indicates that the video live broadcast contains abnormal live broadcast information, acquiring a second identification result corresponding to the live broadcast image frame.
Next to the above preferred embodiment of the present application, after the live broadcast image frames are primarily identified through primary identification in step S13, if the first identification result indicates that the live video includes illegal live broadcast information, picture identification is performed on the illegal live broadcast information through secondary identification to obtain a second identification result corresponding to the live broadcast image frames corresponding to the illegal live broadcast information, so as to reduce the probability of misjudgment, as shown in fig. 4, fig. 4 shows a schematic flow diagram of a picture identification method for identifying abnormal live broadcast information in live video broadcast according to a preferred embodiment of an aspect of the present application, where the picture identification method includes step S41, step S42, and step S43. Wherein, in the step S41, primary authentication is performed on a live image frame of the live video; step S42, determining a first authentication result containing illegal live broadcast information corresponding to the first-level authentication; in step S43, performing secondary authentication on the live video frames of the live video including illegal live information to determine a second authentication result.
Further, the method further comprises:
when the first identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state as a grey list state; or
And when the second identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state to be a blacklist state.
Next, in the above preferred embodiment of the present application, in step S13, adding the live video corresponding to the illegal live broadcast information after the primary authentication into a grey list, and setting the live video flag N to 2; or, in the step S13, the live video corresponding to the illegal live broadcast information after the primary authentication is added to a blacklist, the live video flag N is set to 3, and a live broadcast room corresponding to the live video is warned or signed.
Fig. 5 shows a flow diagram of a method for identifying abnormal live information in a video live according to a preferred embodiment of an aspect of the present application. The method includes step S501, step S502, step S503, step S504, step S505, step S506, step S507, step S508, and step S509. In step S501, a live broadcast room and a corresponding live video are subjected to data processing; step S502, starting a conventional authentication mode; in the step S503, it is determined whether the live broadcast flow corresponding to the live video broadcast suddenly increases, if not, the step S504 is executed, and if yes, the step S505 is executed; in the step S504, the normal authentication mode is continued, and in the normal authentication mode in the step S504, the captured live video frame is also transferred to the primary authentication in the step S507; the step S505 is to determine whether the live video is in a white list, if so, execute the step S504, otherwise, execute the step S506; step S506, starting a super authentication mode; the step S507, performing primary identification on the live broadcast image frame of the video live broadcast; step S508, processing the first authentication result and starting secondary authentication; the step S509 processes the second authentication result.
Fig. 6 illustrates a schematic structural diagram of an apparatus for discriminating abnormal live information in a live video according to an aspect of the present application. The device comprises a switching detection device 11, a switching device 12 and an abnormity determination device 13, wherein the switching detection device 11 detects whether the live video in the first authentication mode meets the triggering condition of switching to the second authentication mode; when the trigger condition is met, the switching device 12 switches the live video broadcast to the second authentication mode, and captures live video frames of the live video broadcast according to an image capture frequency corresponding to the second authentication mode; the anomaly determination means 13 determines from the live video frame whether the live video includes anomalous live information.
Here, the device 1 includes, but is not limited to, a client device, a network device, and a device in which the client device and the network device are integrated through a network. The network device includes an electronic device capable of automatically performing numerical calculation and information processing according to instructions set or stored in advance, and hardware thereof includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a programmable gate array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like. Including, but not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a VPN network, a wireless Ad Hoc network (Ad Hoc network), etc. Preferably, the device 1 may also be a script program running on a device formed by integrating the user device and a network device through a network. Of course, it will be understood by those skilled in the art that the above-described apparatus 1 is merely exemplary, and that other existing or future implementations of the apparatus 1, as applicable to the present application, are also intended to be encompassed within the scope of the present application and are hereby incorporated by reference.
The above devices are operated continuously, and herein, those skilled in the art should understand that "continuously" means that the above devices are operated in real time or according to the set or real-time adjusted operating mode requirement.
It should be noted that the first authentication mode may include, but is not limited to, a conventional authentication mode. In a preferred embodiment of the present application, it is preferred that the regular authentication mode is the first authentication mode. It should be understood by those skilled in the art that the conventional authentication mode is only a preferred embodiment of the first authentication mode, and other existing or future possible first authentication modes may be applicable to the present application and are included within the scope of the present application and are hereby incorporated by reference.
It should be noted that the second authentication mode may include, but is not limited to, a super authentication mode. In a preferred embodiment of the present application, it is preferable that the regular authentication mode is the second authentication mode. It will be understood by those skilled in the art that the super authentication mode is only a preferred embodiment of the second authentication mode, and other existing or future second authentication modes may be applicable to the present application and are included within the scope of the present application and are hereby incorporated by reference.
It should be noted that the abnormal live broadcast information may include, but is not limited to, erroneous live broadcast information, illegal live broadcast information, abnormal video live broadcast information, and the like, and the illegal live broadcast information includes illegal speech live broadcast information, illegal edge live broadcast information, illegal video live broadcast information, and the like. In a preferred embodiment of the present application, it is preferable that the abnormal live broadcast information is illegal live broadcast information, and it should be understood by those skilled in the art that the illegal live broadcast information is only a preferred embodiment of the abnormal live broadcast information, and other existing or future abnormal live broadcast information, if applicable, should be included in the scope of the present application, and is included herein by reference.
In the embodiment of the present application, the switching detection device 11 detects whether the live video in the first authentication mode satisfies the trigger condition for switching to the second authentication mode; when the trigger condition is met, the switching device 12 switches the live video broadcast to the second authentication mode, and captures live video frames of the live video broadcast according to an image capture frequency corresponding to the second authentication mode; the abnormity determining device 13 determines whether the live video broadcast contains abnormal live broadcast information according to the live broadcast image frame, so that switching of different identification modes of the live video broadcast is realized, meanwhile, the live broadcast image frame of the live video broadcast is captured according to the image capture frequency corresponding to the second identification mode, and whether the live video broadcast contains abnormal live broadcast information is determined according to the live broadcast image frame, so that the abnormal live broadcast information in the live video broadcast is to be processed, manual participation is reduced, the cost of manual participation is reduced, and the lasting and high efficiency of abnormal live broadcast information identification in the live video broadcast is increased.
In the preferred embodiment of the present application, before the switching detection device 11, it is necessary to perform data processing on the live video of the live broadcast room, and first, an identification ID is set for each live broadcast room, so as to uniquely identify the live video of the live broadcast room in a live video identification mode; then, recording the live broadcast flow change condition of the live broadcast room under each identification ID, wherein the live broadcast flow is in direct proportion to the number of viewers, so that the change of the number of viewers can be represented by the change of the live broadcast flow, and when illegal live broadcast information is subjected to live broadcast, the number of viewers in various mood states is increased, and the change of the number of viewers in the live broadcast room is represented by the change condition; then, each identification ID corresponds to a flag bit N, and the flag bit is used for indicating the condition that the identification ID is in a white list state, a gray list state and a black list state, wherein the white list is used for identifying the live videos which are reported to be live, such as the live videos of the evening party or the sports event, the gray list is used for identifying the live videos which are warned to be live or the live videos which walk around the edge of the illegal live videos, and the black list is used for marking the live videos which have been recorded in an illegal way; finally, when the default of the flag bit is 0, the current live broadcast room is not in any list; when the flag bit is 1, indicating that the current live broadcast room is in a white list; when the flag bit is 2, the current live broadcast room is in a grey list; when the flag bit is 3, it indicates that the current live broadcast room is in the blacklist.
It should be noted that before the switching of the authentication mode of the live video of the live broadcast room is implemented, it is necessary to determine whether to switch the live broadcast room by obtaining the change condition of the live broadcast traffic corresponding to the live broadcast room, so that it is necessary to calculate the live broadcast traffic of each live broadcast room, and then Q is selected in a preferred embodiment of the present applicationt,iTo indicate the live traffic at time t of the live room with ID i, although it will be understood by those skilled in the art that Q ist,iOther existing or future algorithms for representing live traffic at any time in a live room, such as may be applicable to the present application, are also included within the scope of the present application and are hereby incorporated by reference.
The above preferred embodiments of the present application followThe video live broadcast platform corresponding to the live broadcast room is provided with a specific flow monitoring system for recording historical live broadcast flow and monitoring current live broadcast flow, and Q is used for monitoring the current live broadcast flowt,iThe live broadcast flow of the live broadcast room with ID i at the time t is represented, and due to the fact that the normal live broadcast flow change situation of the live broadcast room does not have too large difference with the historical live broadcast flow change, when the live broadcast flow of one live broadcast room is greatly increased and has change different from the historical increase situation, illegal live broadcast information signals can possibly occur in the live broadcast room. Here, an upper limit of a live traffic growth rate under a normal condition may be calculated according to historical data of live traffic, and in a preferred embodiment of the present application, the live traffic growth rate is calculated as follows:
Figure GDA0002836952130000161
wherein p ist,iIndicating the live traffic growth rate at time t of the live room with identification ID i,
Figure GDA0002836952130000162
representing the sum of the live traffic in the time period between time T-T and time T of the live room with identity ID i,
Figure GDA0002836952130000163
the method comprises the steps that the sum of live broadcast flow of a live broadcast room with an identification ID of i in a time period from time T-2T to time T-T is represented, wherein T is a configuration value of a flow monitoring system in a video live broadcast platform corresponding to the live broadcast room, and the configuration value T can be correspondingly adjusted according to specific requirements of an actual live broadcast room; by QmaxI denotes the live traffic threshold of the live room with ID i, pmax,iIndicating a live traffic growth rate threshold for the live room with identification ID i.
Further, the trigger condition in the handover detection device 11 includes at least any one of:
live broadcast flow of the video live broadcast exceeds a preset live broadcast flow threshold;
the flow rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow rate threshold value;
live broadcast flow of the live video exceeds a preset live broadcast flow upper limit threshold of the first authentication mode;
the flow increase rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow increase rate lower limit threshold of the first identification mode;
live streaming of the live video exceeds a predetermined live streaming threshold of the second authentication mode;
the rate of flow increase of live traffic of the live video exceeds a predetermined live traffic increase rate threshold of the second authentication mode.
Following the above embodiments of the present application, the trigger condition in the switching detection device 11 may be at least any one of the following: live broadcast flow Q of live videot,iExceeding a predetermined live traffic threshold Qmax,i(ii) a If the live broadcast flow of the video live broadcast has the flow growth rate pt,iExceeding a predetermined live traffic growth rate threshold pmax,i(ii) a Live broadcast flow Q of live videot,iExceeding a predetermined upper live traffic threshold Q of the regular authentication modeUpper limit of max,i(ii) a The live broadcast flow rate p of the live broadcast flow of the videot,iExceeding a predetermined lower threshold value P of the rate of increase of live traffic of said regular authentication modeLower limit of max,i(ii) a Live broadcast flow Q of live videot,iExceeding a predetermined live traffic threshold Q of the super authentication modeSuper grade max,i(ii) a The live broadcast flow rate p of the live broadcast flow of the videot,iExceeding a predetermined live traffic growth rate threshold P of the super authentication modeSuper grade max,i。
Further, the switching device 12 is configured to:
and when the triggering condition is met and the state of the target video is a grey list state, a blacklist state or an undetermined state, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode.
Next, in the preferred embodiment of the present application, switching of the authentication mode of live video will be performed by taking the following trigger condition as an example; preferably the trigger condition is: live traffic growth rate p of live broadcast room when identification ID is it,iIs less than pmax,iIf the identification ID is in the normal condition, identifying the live video in the live broadcast room with the identification ID being i by adopting a conventional identification mode; live traffic growth rate p of live broadcast room when identification ID is it,iGreater than pmax,When the identification ID is i, the live video of the live broadcast room with the identification ID being i is in an abnormal mode, and the live video is identified by adopting a super identification mode; namely, if the live broadcast flow of the live broadcast room with the identifier ID i is in a normal state, starting a conventional authentication mode, and if the live broadcast flow of the live broadcast room with the identifier ID i is in an abnormal state, starting a super authentication mode, wherein the conventional authentication mode and the super authentication mode are different in that the conventional authentication mode and the super authentication mode have different image capturing frequencies.
For example, fig. 2 is a flowchart illustrating a switching method of a discrimination mode for discriminating abnormal live information in a live video according to a preferred embodiment of an aspect of the present application, the switching method including step S21, step S22, step S23, step S24, and step S25. Wherein, in the step S21, a normal authentication mode is started; step S22, determining whether the live broadcast traffic increase rate reaches an upper threshold, if yes, executing step S23, otherwise, executing step S24; the step S23, determining whether the target video is in the white list, if yes, performing step S24, otherwise, performing step S25; said step S24, continuing the normal authentication mode; said step S25, starting super authentication mode; further, in the step S23, it is determined whether the target video is in a white list, and if not, the live video is switched to the super authentication mode, where a condition that the state of the target video is not in the white list state at least includes any one of: a blacklist status, or a pending status; further, in the step S25, capturing live broadcast image frames of the live video broadcast based on the image capture frequency corresponding to the super-authentication mode, and if there is no abnormal live broadcast information after the super-authentication mode is T1, automatically switching to the normal mode, where T1 is the super-authentication time of the video live broadcast platform configured to run the super-authentication mode, and the super-authentication time T1 may be adaptively configured according to the specific requirement of the actual live broadcast room.
Further, the step switching device 12 is further configured to:
and when the trigger condition is met and the state of the target video is a white list state, keeping the live video in the first identification mode.
Following the above preferred embodiment of the present application, when the trigger condition is satisfied: live traffic growth rate pt,iHas reached the upper threshold pmax,iAnd if the state of the target video is the white list state, the live video is continuously kept in the normal authentication mode, as shown in fig. 2.
Further, the switching device 12 includes: a switching unit (not shown), a frequency determining unit (not shown) and a grabbing unit (not shown), wherein the switching unit (not shown) is configured to: when the trigger condition is met, switching the live video to the second authentication mode; the frequency determination unit (not shown) is configured to: determining image capture frequency corresponding to the second identification mode; the gripping unit (not shown) is configured to: and capturing live broadcast image frames of the video live broadcast according to the image capturing frequency corresponding to the second identification mode.
Next to the above preferred embodiment of the present application, in the switching device 12, if the preferred triggering condition is satisfied, wherein the preferred triggering condition is: if the live broadcast flow increase rate in fig. 2 reaches the upper limit threshold and the live video is not in the white list state, switching the live video from the conventional authentication mode to the super authentication mode; wherein the regular authentication mode is different from the super authentication mode in having different image capture frequencies. When the live video is in a conventional authentication mode, capturing live video frames from the live video with an image capture period T2 corresponding to the conventional mode in a live video platform, wherein T2 is the inverse number of the image capture frequency of the live video frames in the conventional authentication mode configured by the live video platform of a live video room corresponding to the live video, and T2 can configure different values of the conventional image capture period according to different live video platforms; when the preferable trigger condition is met, the live video is switched from the conventional authentication mode to the super authentication mode; then, an image capture period T3 corresponding to the super authentication mode needs to be determined, where T3 dynamically adjusts the reciprocal of the image capture frequency for capturing live broadcast image frames according to the change condition of live broadcast traffic in the super authentication mode and the condition that the live video is in different list states; then, the capture unit (not shown) captures a live image frame of the video according to an image capture cycle T3 corresponding to the super-authentication mode.
For example, fig. 3 shows a flowchart of a capture method for identifying live video frames of abnormal live information in a video live according to a preferred embodiment of an aspect of the present application, wherein the capture method includes step S31, step S32, step S33 and step S34. In step S31, capturing a live video frame of a live video; the step S32, determining whether a trigger condition corresponding to switching to the super authentication mode is satisfied, if not, performing step S33, and if so, performing step S34; the step S33, capturing live video frames at a set T2 time interval; in step S34, the capture time interval is adjusted to the image capture period T3 corresponding to the super-authentication mode.
Further, the frequency determination unit (not shown) is configured to:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode.
Following the above preferred embodiment of the present application, the image capture period corresponding to the normal authentication mode is T2, and the image capture period corresponding to the second authentication mode is T3, wherein the image capture period T3 corresponding to the super authentication mode is calculated as follows:
Figure GDA0002836952130000201
n is a zone bit corresponding to each live broadcast room of the ID, N is a positive integer greater than or equal to 2 in the super discrimination mode, and pt,iThe live broadcast traffic increase rate of the live broadcast traffic of the live broadcast currently corresponding to the video broadcast in the live broadcast room is shown, therefore, if the image capture period corresponding to the second authentication mode is T3 and is smaller than the image capture period corresponding to the conventional authentication mode is T2, the image capture frequency corresponding to the second authentication mode is greater than the image capture frequency corresponding to the first authentication mode.
Further, the frequency determination unit (not shown) is configured to:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode and the live video state or current flow growth information, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode.
It should be noted that the status of the live video may include, but is not limited to, a white list status, a grey list status, a black list status, that is, a pending status, and the like. Of course, other existing or future situations where live video may be available are also encompassed within the scope of the present application and are hereby incorporated by reference.
It should be noted that the current traffic increase information of the live video may include, but is not limited to, a current traffic increase rate, a current traffic increase amount, and the like. Of course, other existing or future current traffic growth information for the live video, as applicable to the present application, is also included within the scope of the present application and is hereby incorporated by reference.
For example, in the switching device 12, if the image capture period T2 corresponding to the live video in the normal mode is 7 seconds, and the live video is in the grey list state, the flag N corresponding to the live broadcast room of the identifier ID is 2, and the flow increase rate p of the live broadcast flow of the live broadcast currently corresponding to the live video in the live broadcast room is pt,i1.5, based on the calculation formula of the image capture period T3 corresponding to the super authentication mode:
Figure GDA0002836952130000211
and obtaining the image capture period T3 corresponding to the super authentication mode as 2 seconds.
It should be noted that the image authentication algorithm for the live image frame of the present application may include, but is not limited to, the following: color image based discrimination algorithms and contour image based discrimination algorithms, etc. Of course, other image discrimination algorithms for live video frames that are currently or later become available, such as may be applicable to the present application, are also intended to be included within the scope of the present application and are hereby incorporated by reference.
Further, the abnormality determination device 13 includes a result acquisition unit (not shown) and an abnormality determination unit (not shown), wherein the result acquisition unit (not shown) is configured to: acquiring a first authentication result and a second authentication result corresponding to the live broadcast image frame; the abnormality determination unit (not shown) is configured to: and determining whether the video live broadcast contains abnormal live broadcast information or not according to the first identification result and the second identification result.
Next, in the above preferred embodiment of the present application, in order to reduce the false positive probability, the live broadcast image frame is subjected to image identification by two-stage identification in the anomaly determination device 13, where the two-stage identification is one-stage identification and two-stage identification, and the one-stage identification and the two-stage identification use different image identification techniques, so as to reduce the false positive probability, the two-stage identification may be performed by manual identification, and after a second identification result obtained after the image identification technique is used, the two-stage identification using a manual identification mode may also effectively reduce manual input, thereby improving the image processing efficiency. In the result obtaining unit (not shown), first obtaining a first authentication result and a second authentication result corresponding to the live video frame after two-stage authentication; then, in the anomaly determination unit (not shown), it is determined whether the live video includes illegal live broadcast information according to the first authentication result and the second authentication result, wherein in a preferred embodiment of the present application, the illegal live broadcast information is preferred to be the abnormal live broadcast information.
Further, the result obtaining unit (not shown) comprises a first result obtaining sub-unit (not shown) and a second result obtaining sub-unit (not shown), wherein the first result obtaining sub-unit (not shown) is configured to: acquiring a first identification result corresponding to the live broadcast image frame; the second result obtaining subunit (not shown) is configured to: and when the first identification result indicates that the video live broadcast contains abnormal live broadcast information, acquiring a second identification result corresponding to the live broadcast image frame.
Next to the above preferred embodiment of the present application, after the live broadcast image frame is primarily identified through primary identification in the abnormality determining device 13, if the first identification result indicates that the live video includes illegal live broadcast information, picture identification is performed on the illegal live broadcast information through secondary identification, so as to obtain a second identification result corresponding to the live broadcast image frame corresponding to the illegal live broadcast information, so as to reduce the probability of misjudgment, as shown in fig. 4, fig. 4 shows a schematic flow diagram of a picture identification method for identifying abnormal live broadcast information in live video broadcast according to a preferred embodiment of an aspect of the present application, where the picture identification method includes step S41, step S42, and step S43. Wherein, in the step S41, primary authentication is performed on a live image frame of the live video; step S42, determining a first authentication result containing illegal live broadcast information corresponding to the first-level authentication; in step S43, performing secondary authentication on the live video frames of the live video including illegal live information to determine a second authentication result.
Further, the apparatus is further configured to:
when the first identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state as a grey list state; or
And when the second identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state to be a blacklist state.
Next, in the above preferred embodiment of the present application, the video live broadcast corresponding to the illegal live broadcast information after the primary authentication is added to a grey list in the abnormality determining device 13, and the video live broadcast flag N is set to 2; or, the live video corresponding to the illegal live broadcast information after the primary authentication is added to a blacklist in the abnormality determination device 13, the live video flag bit N is set to 3, and the live broadcast room corresponding to the live video is subjected to processing such as warning or number sealing.
Fig. 7 is a schematic structural diagram of an apparatus for discriminating abnormal live information in a video live according to a preferred embodiment of an aspect of the present application. The device 1 comprises a live broadcast room datamation module, a conventional identification mode module, a super identification mode module, a primary identification module, a secondary identification module and a data information recording module. Wherein, the live broadcast room datamation module is used for: performing data processing on the video live broadcast in the live broadcast room; the regular authentication mode module is to: performing conventional authentication on the video live broadcast of a live broadcast room; the super authentication mode module is configured to: performing super authentication on the video live broadcast of the live broadcast room after the triggering condition is met; the primary authentication module is configured to: performing primary picture authentication on live broadcast image frames of the video live broadcast in a conventional authentication mode and a super authentication mode; the secondary authentication module is configured to: performing secondary identification on the first identification result and a corresponding live broadcast image frame when illegal live broadcast information appears in the primary identification result; the data information recording module is used for: recording the digitalized information of the live video broadcast in the live broadcast room after being digitalized, a first identification result corresponding to the primary identification module and a second identification result corresponding to the secondary identification module; further, after the two-stage authentication, the live broadcast room is correspondingly processed based on the first authentication result and the second authentication result.
Compared with the prior art, the method and the equipment for identifying the abnormal live broadcast information in the live video broadcast provided by the application detect whether the live video broadcast in the first identification mode meets the triggering condition for switching to the second identification mode; when the triggering condition is met, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode; determining whether the live video broadcast contains abnormal live broadcast information or not according to the live broadcast image frame, realizing switching of different identification modes of the live video broadcast, simultaneously capturing the live broadcast image frame of the live video broadcast according to the image capture frequency corresponding to the second identification mode, and determining whether the live video broadcast contains the abnormal live broadcast information or not according to the live broadcast image frame so as to treat the abnormal live broadcast information in the live video broadcast, thereby reducing manual participation, reducing the cost of manual participation and increasing the lasting efficiency of the identification of the abnormal live broadcast information in the live video broadcast; further, when the trigger condition is met, switching the live video to the second authentication mode; determining image capture frequency corresponding to the second identification mode; and capturing the live broadcast image frames of the live video according to the image capturing frequency corresponding to the second identification mode, so that the live broadcast image frames of the live video in the second identification mode are captured, manual patrol is avoided, and the real-time performance of capturing and processing the live broadcast image frames of the live video is effectively improved.
It should be noted that the present application may be implemented in software and/or a combination of software and hardware, for example, implemented using Application Specific Integrated Circuits (ASICs), general purpose computers or any other similar hardware devices. In one embodiment, the software programs of the present application may be executed by a processor to implement the steps or functions described above. Likewise, the software programs (including associated data structures) of the present application may be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Additionally, some of the steps or functions of the present application may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.
In addition, some of the present application may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or techniques in accordance with the present application through the operation of the computer. Program instructions which invoke the methods of the present application may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the present application comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or a solution according to the aforementioned embodiments of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Claims (18)

1. A method for identifying abnormal live information in a live video, wherein the method comprises:
detecting whether the live video in a first identification mode meets a trigger condition for switching to a second identification mode according to the change condition of live video flow corresponding to the live video, wherein the image capture frequency corresponding to the second identification mode is greater than the image capture frequency corresponding to the first identification mode;
when the triggering condition is met and the state of a target video is a grey list state, a blacklist state or an undetermined state, switching the live video broadcast to the second identification mode, and capturing live video frames of the live video broadcast according to the image capturing frequency corresponding to the second identification mode, wherein the target video is the live video broadcast serving as a detection target;
and determining whether the video live broadcast contains abnormal live broadcast information or not according to the live broadcast image frame.
2. The method of claim 1, wherein the determining from the live image frame whether the video live includes anomalous live information comprises:
acquiring a first authentication result and a second authentication result corresponding to the live broadcast image frame;
and determining whether the video live broadcast contains abnormal live broadcast information or not according to the first identification result and the second identification result.
3. The method of claim 2, wherein the method further comprises:
when the first identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state as a grey list state; or
And when the second identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state to be a blacklist state.
4. The method of claim 2 or 3, wherein the obtaining of the first and second authentication results corresponding to the live image frame comprises:
acquiring a first identification result corresponding to the live broadcast image frame;
and when the first identification result indicates that the video live broadcast contains abnormal live broadcast information, acquiring a second identification result corresponding to the live broadcast image frame.
5. The method of claim 1, wherein the method further comprises:
and when the trigger condition is met and the state of the target video is a white list state, keeping the live video in the first identification mode.
6. The method of claim 1, wherein the trigger condition comprises at least any one of:
live broadcast flow of the video live broadcast exceeds a preset live broadcast flow threshold;
the flow rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow rate threshold value;
live broadcast flow of the live video exceeds a preset live broadcast flow upper limit threshold of the first authentication mode;
the flow increase rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow increase rate lower limit threshold of the first identification mode;
live streaming of the live video exceeds a predetermined live streaming threshold of the second authentication mode;
the rate of flow increase of live traffic of the live video exceeds a predetermined live traffic increase rate threshold of the second authentication mode.
7. The method of claim 1, wherein capturing live video frames of the video live at an image capture frequency corresponding to the second authentication mode comprises:
determining image capture frequency corresponding to the second identification mode;
and capturing live broadcast image frames of the video live broadcast according to the image capturing frequency corresponding to the second identification mode.
8. The method of claim 7, wherein the determining an image capture frequency to which the second authentication mode corresponds comprises:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode.
9. The method of claim 8, wherein the determining an image capture frequency to which the second authentication mode corresponds comprises:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode and the live video state or the current flow increasing information, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode, and the live video state comprises a white list state, a grey list state, a black list state and an undetermined state.
10. An apparatus for discriminating anomalous live information in a live video, wherein the apparatus comprises:
the switching detection device is used for detecting whether the live video in the first authentication mode meets the triggering condition for switching to the second authentication mode or not according to the change condition of the live video flow corresponding to the live video;
the switching device is used for switching the live video to the second identification mode when the trigger condition is met and the state of the target video is a grey list state, a blacklist state or an undetermined state, capturing live video frames of the live video according to the image capturing frequency corresponding to the second identification mode, wherein the target video is the live video serving as a detection target;
and the abnormity determining device is used for determining whether the video live broadcast contains abnormal live broadcast information according to the live broadcast image frame.
11. The apparatus of claim 10, wherein the anomaly determination device comprises:
the result acquisition unit is used for acquiring a first identification result and a second identification result corresponding to the live broadcast image frame;
and the abnormity determining unit is used for determining whether the video live broadcast contains abnormal live broadcast information according to the first identification result and the second identification result.
12. The apparatus of claim 11, wherein the apparatus is further configured to:
when the first identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state as a grey list state; or
And when the second identification result indicates that the video live broadcast contains abnormal live broadcast information, setting the video live broadcast state to be a blacklist state.
13. The apparatus according to claim 11 or 12, wherein the result obtaining unit comprises:
the first result acquiring subunit is used for acquiring a first identification result corresponding to the live broadcast image frame;
and the second result obtaining subunit is configured to obtain a second identification result corresponding to the live broadcast image frame when the first identification result indicates that the live broadcast of the video contains abnormal live broadcast information.
14. The apparatus of claim 10, wherein the switching means is further for:
and when the trigger condition is met and the state of the target video is a white list state, keeping the live video in the first identification mode.
15. The device of claim 10, wherein the trigger condition comprises at least any one of:
live broadcast flow of the video live broadcast exceeds a preset live broadcast flow threshold;
the flow rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow rate threshold value;
live broadcast flow of the live video exceeds a preset live broadcast flow upper limit threshold of the first authentication mode;
the flow increase rate of live broadcast flow of the video live broadcast exceeds a preset live broadcast flow increase rate lower limit threshold of the first identification mode;
live streaming of the live video exceeds a predetermined live streaming threshold of the second authentication mode;
the rate of flow increase of live traffic of the live video exceeds a predetermined live traffic increase rate threshold of the second authentication mode.
16. The apparatus of claim 10, wherein the switching means comprises:
the switching unit is used for switching the live video to the second authentication mode when the triggering condition is met;
the frequency determining unit is used for determining the image capturing frequency corresponding to the second identification mode;
and the capturing unit is used for capturing the live broadcast image frames of the video live broadcast according to the image capturing frequency corresponding to the second identification mode.
17. The device of claim 16, wherein the frequency determination unit is to:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode.
18. The device of claim 17, wherein the frequency determination unit is to:
and determining the image capturing frequency corresponding to the second identification mode according to the image capturing frequency corresponding to the first identification mode and the live video state or the current flow increasing information, wherein the image capturing frequency corresponding to the second identification mode is greater than the image capturing frequency corresponding to the first identification mode, and the live video state comprises a white list state, a grey list state, a black list state and an undetermined state.
CN201610430465.XA 2016-06-16 2016-06-16 Method and equipment for identifying abnormal live broadcast information in video live broadcast Active CN107517391B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610430465.XA CN107517391B (en) 2016-06-16 2016-06-16 Method and equipment for identifying abnormal live broadcast information in video live broadcast

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610430465.XA CN107517391B (en) 2016-06-16 2016-06-16 Method and equipment for identifying abnormal live broadcast information in video live broadcast

Publications (2)

Publication Number Publication Date
CN107517391A CN107517391A (en) 2017-12-26
CN107517391B true CN107517391B (en) 2021-04-09

Family

ID=60720616

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610430465.XA Active CN107517391B (en) 2016-06-16 2016-06-16 Method and equipment for identifying abnormal live broadcast information in video live broadcast

Country Status (1)

Country Link
CN (1) CN107517391B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110505490A (en) * 2018-05-18 2019-11-26 阿里巴巴集团控股有限公司 Multi-medium data recognition methods and system
CN108881938B (en) * 2018-08-02 2021-05-28 广东龙眼数字科技有限公司 Live broadcast video intelligent cutting method and device
CN110839010B (en) * 2018-08-17 2023-03-31 阿里巴巴集团控股有限公司 Streaming media data processing method, device and equipment
CN109257617B (en) * 2018-09-30 2021-11-09 武汉斗鱼网络科技有限公司 Method for determining suspected user in live broadcast platform and related equipment
CN111314716B (en) * 2018-12-11 2022-06-03 阿里巴巴集团控股有限公司 Live broadcast processing method, device and system and real-time data processing method
CN112800250A (en) * 2021-02-05 2021-05-14 联想(北京)有限公司 Multimedia data stream processing method and device and electronic equipment

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101184241B (en) * 2007-12-03 2010-07-14 中兴通讯股份有限公司 Image automatic detection method and apparatus
CN107707929A (en) * 2011-05-12 2018-02-16 索林科集团 Video analytic system
CN103049456B (en) * 2011-10-14 2016-03-16 腾讯科技（深圳）有限公司 A kind of method and device screening webpage
CN103716690B (en) * 2013-12-27 2017-09-01 广州华多网络科技有限公司 Method, terminal, server and the system of multimedia live broadcast report

Also Published As

Publication number Publication date
CN107517391A (en) 2017-12-26

Similar Documents

Publication Publication Date Title
CN107517391B (en) Method and equipment for identifying abnormal live broadcast information in video live broadcast
EP3321844B1 (en) Action recognition in a video sequence
US10121069B2 (en) Method of switching video images and surveillance system
US9582709B2 (en) Privacy for camera with people recognition
Singh et al. I always feel like somebody's sensing me! A framework to detect, identify, and localize clandestine wireless sensors
US10360456B2 (en) Methods and systems of maintaining lost object trackers in video analytics
WO2013118491A1 (en) Camera device, server device, image monitoring system, control method of image monitoring system, and control program of image monitoring system
US10152630B2 (en) Methods and systems of performing blob filtering in video analytics
WO2018031102A1 (en) Methods and systems of performing content-adaptive object tracking in video analytics
CN101720031A (en) Difference method-based indoor video monitoring and alarming method
JP2006011728A (en) Suspicious person countermeasure system and suspicious person detection device
CN107330414A (en) Act of violence monitoring method
WO2018018774A1 (en) Remote monitoring method and device, and storage medium
WO2018032270A1 (en) Low complexity tamper detection in video analytics
EP2770740A1 (en) System and method for detecting motion in compressed video
CN105872466A (en) Alarm method and alarm device based on smart TV
WO2018076508A1 (en) Method and apparatus for security monitoring of greenhouse
CN108597168A (en) Security alarm method, apparatus based on optical filter and household appliance
CN105450992A (en) Smart home monitoring method and system based on motion detection
CN113706573B (en) Method and device for detecting moving object and storage medium
US20160100160A1 (en) Blurry image detecting method and related camera and image processing system
CN112291282A (en) A kind of motion detection alarm method and device
KR20210024935A (en) Apparatus for monitoring video and apparatus for analyzing video, and on-line machine learning method
CN108564750A (en) Security alarm method, apparatus and household appliance
CN105187803B (en) A kind of safe video signal monitoring method based on TV

Legal Events

Date Code Title Description
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant