CN113435414B - Video content auditing system, method and device - Google Patents

Video content auditing system, method and device Download PDF

Info

Publication number
CN113435414B
CN113435414B CN202110873450.1A CN202110873450A CN113435414B CN 113435414 B CN113435414 B CN 113435414B CN 202110873450 A CN202110873450 A CN 202110873450A CN 113435414 B CN113435414 B CN 113435414B
Authority
CN
China
Prior art keywords
auditing
module
video
audit
decoding
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
CN202110873450.1A
Other languages
Chinese (zh)
Other versions
CN113435414A (en
Inventor
郭志全
罗鑫
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.)
Beijing Zhongke Flux Technology Co ltd
Original Assignee
Beijing Zhongke Flux Technology Co 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 Beijing Zhongke Flux Technology Co ltd filed Critical Beijing Zhongke Flux Technology Co ltd
Priority to CN202110873450.1A priority Critical patent/CN113435414B/en
Publication of CN113435414A publication Critical patent/CN113435414A/en
Application granted granted Critical
Publication of CN113435414B publication Critical patent/CN113435414B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/71Indexing; Data structures therefor; Storage structures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/75Clustering; Classification

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

The invention provides a video content auditing system, method and device, wherein the system comprises the following steps: the web service module performs data interaction with the video storage module and the configuration module; the web service module receives the video file to be checked, sends the video file to the video storage module, and sends the configuration information to the configuration module; the video storage module is also used for carrying out data interaction with the decoding module and sending the video file to the decoding module; the configuration module is also in data interaction with the auditing module and is used for analyzing the configuration information and notifying the auditing module of the dynamic adjustment function; the decoding module is also in data interaction with the auditing system and is used for decoding the received video file and outputting a video frame sequence to the auditing module; the auditing module is used for auditing the received video frame sequence. The scheme has high processing efficiency and strong expansibility, and can greatly reduce the subsequent development and maintenance cost of the system.

Description

Video content auditing system, method and device
Technical Field
The invention relates to the field of computer body image recognition, in particular to a method, a system and a device for recognizing and auditing related computer video image contents.
Background
With the development of internet broadband technology, numerous and various audio and video platforms have appeared in the network. In recent years, video sharing websites with videos as main information carriers are also rapidly developed, and on the mainstream video sharing websites, the video click rate per day can reach hundreds of millions, and the uploading amount of videos can reach tens of thousands of segments. Users can release audio and video through the platforms, the quality of the video is good, the platforms need to carry out auditing before release, low-quality unreasonable video is filtered, namely, video auditing needs to consume a lot of time, manpower and material resources, and the video auditing can not adapt to the development of society at all only by means of manual auditing, and computer video auditing occurs.
The auditing problem of the video platform is solved through a computer, and the platform can be naturally helped to solve the pain point of huge manpower consumption.
Because the content auditing is to label and classify the data, for audio content, especially for picture content, the conventional thinking is to adopt a classification algorithm to identify whether the picture is pornographic or not, adopt target detection to identify whether the face is political figure or not, identify the text content on the picture through OCR, and then pass the text content through a sensitive word and semantic analysis model so as to identify whether the content is illegal or not.
However, in the prior art, the current auditing function is realized in a single content auditing mode, and no better method is provided for integrating and processing various types of content auditing so as to realize an efficient pipeline processing mode, which influences the further application of the identification technology in audio and video content auditing to a certain extent.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a video content auditing system, a method and a device, which realize the pipelined processing of multiple categories of video contents in the same system in a low-coupling and modularized mode.
Specifically, the invention provides the following technical scheme:
In one aspect, the present invention provides a video content auditing system, the system comprising:
The system comprises a Web service module, a video storage module, a decoding module, a configuration module and an auditing module;
The web service module performs data interaction with the video storage module and the configuration module; the web service module receives the video file to be checked, sends the video file to the video storage module, and is also used for sending the configuration information to the configuration module;
The video storage module is also used for carrying out data interaction with the decoding module and sending the video file to the decoding module;
The configuration module is also in data interaction with the auditing module and is used for analyzing the configuration information and notifying the auditing module of the dynamic adjustment function;
the decoding module is also in data interaction with the auditing system and is used for decoding the received video file and outputting a video frame sequence to the auditing module;
The auditing module is used for auditing the received video frame sequence.
Preferably, the web service also performs data interaction with the database, and transmits the auditing rules and auditing tasks to the database.
Preferably, the auditing module performs data interaction with the database, acquires auditing rules and auditing tasks from the database, and sends auditing results to the database.
Preferably, in the auditing module, an assembly line mode is adopted for auditing the video frame sequence; the auditing module comprises a plurality of auditing units, wherein the auditing units are provided with unified interfaces, and the interfaces are used for carrying out a lower package recognition algorithm and adding a layer of package to support multiple language realization;
preferably, the configuration information is used for defining the number of auditing pipelines, the order of auditing units and the number of auditing units in the auditing module.
Preferably, the same type of auditing unit is arranged in a pipeline in 1 or more.
Preferably, the auditing unit comprises an input port and an output port; one of the input ports can only bind one of the output ports; the same output port can bind one or more of the input ports.
Preferably, in the audit rule setting, two types may be set:
Setting an audit confidence threshold, and considering that the audit is passed when the audit result of the audit unit is passed and the audit confidence is greater than the audit confidence threshold; and/or
Setting a blacklist, and if the content in the blacklist appears in the video frame, checking is not passed.
In addition, the invention also provides a video content auditing method, which is applied to the system, and comprises the following steps:
s1, establishing an auditing rule through a web service module, and setting configuration information to perform auditing assembly line configuration;
s2, creating and storing a unique identifier for the input video file to be checked, and creating a checking task;
S3, the decoding module decodes the video file to be audited, and inputs the decoded video frame into the auditing module for pipeline auditing;
And S4, after the auditing is completed, transmitting the auditing result to a database for storage based on the unique identifier.
Preferably, the auditing rule includes:
Setting an audit confidence threshold, and considering that the audit is passed when the audit result of the audit unit is passed and the audit confidence is greater than the audit confidence threshold; and/or
Setting a blacklist, and if the content in the blacklist appears in the video frame, checking is not passed.
In addition, the invention also provides a video content auditing device, which comprises a processor and a storage device, wherein the processor calls instructions in the storage device to execute the video content auditing method.
Compared with the prior art, the auditing module based on the assembly line has high processing efficiency and strong expansibility, can reduce the cost of video auditing in the actual production environment, and in addition, the requirements and the requirements for video auditing can change along with the time.
Drawings
FIG. 1 is a block diagram of a video content auditing system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a video content auditing function module pipeline according to an embodiment of the present invention;
fig. 3 is a schematic diagram of the interior of the pipeline_2 module according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without any inventive effort, are intended to be within the scope of the invention.
In a specific embodiment, the system architecture of the present invention may take the form of that shown in fig. 1.
The system mainly comprises a web service, a configuration system, a video storage system, a decoding system and an auditing system.
An external web network or mobile terminal (mobile) interacts with the web service module by sending a request.
The Web service is mainly used for processing requests from a browser and a mobile terminal APP, and comprises video file uploading, auditing tasks, auditing rule submitting, configuration information submitting and user rights management.
The video storage system mainly stores video files uploaded by system users.
The decoding system is mainly used for decoding the video file and outputting the video frame sequence to the auditing system.
The configuration system is mainly used for analyzing the configuration information submitted by the user and informing the auditing system of the dynamic adjustment function.
The auditing system is mainly used for auditing the video frame sequence.
In the invention, the auditing of the video frame sequence adopts a pipeline mode, and the working process of the pipeline is further described with reference to fig. 1. And (5) sequentially sending the image frames obtained after the video is decoded into an auditing system, and starting the pipeline to work. The video frame enters an auditing unit for detection, if the video frame passes the detection, the next auditing unit is entered for continuous detection, otherwise, the whole assembly line work is stopped, and the auditing is not passed; if all frames pass each auditing unit, the auditing passes.
The auditing rules defined by the invention are mainly two types, namely an artificial intelligence auditing result self-reliability threshold and a blacklist, such as pornography detection, wherein the artificial intelligence auditing result is passing, and the auditing is considered to pass when the auditing self-reliability is greater than a preset threshold, or else, the auditing is considered to not pass; and detecting the person, and if the person is detected to appear in the person blacklist, checking is not passed.
The auditing unit adopts an artificial intelligence algorithm or an image recognition algorithm, and the auditing unit defined in the invention provides a unified interface, and in a preferred embodiment, the interface can be set as follows:
Detect (image, result)
The interface is packaged through a dynamic link library in the C++/C language, and only a link file is required to be updated thermally when the algorithm is upgraded for the lower packaging artificial intelligent algorithm, so that the upper system does not need to pay attention to the realization of the lower layer; and adding a layer of encapsulation, supporting Python call through swig encapsulation, supporting Java call through JNI encapsulation, supporting Golang call through CGO encapsulation, supporting realization of multiple high-level languages, and increasing the expansibility of the system.
The configuration information defined by the invention mainly defines the number of auditing pipelines in the auditing system, and the order of auditing units and the number of auditing units in the pipelines. The order of the auditing units can be set according to practical application, and in order to save resources, a module with accurate detection results can be placed at the head of the assembly line. The number of the auditing units is set according to the capability of the machine, the frame loss processing in the conventional detection is not adopted in the system, the performance requirement of the system on the machine is higher, the complexity is high, the number of the auditing units on the assembly line can be properly increased by the auditing units with long operation time, and the efficiency of the whole assembly line is improved.
In a more specific embodiment, the configuration information definition may be set to pipeline name= (number of pipelines, module definition)
The auditing units are defined as auditing units=auditing unit number, the auditing units are divided by I, and the sequence of each auditing unit in the definition is the sequence of processing on the production line.
For example:
pipeline_1= (1, pornography=1|person=1|sensitivity=1|riot=1|watermark=1.)
The meaning of the above example is: 1 pipeline named pipeline_1 is started, and the pipeline sequentially comprises pornography/personage/sensitivity/riot/watermark audit and is 1 module.
Pipyline_2= (2, sensitivity=1|person=3|riot=1|watermark=1)
The meaning of the above example is: 2 pipelines named pipeline_2 are started, and the pipeline sequentially comprises sensitive/persona/riot/watermark auditing, wherein the persona comprises 3 auditing units.
Taking pipeline_2 as an example, the details of the pipeline interior are shown in fig. 3, and each auditing unit is defined to have 2 ports, namely an output port and an input port, which are in a binding relationship.
One input port can only be bound on one output port, a plurality of input ports can be bound on the same output port, the configuration of pipeline prescribes the binding sequence of the ports of the auditing unit, and the image frame information is transmitted to the input port of the next-stage auditing unit through the output port of the auditing unit.
The design benefit based on port binding is that the load on the next-stage auditing unit is visible to the previous-stage auditing unit, the output port can acquire the state of the input port auditing unit through binding, and when a new image needs to be transmitted, the next-stage auditing unit can be determined according to the state, so that a better load is achieved.
Referring to fig. 2, the main flow of the video auditing system of the present invention is as follows:
1. After logging in by the browser, the user creates an auditing rule, carries out auditing pipeline configuration according to system prompt, and initializes the system.
2. The user uploads the video file to the rear end of the server, the rear end creates a unique identifier UUID for the file, stores the UUID, and automatically creates an audit task.
3. The decoding system decodes the video file and sends the decoded video frame to the auditing system, and the auditing system pipeline starts to work.
4. And after the auditing is finished, writing the result into a database for persistent storage according to the UUID of the video.
In another specific embodiment, the technical solution of the present invention may also be implemented by a video content auditing apparatus, where the apparatus may include a processor, a storage device, and necessary device operation components, where the processor invokes instructions in the storage device to execute a video content auditing method as described above. Or the device can be arranged to accommodate the video content auditing system as described above, thereby realizing auditing of the video content with the technical scheme of the invention as the core.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random-access Memory (Random Access Memory, RAM), or the like.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. A video content auditing system, the system comprising:
The system comprises a web service module, a video storage module, a decoding module, a configuration module and an auditing module;
The web service module performs data interaction with the video storage module and the configuration module; the web service module receives the video file to be checked, sends the video file to the video storage module, and is also used for sending the configuration information to the configuration module;
The video storage module is also used for carrying out data interaction with the decoding module and sending the video file to the decoding module;
The configuration module is also in data interaction with the auditing module and is used for analyzing the configuration information and notifying the auditing module of the dynamic adjustment function;
The decoding module is also in data interaction with the auditing module and is used for decoding the received video file and outputting a video frame sequence to the auditing module;
The auditing module is used for auditing the received video frame sequence;
The configuration information is used for defining the number of auditing pipelines, the order of auditing units and the number of auditing units in the auditing module;
in the auditing module, the auditing of the video frame sequence adopts a pipeline mode; the auditing module comprises a plurality of auditing units, wherein the auditing units are provided with unified interfaces, and the interfaces are used for identifying an algorithm by a lower package and a layer of package to support multiple languages.
2. The system of claim 1, wherein the web service module further interacts with the database and transmits audit rules, audit tasks to the database.
3. The system of claim 1, wherein the auditing module further interacts with a database, and wherein the auditing module obtains auditing rules and auditing tasks from the database and sends auditing results to the database.
4. The system of claim 1, wherein the audit units of the same type are arranged in a pipeline in a number of 1 or more.
5. The system of claim 1, wherein the auditing unit includes an input port, an output port; one of the input ports can only bind one of the output ports; the same output port can bind one or more of the input ports.
6. A method of auditing video content, the method being applied to the system of any one of claims 1-5, the method comprising:
s1, establishing an auditing rule through a web service module, and setting configuration information to perform auditing assembly line configuration;
s2, creating and storing a unique identifier for the input video file to be checked, and creating a checking task;
S3, the decoding module decodes the video file to be audited, and inputs the decoded video frame into the auditing module for pipeline auditing;
And S4, after the auditing is completed, transmitting the auditing result to a database for storage based on the unique identifier.
7. The method of claim 6, wherein the auditing rules include:
Setting an audit confidence threshold, and considering that the audit is passed when the audit result of the audit unit is passed and the audit confidence is greater than the audit confidence threshold; and/or
Setting a blacklist, and if the content in the blacklist appears in the video frame, checking is not passed.
8. A video content auditing apparatus, characterized in that the apparatus comprises a processor, a storage device, the processor invoking instructions in the storage device to perform the video content auditing method of any of claims 6-7.
CN202110873450.1A 2021-07-30 2021-07-30 Video content auditing system, method and device Active CN113435414B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110873450.1A CN113435414B (en) 2021-07-30 2021-07-30 Video content auditing system, method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110873450.1A CN113435414B (en) 2021-07-30 2021-07-30 Video content auditing system, method and device

Publications (2)

Publication Number Publication Date
CN113435414A CN113435414A (en) 2021-09-24
CN113435414B true CN113435414B (en) 2024-04-26

Family

ID=77762542

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110873450.1A Active CN113435414B (en) 2021-07-30 2021-07-30 Video content auditing system, method and device

Country Status (1)

Country Link
CN (1) CN113435414B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103188530A (en) * 2011-12-30 2013-07-03 大连天维科技有限公司 Hierarchical content audit system
CN111901608A (en) * 2020-06-28 2020-11-06 上海文化广播影视集团有限公司 Video program film examination system and method
CN112153378A (en) * 2020-09-27 2020-12-29 福建天晴数码有限公司 Method and system for testing video auditing capability
CN112351337A (en) * 2021-01-04 2021-02-09 腾讯科技(深圳)有限公司 Video quality inspection method and device, computer equipment and storage medium

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103188530A (en) * 2011-12-30 2013-07-03 大连天维科技有限公司 Hierarchical content audit system
CN111901608A (en) * 2020-06-28 2020-11-06 上海文化广播影视集团有限公司 Video program film examination system and method
CN112153378A (en) * 2020-09-27 2020-12-29 福建天晴数码有限公司 Method and system for testing video auditing capability
CN112351337A (en) * 2021-01-04 2021-02-09 腾讯科技(深圳)有限公司 Video quality inspection method and device, computer equipment and storage medium

Also Published As

Publication number Publication date
CN113435414A (en) 2021-09-24

Similar Documents

Publication Publication Date Title
CA2865187C (en) Method and system relating to salient content extraction for electronic content
WO2016173200A1 (en) Malicious website detection method and system
US9398117B2 (en) Protocol data unit interface
CN100518183C (en) Server equipment and server method
CN102664967A (en) Method and system for interacting cross-platform personal information and background server
US11294740B2 (en) Event to serverless function workflow instance mapping mechanism
CN113568626B (en) Dynamic packaging and application package opening method and device and electronic equipment
CN109145040A (en) A kind of data administering method based on double message queues
CN111314388A (en) Method and apparatus for detecting SQL injection
CN115357663A (en) Data synchronization method, system and device based on incremental data synchronization component
CN102203734A (en) Conditional processing method and apparatus
CN112866279B (en) Webpage security detection method, device, equipment and medium
CN113435414B (en) Video content auditing system, method and device
CN113485607A (en) Public general video capturing method
US20130230248A1 (en) Ensuring validity of the bookmark reference in a collaborative bookmarking system
CN111198983A (en) Sensitive information detection method, device and storage medium
CN113064987A (en) Data processing method, apparatus, electronic device, medium, and program product
CN113742235A (en) Method and device for checking codes
CN108920589B (en) Browsing hijacking identification method, device, server and storage medium
CN115730104A (en) Live broadcast room processing method, device, equipment and medium
CN112487218A (en) Content processing method, system, device, computing equipment and storage medium
US10963510B2 (en) Dynamic natural language processing tagging
CN113050987A (en) Interface document generation method and device, storage medium and electronic equipment
CN116887197B (en) Bank based on 5G message service processing method and device
CN110908958B (en) File processing method, device, terminal and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB02 Change of applicant information
CB02 Change of applicant information

Address after: Room 711c, 7 / F, block a, building 1, yard 19, Ronghua Middle Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing 102600

Applicant after: Beijing Zhongke Flux Technology Co.,Ltd.

Address before: Room 711c, 7 / F, block a, building 1, yard 19, Ronghua Middle Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing 102600

Applicant before: Beijing Ruixin high throughput technology Co.,Ltd.

GR01 Patent grant
GR01 Patent grant