CN108024148A - The multimedia file recognition methods of Behavior-based control feature, processing method and processing device - Google Patents

The multimedia file recognition methods of Behavior-based control feature, processing method and processing device Download PDF

Info

Publication number
CN108024148A
CN108024148A CN201610929276.7A CN201610929276A CN108024148A CN 108024148 A CN108024148 A CN 108024148A CN 201610929276 A CN201610929276 A CN 201610929276A CN 108024148 A CN108024148 A CN 108024148A
Authority
CN
China
Prior art keywords
multimedia file
video
certain content
wish
value
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.)
Granted
Application number
CN201610929276.7A
Other languages
Chinese (zh)
Other versions
CN108024148B (en
Inventor
刘杰
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.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen 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 Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN201610929276.7A priority Critical patent/CN108024148B/en
Priority to PCT/CN2017/104275 priority patent/WO2018068664A1/en
Publication of CN108024148A publication Critical patent/CN108024148A/en
Priority to US16/026,786 priority patent/US10805255B2/en
Application granted granted Critical
Publication of CN108024148B publication Critical patent/CN108024148B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
    • H04N21/44204—Monitoring of content usage, e.g. the number of times a movie has been viewed, copied or the amount which has been watched
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/7837—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using objects detected or recognised in the video content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/4508—Management of client data or end-user data
    • H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • General Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Information Transfer Between Computers (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

The present invention relates to Internet technical field, more particularly to a kind of multimedia file recognition methods of Behavior-based control feature, processing method and processing device.The internet behavior of present invention analysis user is associated with viewing certain content, it is proposed in multimedia file playing process, by obtaining the Figure Characteristics value of spectators user and wishing to watch the first wish characteristic value of certain content for representing user in preset time, and then the probability that the multimedia file includes certain content is calculated according to the Figure Characteristics value of each user and the first wish characteristic value, the probability is compared to determine the need for further detecting the multimedia file with preset value, multimedia file to be analyzed is obtained from there through user behavior feature assisting sifting, certain content detection is carried out to the multimedia file screened, improve the recognition efficiency and accuracy of multimedia file.The present invention is used for pornographic video detection and can greatly improve verification and measurement ratio and reliability, facilitates the management and control of multimedia video.

Description

The multimedia file recognition methods of Behavior-based control feature, processing method and processing device
Technical field
The present invention relates to Internet technical field, more particularly to a kind of multimedia file identification side of Behavior-based control feature Method, processing method and processing device.
Background technology
With the rapid development of internet and multimedia technology, video media is ubiquitous in network, becomes us and gives birth to Amusement important component living.But a large amount of pornographic videos are wherein full of, caused to Social Culture life extremely serious negative Face is rung.How the propagation of network on pornographic video is effectively prevented to safeguarding teen-age physically and mentally healthy and purification Internet video money Source is of great significance.
In the past few years, pornographic video detecting method continues to bring out.These methods are divided into three classes：The first kind is to extract The key frame of video, Pornograph detection is carried out to key frame；Second class is the audio extracted in video, and audio content is carried out Identification；Three classes obtain the motion vector of video, and motion feature is analyzed.But since pornographic video is relative to normal video It is maximum difference lies in：Containing a large amount of exposed skins in pornographic video frame.Therefore, compared with other two classes methods, first kind side Method has higher accuracy.Pornograph inspection just is carried out to key frame using multiple key frames in extraction video in existing patent Survey, according to the testing result to single key frame Pornograph, together with the testing result of other key frames obtained before, carry out Decision fusion judges.But first kind method is there is also shortcoming, i.e.,：If Pornograph only accounts for the smaller portions in a length section video When, the possibility for identifying Pornograph of this method will reduce, in addition, to know in this way in massive video Do not go out pornographic video, its recognition efficiency is very low.
The content of the invention
Inventor is had found by studying：The internet behavior of user can reflect the preference of user, for example have deep love for American series User can generally go to browse the American series plate of various film review websites, and it is also mostly to see American series to see video；And the use of preference Pornograph Family can then pay close attention to various adult web sites, pornographic forum etc., see the video that can see mature content during video more.User preference and user Behavior can be directed toward the video of certain content, such as after a user is browsing pornographic text or porny, then go Watch video or live, then this video or the live possibility comprising Pornograph are with regard to larger；A meanwhile if video There are 10 people watching, wherein have user or just browsed pornographic text or the porny that 8 people are preference Pornograph, The possibility that this video includes Pornograph is very big.
Associating based on user preference and user behavior and certain content video, can be by the behavior of user on the internet It is added to preference as an auxiliary dimension in Pornograph identification, more effectively identifies whether multimedia file includes color Feelings content；Based on this, the present invention provides a kind of multimedia file recognition methods of Behavior-based control feature, processing method and processing device.
In a first aspect, the present invention provides a kind of multimedia file recognition methods of Behavior-based control feature, including：
In multimedia file playing process, the Figure Characteristics value and the first wish characteristic value of spectators user are obtained, it is described Figure Characteristics value is used to identify hobby of the user to certain content, and the first wish characteristic value is used to identifying user when default Between wish in section to watch the wish of certain content；
The multimedia file is calculated according to the Figure Characteristics value and the first wish characteristic value and includes the general of certain content Rate；
Judge whether the probability exceedes preset value, if so, then carrying out feature detection to the multimedia file；
According to feature testing result judge the multimedia file whether be certain content multimedia file.
Preferably, it is described that the multimedia file is calculated comprising special according to the Figure Characteristics value and the first wish characteristic value Determine the probability of content, including：
The second wish characteristic value of each user is determined according to the Figure Characteristics value and the first wish characteristic value；
The probability of certain content is included according to the second wish characteristic value of all users calculating multimedia file.
Preferably, it is described to determine that the second wish of each user is special according to the Figure Characteristics value and the first wish characteristic value Value indicative, including：Sum to the Figure Characteristics value and the first wish characteristic value, obtain the second wish characteristic value, or Person, according to the weight set in advance for Figure Characteristics value and the first wish characteristic value, to the Figure Characteristics value and the first wish Characteristic value is weighted averagely, obtains the second wish characteristic value.
Preferably, it is described that certain content is included according to the second wish characteristic value of all users calculating multimedia file Probability, including：The second wish characteristic value of each user is compared with default threshold value respectively；Calculate second meaning It is willing to that characteristic value exceedes the number of users of threshold value and the ratio of total number of users amount, obtains the multimedia file and include certain content Probability.
Further, before the Figure Characteristics value and the first wish characteristic value of spectators user is obtained, further include：Analysis is used The behavioral data at family, determines the Figure Characteristics value and the first wish characteristic value of user.
Preferably, the behavioral data of the analysis user, determines the Figure Characteristics value of user, including：Obtain the row of user For data, the behavioral data includes browsing the first behavioral data of certain content related text, browses the related figure of certain content Second behavioral data of piece, access the third line of certain content correlation forum for data and in the relevant chat group of certain content The fourth line of chat is data；Judge first behavioral data, the second behavioral data, the third line for data and fourth line respectively Whether it is empty for data, correspondence obtains the first judging result, the second judging result, the 3rd judging result and the 4th judging result； According to the first weight of first judging result set in advance, the second weight of second judging result, the described 3rd 3rd weight of judging result and the 4th weight of the 4th judging result, judge to tie to first judging result, second Fruit, the 3rd judging result and the 4th judging result are allocated integration, obtain the behavioural characteristic value of the user.
Preferably, the behavioral data of the analysis user, determines the first wish characteristic value of user, including：Obtain user Behavioral data within nearest a period of time, the behavioral data include browsing first time, clear of certain content related text Look at certain content picture concerned the second time, access the 3rd time of certain content correlation forum and relevant in certain content The 4th time chatted in chat group；First weight is assigned for the first time, second time assigns described the Two weights, the 3rd time assign the 3rd weight, the 4th time assigns the 4th weight, during to described first Between, the second time, the 3rd time and the 4th time be weighted average, obtain the wish characteristic value of user.
Preferably, the multimedia file is video；It is described to the multimedia file carry out feature detection, including：It is right Video constant duration extracts the image of preset number；Feature detection is carried out to every image, judges whether described image includes Special characteristic, the feature detection include sensitive part detection and skin pixel detection.
Preferably, it is described according to feature testing result judge the multimedia file whether be certain content multimedia text Part, including：When judging that the picture number comprising special characteristic is more than threshold value P set in advance, judge the video for spy Determine audio content, otherwise judge the video for normal video；Or, determine the picture number comprising special characteristic judged with The total number of images purpose ratio extracted for the video detection, when definite ratio is more than threshold value Q, judges the video For certain content video, otherwise judge the video for normal video.
Preferably, the certain content is Pornograph, and the video is order video or live video.
Second aspect, the present invention provide a kind of multimedia file processing method, including：
In multimedia file playing process, the Figure Characteristics value and the first wish characteristic value of spectators user are obtained, it is described Figure Characteristics value is used to identify hobby of the user to certain content, and the first wish characteristic value is used to identifying user when default Between wish in section to watch the wish of certain content；
The multimedia file is calculated according to the Figure Characteristics value and the first wish characteristic value and includes the general of certain content Rate；
Judge whether the probability exceedes preset value, if so, then carrying out feature detection to the multimedia file；
According to feature testing result judge the multimedia file whether be certain content multimedia file；
The multimedia file is handled according to the judging result.
Preferably, the multimedia file is order video or live video, and the certain content is Pornograph；It is described Carrying out processing to the multimedia file according to the judging result includes：If multimedia file is the pornographic video of program request, Exit the broadcasting of order video；If multimedia file is live pornographic video, close between the net cast for playing the video.
The third aspect, the present invention provide a kind of multimedia file identification device of Behavior-based control feature, including：
Acquiring unit, in multimedia file playing process, obtaining the Figure Characteristics value and the first meaning of spectators user It is willing to characteristic value, the Figure Characteristics value is used to identify hobby of the user to certain content, and the first wish characteristic value is used to mark Know user to wish to watch the wish of certain content in preset time period；
Computing unit, includes for calculating the multimedia file according to the Figure Characteristics value and the first wish characteristic value The probability of certain content；
Detection unit, for judging whether the probability exceedes preset value, if so, then being carried out to the multimedia file special Sign detection；
Determination unit, for according to feature testing result judge the multimedia file whether be certain content multimedia File.
Preferably, the computing unit includes：First computation subunit, for being anticipated according to the Figure Characteristics value and first It is willing to that characteristic value determines the second wish characteristic value of each user；Second computation subunit, for according to the second of all users the meaning It is willing to that characteristic value calculates the probability that the multimedia file includes certain content.
Preferably, first computation subunit includes：First computing module, for the Figure Characteristics value and first Wish characteristic value is summed, and obtains the second wish characteristic value；Second computing module, is in advance Figure Characteristics for basis Value and the weight of the first wish characteristic value setting, are weighted the Figure Characteristics value and the first wish characteristic value averagely, obtain To the second wish characteristic value.
Preferably, second computation subunit includes：Comparing module, for respectively that the second wish of each user is special Value indicative is compared with default threshold value；Probability evaluation entity, the use of threshold value is exceeded for calculating the second wish characteristic value The ratio of amount amount and total number of users amount, obtains the probability that the multimedia file includes certain content.
Further, described device further includes：Pretreatment unit, for analyzing the behavioral data of user, determines user's Figure Characteristics value and the first wish characteristic value.
Preferably, the pretreatment unit includes：
First processing subelement, is used for：The behavioral data of user is obtained, the behavioral data includes browsing certain content phase First behavioral data of pass text, the second behavioral data for browsing certain content picture concerned, access certain content correlation forum The third line be data and the fourth line chatted is data in the relevant chat group of certain content；The first row is judged respectively For data, the second behavioral data, the third line be data and fourth line be data whether be it is empty, it is corresponding obtain the first judging result, Second judging result, the 3rd judging result and the 4th judging result；According to the first of first judging result set in advance Weight, the second weight of second judging result, the 3rd weight of the 3rd judging result and the 4th judging result The 4th weight, first judging result, the second judging result, the 3rd judging result and the 4th judging result are allocated Integrate, obtain the behavioural characteristic value of the user；
Second processing subelement, is used for:Obtain behavioral data of the user within nearest a period of time, the behavioral data bag The second time for include and browse the first time of certain content related text, browsing certain content picture concerned, access certain content 3rd time of related forum and the 4th time chatted in the relevant chat group of certain content；Assigned for the first time First weight, second time assign second weight, the 3rd time assigns the 3rd weight, described the Four times assigned the 4th weight, the first time, the second time, the 3rd time and the 4th time were weighted flat , the wish characteristic value of user is obtained.
Preferably, the multimedia file is video；The detection unit includes：Subelement is extracted, for video etc. Time interval extracts the image of preset number；Detection sub-unit, for carrying out feature detection to every image, judges described image Whether special characteristic is included, and the feature detection includes sensitive part detection and skin pixel detection.
Preferably, the determination unit includes：First determination subelement, in the image for judging to include special characteristic When number is more than threshold value P set in advance, the video is judged for certain content video, otherwise judges the video normally to regard Frequently；Second determination subelement, the picture number comprising special characteristic for determining to judge are extracted with being directed to the video detection Obtained total number of images purpose ratio, when definite ratio is more than threshold value Q, judges the video for certain content video, no Then judge the video for normal video.
Preferably, the certain content is Pornograph, and the video is order video or live video.
Fourth aspect, the present invention provide a kind of multimedia file processing unit, and described device includes：
Acquiring unit, in multimedia file playing process, obtaining the Figure Characteristics value and the first meaning of spectators user It is willing to characteristic value, the Figure Characteristics value is used to identify hobby of the user to certain content, and the first wish characteristic value is used to mark Know user to wish to watch the wish of certain content in preset time period；
Computing unit, includes for calculating the multimedia file according to the Figure Characteristics value and the first wish characteristic value The probability of certain content；
Detection unit, for judging whether the probability exceedes preset value, if so, then being carried out to the multimedia file special Sign detection；
Determination unit, for according to feature testing result judge the multimedia file whether be certain content multimedia File；
Processing unit, for being handled according to the judging result the multimedia file.
Preferably, the multimedia file is order video or live video, and the certain content is Pornograph；It is described Processing unit, specifically for when definite multimedia file is the pornographic video of program request, exiting the broadcasting of order video；Determining When multimedia file is live pornographic video, close between the net cast for playing the video.
The present invention has the advantages that：
The internet behavior of present invention analysis user is associated with viewing certain content, is proposed in multimedia file playing process In, by obtaining the Figure Characteristics value of spectators user and wishing to watch the of certain content for representing user in preset time One wish characteristic value, and then the multimedia file bag is calculated according to the Figure Characteristics value of each user and the first wish characteristic value Probability containing certain content, and the probability is compared to determine the need for further detecting the multimedia with preset value File, multimedia file to be analyzed is obtained from there through user behavior feature assisting sifting, to the multimedia text screened Part carries out certain content detection, improves the recognition efficiency and accuracy to the multimedia file of certain content.The present invention is used In the detection of the harmful contents such as pornographic, terror in multimedia file, detection efficiency and reliability can be greatly improved, it is convenient The management and control of multimedia file.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art and advantage, below will be to implementing Example or attached drawing needed to be used in the description of the prior art are briefly described, it should be apparent that, drawings in the following description are only Only it is some embodiments of the present invention, for those of ordinary skill in the art, without creative efforts, Other attached drawings can also be obtained according to these attached drawings.
Fig. 1 is the computer that can be used for implementing the multimedia file recognition methods of the Behavior-based control feature of the embodiment of the present invention The hardware block diagram of terminal；
Fig. 2 is the flow chart of the multimedia file recognition methods for the Behavior-based control feature that the embodiment of the present invention 1 discloses；
Fig. 3 is the flow chart of the multimedia file recognition methods for the Behavior-based control feature that the embodiment of the present invention 2 discloses；
Fig. 4 is the flow chart for the multimedia file processing method that the embodiment of the present invention 3 discloses；
Fig. 5 is the schematic diagram of the multimedia file identification device for the Behavior-based control feature that the embodiment of the present invention 4 discloses；
Fig. 6 is the schematic diagram of the multimedia file identification device for the Behavior-based control feature that the embodiment of the present invention 4 discloses；
Fig. 7 is the schematic diagram for the multimedia file processing unit that the embodiment of the present invention 5 discloses；
Fig. 8 is the structure diagram of terminal according to embodiments of the present invention.
Embodiment
In order to make those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention Attached drawing, is clearly and completely described the technical solution in the embodiment of the present invention, it is clear that described embodiment is only The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people Member's all other embodiments obtained without making creative work, should all belong to the model that the present invention protects Enclose.
It should be noted that term " first " in description and claims of this specification and above-mentioned attached drawing, " Two " etc. be for distinguishing similar object, without for describing specific order or precedence.It should be appreciated that so use Data can exchange in the appropriate case, so as to the embodiment of the present invention described herein can with except illustrating herein or Order beyond those of description is implemented.In addition, term " comprising " and " having " and their any deformation, it is intended that cover Cover it is non-exclusive include, be not necessarily limited to for example, containing the process of series of steps or unit, method, system, product or equipment Those steps or unit clearly listed, but may include not list clearly or for these processes, method, product Or the intrinsic other steps of equipment or unit.
Embodiment 1
The present embodiment provides a kind of embodiment of the multimedia file recognition methods of Behavior-based control feature, it is necessary to explanation It is that step shown in the flowchart of the accompanying drawings can perform in the computer system of such as a group of computer-executable instructions, Also, although logical order is shown in flow charts, in some cases, can be with different from order execution herein Shown or described step.
Embodiment of the method provided herein can be in mobile terminal, terminal or similar arithmetic unit Perform.Exemplified by running on computer terminals, Fig. 1 is the multimedia file that can be used for implementing the Behavior-based control feature of the present invention The hardware block diagram of the terminal of recognition methods.As shown in Figure 1, terminal 100 can include one or more (processor 102 can include but is not limited to Micro-processor MCV or programmable logic device to (one is only shown in figure) processor 102 The processing unit of part FPGA etc.), the memory 104 for storing data and the transmitting device 106 for communication function.This Field those of ordinary skill is appreciated that the structure shown in Fig. 1 is only to illustrate, it does not cause the structure of above-mentioned electronic device Limit.For example, terminal 100 may also include more either less components than shown in Fig. 1 or have and Fig. 1 institutes Show different configurations.
Memory 104 can be used for storage application software software program and module, as in the embodiment of the present invention based on Corresponding programmed instruction/the module of multimedia file recognition methods of behavioural characteristic, processor 102 are stored in memory by operation Software program and module in 104, so as to perform various functions application and data processing, that is, realize above-mentioned Behavior-based control The multimedia file recognition methods of feature.Memory 104 may include high speed random access memory, may also include non-volatile memories Device, such as one or more magnetic storage device, flash memory or other non-volatile solid state memories.In some instances, deposit Reservoir 104 can further comprise that relative to the remotely located memory of processor 102, these remote memories network can be passed through It is connected to terminal 10.The example of above-mentioned network includes but not limited to internet, intranet, LAN, mobile logical Letter net and combinations thereof.
Transmitting device 106 is used to data are received or sent via a network.Above-mentioned network instantiation may include The wireless network that the communication providers of terminal 100 provide.In an example, transmitting device 106 includes a network Adapter (Network Interface Controller, referred to as NIC), it can be connected by base station with other network equipments So as to be communicated with internet.In an example, transmitting device 106 can be radio frequency (Radio Frequency, letter Referred to as RF) module, it is used to wirelessly be communicated with internet.
Under above-mentioned running environment, this application provides a kind of multimedia file of Behavior-based control feature as shown in Figure 2 Recognition methods.This method can be applied in intelligent terminal, be performed by the processor in intelligent terminal, intelligent terminal Equipment can be smart mobile phone, tablet computer etc..At least one application program is installed, the present invention is implemented in intelligent terminal The species of example not defining application, can be system class application program, or software class application program.
Fig. 2 is the flow chart of the multimedia file recognition methods for the Behavior-based control feature that the embodiment of the present invention one discloses.Such as Shown in Fig. 2, a kind of optional scheme of this method includes the following steps：
Step S201, in multimedia file playing process, Figure Characteristics value and the first wish for obtaining spectators user are special Value indicative, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value, which is used to identify, to be used Wish to watch the wish of certain content in preset time period in family；
Step S202, the multimedia file is calculated comprising specific according to the Figure Characteristics value and the first wish characteristic value The probability of content；
Step S203, judges whether the probability exceedes preset value, if so, then carrying out feature inspection to the multimedia file Survey；
Step S204, according to feature testing result judge the multimedia file whether be certain content multimedia text Part.
It is described according to the Figure Characteristics value and the first wish characteristic value as a kind of preferred embodiment of step S202 The probability that the multimedia file includes certain content is calculated, including：
The second wish characteristic value of each user is determined according to the Figure Characteristics value and the first wish characteristic value；
The probability of certain content is included according to the second wish characteristic value of all users calculating multimedia file.
The internet behavior of present invention analysis user is associated with viewing certain content, is proposed in multimedia file playing process In, by obtaining the Figure Characteristics value of spectators user and wishing to watch the of certain content for representing user in preset time One wish characteristic value, and then the multimedia file bag is calculated according to the Figure Characteristics value of each user and the first wish characteristic value Probability containing certain content, and the probability is compared to determine the need for further detecting the multimedia with preset value File, multimedia file to be analyzed is obtained from there through user behavior feature assisting sifting, to the multimedia text screened Part carries out certain content detection, improves the recognition efficiency and accuracy to the multimedia file of certain content.The present invention is used In the detection of the harmful contents such as pornographic, terror in multimedia file, detection efficiency and reliability can be greatly improved, it is convenient The management and control of multimedia file.
Embodiment 2
The present embodiment provides a kind of video content recognition method of Behavior-based control feature.In the running environment of such as embodiment 1 Under, this application provides the multimedia file recognition methods of Behavior-based control feature as shown in Figure 3.As shown in figure 3, Fig. 3 is root According to the flow chart of the multimedia file recognition methods of the Behavior-based control feature of the embodiment of the present invention, a kind of optional side of this method Case includes the following steps：
Step 1：The behavioral data of user is analyzed, determines the Figure Characteristics value and the first wish characteristic value of user.
The internet behavior of user can reflect the hobby of user, by analyzing the search of user, browsing, click on recommendation information It can determine that user draws a portrait Deng behavior, such as user's portrait is the pornographic video of hobby, correspondingly, user's portrait can also aid in sentencing The internet behavior of disconnected user's present or future, for example, the pornographic video of hobby user instantly or the pornographic video of following viewing probability Less like user's bigger of pornographic video.User, which draws a portrait, can often reflect a variety of hobbies of user, thus only rely on user Draw a portrait to judge user instantly or following behavior is accurate not enough, it is right since the internet behavior of user often has continuity The search of a certain content or browse often continues for some time, such as user has paid close attention to Pornograph in first few minutes, is working as Preceding or following a period of time continues to browse the probability of pornographic related content with regard to larger, based on this, before may be referred to current time A period in the behavioral characteristic of user carry out auxiliary judgment user instantly or following behavior.
User behavior data is analyzed, hobby of the user to certain content can be identified with Figure Characteristics value, with the first wish Characteristic value mark user pays close attention to the wish of certain content within a period of time before current time.
Wherein, the behavioral data of user is analyzed, determines the Figure Characteristics value of user, including：The behavioral data of user is obtained, The behavioral data includes browsing the first behavioral data of certain content related text, browses the second of certain content picture concerned Behavioral data, access that the third line of certain content correlation forum chats for data and in the relevant chat group of certain content the Four behavioral datas；Judge that first behavioral data, the second behavioral data, the third line are data and fourth line is that data are respectively It is no to be if it is empty then denoted as 0, be then denoted as 1 if not empty for sky, it is corresponding to obtain the first judging result R1, the second judging result R2, the Three judging result R3 and the 4th judging result R4；
According to the first weight W1 of first judging result set in advance, the second weight of second judging result The 4th weight W4 of W3, the 3rd weight W3 of the 3rd judging result and the 4th judging result, judge described first As a result, the second judging result, the 3rd judging result and the 4th judging result are allocated integration, and the behavior for obtaining the user is special Value indicative.As a kind of preferred embodiment, behavioural characteristic value B=W1*R1+W2*R2+W3*R3+W4*R4, as another preferred side Formula, behavioural characteristic value B=(W1*R1+W2*R2+W3*R3+W4*R4)/4.
The behavioral data of user is analyzed, the first wish characteristic value of user is determined, can be realized by two ways：(1) By the similar software such as computer house keeper for being run on user terminal, the screen display content of user is obtained to judge；(2) can be The flow of user, such as packet capturing on router are captured on network, so as to analyze the ongoing operation of user.Specific steps bag Include：Behavioral data of the user within nearest a period of time is obtained, the behavioral data includes browsing certain content related text At the first time, browse the second time of certain content picture concerned, access the 3rd time of certain content correlation forum and in spy Determine the 4th time chatted in the relevant chat group of content；The first weight W1, described second are assigned for the first time Time assigns the second weight W2, the 3rd time assigns the 3rd weight W3, the 4th time assigns described the Four weight W4, are weighted the first time, the second time, the 3rd time and the 4th time average, obtain the of user One wish characteristic value.
Illustrate, it is assumed that certain content is Pornograph, and Figure Characteristics value represents hobby journey of the user to Pornograph Degree, the first wish characteristic value represent that the wish of pornographic video is watched in the hope interior for the previous period of user at the moment, analyze the use The internet behavior at family, mainly include user nearest a period of time whether the relevant word of browsed pornographic, picture and whether Access pornographic relevant forum, whether make a speech in pornographic chats group, wherein, browse erotic novel, pornographic relevant cross-talk or Microblogging etc. can be considered as the browsed relevant word of pornographic, browse be marked as the picture of pornographic, the picture on porn site with And various beauty's pictures on normal website can be considered as the browsed relevant picture of pornographic；Then according to these behavioural characteristics Weight, calculates the Figure Characteristics value of user, and it is 0.4 such as to browse the corresponding weight of pornographic relevant word, is browsed pornographic relevant The corresponding weight of picture is 0.3, and the weight for accessing pornographic forum is 0.6, and the weight made a speech in the pornographic group that chats is 0.5, such as The relevant picture of fruit user browsed pornographic within nearest a period of time, have accessed pornographic relevant forum and also chatted in pornographic Make a speech in its group, then the behavioural characteristic value B=0.4*0+0.6*1+0.3*1+0.5*1=1.4 of the user, according to historical data point Analysis, more than the 1 explanation more concern Pornograph of user, can mark the user as pornographic user.If user is before current time 40 minutes in spend 10 minutes and see erotic novel, see within 10 minutes porny, access within 20 minutes pornographic forum, then the first wish Characteristic value is (0.4*10+0.3*10+0.6*20)/40=0.475.
Step 2：In multimedia file playing process, the Figure Characteristics value and the first wish feature of spectators user are obtained Value.
Multimedia file includes text, picture, video and audio file, and this programme can be used for identifying whether these files wrap Containing certain content, certain content can be terrified and/or Pornograph, identify whether text is color for example with the present invention program Whether feelings text, picture are porny, whether video is pornographic video.
When multimedia file is video file, the video can be order video or live video, described live to regard Frequency includes the video that direct broadcasting room plays.In video display process, Figure Characteristics value and the first wish for obtaining spectators user are special Value indicative, the Figure Characteristics value are used to identify fancy grade of the user to certain content, and the first wish characteristic value is used to mark Know user to wish to watch the wish of certain content in preset time period, preset time period generally refers to current time and elapses forward A period, such as 40 minutes before current time.
Step 3：The multimedia file is calculated according to the Figure Characteristics value and the first wish characteristic value to include in specific The probability of appearance.
It is described that certain content is included according to the Figure Characteristics value and the first wish characteristic value calculating multimedia file Probability specifically include：The second wish feature of each user is determined according to the Figure Characteristics value and the first wish characteristic value Value；The probability of certain content is included according to the second wish characteristic value of all users calculating multimedia file.Composite portrait Characteristic value and the first wish characteristic value, can improve the accuracy for judging whether the multimedia file includes certain content.
In one embodiment, can be obtained by summing to the Figure Characteristics value and the first wish characteristic value The second wish characteristic value；By the way that the second wish characteristic value of each user is compared with default threshold value respectively, meter The second wish characteristic value is calculated more than the number of users of threshold value and the ratio of total number of users amount, obtains the multimedia file bag Probability containing certain content.
In another embodiment, can also be according to the power set in advance for Figure Characteristics value and the first wish characteristic value Weight, is weighted the Figure Characteristics value and the first wish characteristic value averagely, obtains the second wish characteristic value；By respectively will The second wish characteristic value of each user is compared with default threshold value, calculates the second wish characteristic value and exceedes threshold value Number of users and the ratio of total number of users amount, obtain the probability that the multimedia file includes certain content.
Step 4：Judge whether the probability exceedes preset value, if so, then carrying out feature inspection to the multimedia file Survey.
Preset value can be manually set, preset value can combine multimedia file whether be certain content multimedia file Judging result be adjusted, to improve the accuracy of final judging result.If the probability is not above preset value, explanation The possibility that the multimedia file played includes certain content is smaller, to improve detection efficiency and accuracy, can abandon Further detection to this kind of multimedia file, does not carry out any processing to it.If the probability exceedes preset value, illustrate institute It is larger, it is necessary to further detect multimedia file content to state the possibility that video includes certain content.
For text, further detection includes carrying out character machining to content of text.Character spy can be pre-established Storehouse is levied, for storing the characteristic character of the extraction from particular content file (such as erotic novel, porny etc.), is then utilized Characteristic character in character feature storehouse is matched with content of text, after matching result exceedes default matching threshold, explanation The characteristic character that text includes is more, it may be determined that it is the text of certain content.
For picture file, further detection includes examining picture progress character machining, sensitive part detection, skin pixel Survey, color pixel detection etc..Character machining carries out characteristic character matching to be detected using character feature storehouse, sensitive part inspection Survey and carry out sensitive part matching using sensitive part feature database and be detected, color pixel detection and skin pixel detection can be with Color model and complexion model are initially set up, color pixel detection and skin are carried out to picture further according to color model and complexion model Color pixel detects.The construction method of color model and complexion model is the prior art, and details are not described herein.
For audio file, further during detection, an audio detection model can be trained, by audio file to be detected Audio detection model is inputted, to obtain the testing result for whether including certain content.The construction method of audio detection model is existing There is technology, details are not described herein.
For video file, further detection includes audio detection and image detection；Wherein, audio detection can use sound Frequency detection model is detected；Image detection includes extracting the image of the video, and feature detection is carried out to described image.Specifically , the image of the extraction video, carries out described image feature detection, including：The video constant duration is carried The image of preset number is taken, such as image is extracted by being interrupted 10s screenshotss to video；Then feature inspection is carried out to every image Survey, judge whether image includes special characteristic, feature detection includes motion detection, character machining, sensitive part detection, colour of skin picture Plain detection, color pixel detection etc..
Step 5：According to feature testing result judge the multimedia file whether be certain content multimedia file.
The multimedia file is video file, in one embodiment, the figure of special characteristic can be included by statistics Piece number, judges whether the number is more than threshold value P set in advance, when the picture number for judging to include special characteristic is more than pre- During the threshold value P first set, it is certain content video to determine the video, otherwise judges the video for normal video.Another In a embodiment, the number of pictures of special characteristic can be included by statistics, determines that judges includes the image of special characteristic Number and the total number of images purpose ratio extracted for the video detection, when definite ratio is more than threshold value Q, judge institute It is certain content video to state video, otherwise judges the video for normal video.
Video for being judged as certain content, can be further processed, for example, to video carry out classification or Exit broadcasting.
The above method can be used for the pornographic video of identification, wherein, the feature detection to video image includes sensitive part detection Detected with skin pixel.
A kind of implemented method of sensitive part detection includes：
S11, search prestore human body sensitive part index in the matched human body sensitive part figure of images to be recognized Characteristic corresponding to piece.The characteristic of human body sensitive part picture can be pressed certain way by human body sensitive part index Organize, store in an orderly manner, it is easy-to-look-up.Human body sensitive part picture can be quick by marking out human body in porny Sense position simultaneously generates picture and obtains.Characteristic can be vector characteristics, which can be conventional images identification side Arbitrary characteristics in method, for example, texture is described, (Histogram ofOriented Gradient, image gradient direction are straight by HOG Side's figure) or LBP (Local Binary Patterns, local binary patterns) etc..Can be by extracting the spy of images to be recognized Data are levied, and calculate the distance of the characteristic of images to be recognized and the characteristic of human body sensitive part picture, so that according to Whether Distance Judgment picture to be identified matches with human body sensitive part picture.For example can represent distance using Euclidean distance, If the characteristic of images to be recognized and the Euclidean distance of the characteristic of one of human body sensitive part picture are most short, and The Euclidean distance is less than Euclidean distance threshold value, then images to be recognized and the human body sensitive part picture are matched.It is appreciated that , can also be by other similarity measurements to determine whether matching, such as related coefficient etc., do not enumerate here.
S12, the corresponding confidence level of images to be recognized is calculated according to matched characteristic.Confidence level is sentenced for weighing certain The function of the disconnected matching degree between actual observation result.Confidence level is higher, images to be recognized and human body sensitive part picture Matching degree is higher.In one embodiment, between the characteristic of images to be recognized and matched characteristic it is European away from From being negatively correlated relation with confidence level, the relation of a negatively correlated function representation between the two, such as c=e- can be used X, wherein x are the Euclidean distances between the characteristic of images to be recognized and matched characteristic, and c is confidence level.
S13, judges whether images to be recognized is pornographic image according to the corresponding confidence level of images to be recognized.When confidence level height When the first confidence threshold value, illustrate that images to be recognized and the matching degree of matched human body sensitive part picture are very high, wait to know Other image is pornographic image.
A kind of implemented method of skin pixel detection includes：
S21, detects human region pixel and number of people area pixel in video image.
Human testing generally using a kind of Adaboost (iterative algorithm) human testing algorithm (it is of course also possible to use its His algorithm), whether judged by the Adaboost human testing algorithms based on edge histogram feature in image with the presence of human body, The integrogram of video image is calculated first, edge histogram feature is extracted, according to the grader feature database set, runtime class The method of connection searches for human region in the picture.Wherein grader feature database training method includes：Calculate the integration of sample image Figure, extracts the class rectangle feature of sample image；Effective feature is screened according to Adaboost algorithm, forms Weak Classifier；Pass through Multiple Weak Classifiers are combined, form strong classifier；Multiple strong classifiers are cascaded, form the grader feature database of human testing. Human testing unit is detected there are during human body, then video image is detected, and judges whether the number of people.
Number of people detection uses Adaboost number of people detection algorithms, is detected by the Adaboost numbers of people based on class rectangle feature Whether algorithm judges in image, with the presence of the number of people, to calculate the integrogram of image first, extract edge histogram feature, according to having instructed The grader feature database perfected, method seeker's head region in the picture of operation cascade cascades.Wherein grader feature database Training method includes：The integrogram of sample image is calculated, extracts the class rectangle feature of sample image；Sieved according to Adaboost algorithm Effective feature is selected, forms Weak Classifier；By combining multiple Weak Classifiers, strong classifier is formed；Multiple strong classifiers are cascaded, Form the grader feature database of number of people detection.
S22, counts the ratio of skin pixel and image pixel, skin pixel and human region pixel in every video image Ratio and number of people area pixel and the ratio of skin pixel.
S23, according to the first proportion threshold value of skin pixel set in advance and image pixel, skin pixel and human region The second proportion threshold value, number of people area pixel and the 3rd proportion threshold value of skin pixel of pixel and default determination strategy judge to regard Whether frequency image is pornographic image.
First determine whether the ratio of the skin pixel and image pixel is more than the first proportion threshold value, the skin pixel With the ratio of human region pixel whether be more than the second proportion threshold value, the number of people area pixel and skin pixel ratio whether More than the 3rd proportion threshold value, the first result, the second result and the 3rd result are respectively obtained；Then the first result, the second knot are judged Whether fruit and the 3rd result meet determination strategy, if satisfied, the skin pixel of explanation video image meets pornographic image feature, really The fixed video image is pornographic image.Determination strategy can be that the ratio for meeting skin pixel and image pixel is more than the first ratio The ratio of threshold value, skin pixel and human region pixel is more than the ratio of the second proportion threshold value, number of people area pixel and skin pixel Example is more than at least two conditions in the 3rd proportion threshold value.
It should be noted that for foregoing embodiment of the method, in order to be briefly described, therefore it is all expressed as a series of Combination of actions, but those skilled in the art should know, the present invention and from the limitation of described sequence of movement, because according to According to the present invention, some steps can use other orders or be carried out at the same time.Secondly, those skilled in the art should also know, Embodiment described in this description belongs to preferred embodiment, and not necessarily the present invention must for involved action and module Must.
Through the above description of the embodiments, those skilled in the art can be understood that according to above-mentioned implementation The method of example can add the mode of required general hardware platform to realize by software, naturally it is also possible to by hardware, but very much In the case of the former be more preferably embodiment.Based on such understanding, technical scheme is substantially in other words to existing The part that technology contributes can be embodied in the form of software product, which is stored in a storage In medium (such as ROM/RAM, magnetic disc, CD), including some instructions are used so that a station terminal equipment (can be mobile phone, calculate Machine, server, or network equipment etc.) perform method described in each embodiment of the present invention.
Embodiment 3
The present embodiment provides a kind of multimedia file processing method.Under such as running environment of embodiment 1, the application provides Multimedia file processing method as shown in Figure 4.As shown in figure 4, Fig. 4 is at multimedia file according to embodiments of the present invention The flow chart of reason method, a kind of optional scheme of this method include the following steps：
S401：In multimedia file playing process, the Figure Characteristics value and the first wish characteristic value of spectators user are obtained, The Figure Characteristics value is used to identify hobby of the user to certain content, and the first wish characteristic value is used to identify user pre- If wish to watch the wish of certain content in the period；
S402：The multimedia file is calculated according to the Figure Characteristics value and the first wish characteristic value and includes certain content Probability；
S403：Judge whether the probability exceedes preset value, if so, then carrying out feature detection to the multimedia file；
S404：According to feature testing result judge the multimedia file whether be certain content multimedia file；
S405：The multimedia file is handled according to the judging result.
Preferably, the multimedia file is order video or live video, and the certain content is Pornograph；It is described Carrying out processing to the multimedia file according to the judging result includes：If multimedia file is the pornographic video of program request, Exit the broadcasting of order video；If multimedia file is live pornographic video, close between the net cast for playing the video.
The present embodiment is realized carries out multimedia file preliminary screening by user behavioural characteristic, then to screening Multimedia file carries out certain content detection, improves the recognition efficiency and accuracy to certain content.The present invention is used for color Feelings video detection can greatly improve detection efficiency and reliability, facilitate the management and control of multimedia video.
Embodiment 4
The present embodiment provides a kind of multimedia file identification device of Behavior-based control feature.As shown in figure 5, the device includes Acquiring unit 20, computing unit 30, detection unit 40 and determination unit 50.
Acquiring unit 20, in multimedia file playing process, obtaining the Figure Characteristics value and first of spectators user Wish characteristic value, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value is used for Mark user wishes to watch the wish of certain content in preset time period；
Computing unit 30, for calculating the multimedia file bag according to the Figure Characteristics value and the first wish characteristic value Probability containing certain content；
Detection unit 40, for judging whether the probability exceedes preset value, if so, then being carried out to the multimedia file Feature detects；
Determination unit 50, for according to feature testing result judge the multimedia file whether be certain content more matchmakers Body file.
In the multimedia file identification device of the Behavior-based control feature of the embodiment, acquiring unit 20 is used to perform the present invention Step S201 in embodiment 1, computing unit 30 are used to perform the step S202 in the embodiment of the present invention 1, and detection unit 40 is used In performing the step S203 in the embodiment of the present invention 1, determination unit 50 is used to perform the step S204 in the embodiment of the present invention 1.
Referring to Fig. 6, as an alternative embodiment, the computing unit 30 includes：
First computation subunit 301, for determining each user according to the Figure Characteristics value and the first wish characteristic value The second wish characteristic value；
Second computation subunit 302, for calculating the multimedia file according to the second wish characteristic value of all users Probability comprising certain content.
As an alternative embodiment, first computation subunit 301 includes：
First computing module 3011, for summing to the Figure Characteristics value and the first wish characteristic value, obtains institute State the second wish characteristic value；
Second computing module 3012, the weight set in advance for Figure Characteristics value and the first wish characteristic value for basis, The Figure Characteristics value and the first wish characteristic value are weighted averagely, obtain the second wish characteristic value.
Further, second computation subunit 302 includes：
Comparing module 3021, for the second wish characteristic value of each user to be compared with default threshold value respectively；
Probability evaluation entity 3022, it is total more than the number of users of threshold value and user for calculating the second wish characteristic value The ratio of quantity, obtains the probability that the multimedia file includes certain content.
Further, described device further includes pretreatment unit 10, and pretreatment unit 10 is used for the behavior number for analyzing user According to determining the Figure Characteristics value and the first wish characteristic value of user.The pretreatment unit 10 includes the first pretreatment subelement 101 and second pre-process subelement 102.
First processing subelement 101, is used for：The behavioral data of user is obtained, the behavioral data includes browsing in specific First behavioral data of appearance related text, the second behavioral data for browsing certain content picture concerned, access certain content correlation The third line of forum is data and the fourth line chatted in the relevant chat group of certain content is data；Described is judged respectively One behavioral data, the second behavioral data, the third line are data and fourth line is whether data are empty, and correspondence obtains the first judgement knot Fruit, the second judging result, the 3rd judging result and the 4th judging result；According to the of first judging result set in advance One weight, the second weight of second judging result, the 3rd weight of the 3rd judging result and the described 4th judge knot The 4th weight of fruit, divides first judging result, the second judging result, the 3rd judging result and the 4th judging result With integration, the behavioural characteristic value of the user is obtained；
Second processing subelement 102, is used for:Obtain behavioral data of the user within nearest a period of time, the behavior number According to the first time including browsing certain content related text, browse certain content picture concerned the second time, access it is specific 3rd time of content correlation forum and the 4th time chatted in the relevant chat group of certain content；For the first time Assign first weight, second time assigns second weight, the 3rd time assigns the 3rd weight, institute The 4th time imparting the 4th weight is stated, the first time, the second time, the 3rd time and the 4th time are weighted It is average, obtain the wish characteristic value of user.
As an alternative embodiment, the multimedia file is video, the detection unit 40 includes：
Subelement 401 is extracted, for the image to video constant duration extraction preset number；
Detection sub-unit 402, for carrying out feature detection to every image, judges whether described image includes specific spy Sign, the feature detection include sensitive part detection and skin pixel detection.
As an alternative embodiment, the determination unit 50 includes：
First determination subelement 501, for being more than threshold set in advance in the picture number for judging to include special characteristic During value P, judge that the video for certain content video, otherwise judges the video for normal video；Or
Second determination subelement 502, for the picture number comprising special characteristic for determining to judge and is directed to the video The total number of images purpose ratio that Detection and Extraction obtain, when definite ratio is more than threshold value Q, judges the video for certain content Video, otherwise judges the video for normal video.
As the preferred embodiment of the present embodiment, the certain content is Pornograph, and the video is order video or straight Broadcast video.
Embodiment 5
The present embodiment provides a kind of multimedia file processing unit.As shown in fig. 7, the device includes acquiring unit 20, meter Calculate unit 30, detection unit 40, determination unit 50 and processing unit 60.
Acquiring unit 20, in multimedia file playing process, obtaining the Figure Characteristics value and first of spectators user Wish characteristic value, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value is used for Mark user wishes to watch the wish of certain content in preset time period；
Computing unit 30, for calculating the multimedia file bag according to the Figure Characteristics value and the first wish characteristic value Probability containing certain content；
Detection unit 40, for judging whether the probability exceedes preset value, if so, then being carried out to the multimedia file Feature detects；
Determination unit 50, for according to feature testing result judge the multimedia file whether be certain content more matchmakers Body file；
Processing unit 60, for being handled according to the judging result the multimedia file.
In the multimedia file processing unit of the embodiment, acquiring unit 20 is used to perform the step in the embodiment of the present invention 3 Rapid S401, computing unit 30 are used to perform the step S402 in the embodiment of the present invention 3, and detection unit 40 is used to perform of the invention real The step S403 in example 3 is applied, determination unit 50 is used to perform the step S404 in the embodiment of the present invention 3, and processing unit 60 is used for Perform the step S405 in the embodiment of the present invention 3.
Preferably, the multimedia file is order video or live video, and the certain content is Pornograph.It is described Processing unit 60 is specifically used for：When definite multimedia file is the pornographic video of program request, the broadcasting of order video is exited；True When to determine multimedia file be live pornographic video, close between the net cast for playing the video.
Embodiment 6
The embodiment of the present invention additionally provides a kind of storage medium.Alternatively, in the present embodiment, above-mentioned storage medium can For preserving the program code performed by a kind of multimedia file recognition methods of Behavior-based control feature of above-described embodiment.
Alternatively, in the present embodiment, above-mentioned storage medium can be located in multiple network equipments of computer network At least one network equipment.
Alternatively, in the present embodiment, storage medium is arranged to the program code that storage is used to perform following steps：
The first step, in multimedia file playing process, obtains the Figure Characteristics value and the first wish feature of spectators user Value, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value is used to identify user The wish of viewing certain content is wished in preset time period；
Second step, calculates the multimedia file according to the Figure Characteristics value and the first wish characteristic value and includes in specific The probability of appearance；
3rd step, judges whether the probability exceedes preset value, if so, then carrying out feature inspection to the multimedia file Survey；
4th step, according to feature testing result judge the multimedia file whether be certain content multimedia file.
Alternatively, storage medium is also configured to the program code that storage is used to perform following steps：According to the portrait Characteristic value and the first wish characteristic value determine the second wish characteristic value of each user；According to the second wish feature of all users Value calculates the probability that the multimedia file includes certain content.
Alternatively, storage medium is also configured to the program code that storage is used to perform following steps：It is special to the portrait Value indicative and the first wish characteristic value are summed, and obtain the second wish characteristic value, alternatively, according in advance be Figure Characteristics value With the weight of the first wish characteristic value setting, the Figure Characteristics value and the first wish characteristic value are weighted averagely, obtained Second wish characteristic value.
Alternatively, storage medium is also configured to the program code that storage is used to perform following steps：Respectively by each use The second wish characteristic value at family is compared with default threshold value；Calculate the number of users that the second wish characteristic value exceedes threshold value The ratio of amount and total number of users amount, obtains the probability that the multimedia file includes certain content.
Alternatively, storage medium is also configured to the program code that storage is used to perform following steps：Analyze the row of user For data, the Figure Characteristics value and the first wish characteristic value of user are determined.
Alternatively, storage medium is also configured to the program code that storage is used to perform following steps：The multimedia text When part is video, to the image of video constant duration extraction preset number；Feature detection is carried out to every image, described in judgement Whether image includes special characteristic, and the feature detection includes sensitive part detection and skin pixel detection.
Alternatively, storage medium is also configured to the program code that storage is used to perform following steps：When judging to include When the picture number of special characteristic is more than threshold value P set in advance, judge that the video for certain content video, otherwise judges institute It is normal video to state video；Or, the picture number comprising special characteristic for determining to judge for the video detection with extracting The total number of images purpose ratio arrived, when definite ratio is more than threshold value Q, judges the video for certain content video, otherwise Judge the video for normal video.
Alternatively, in the present embodiment, above-mentioned storage medium can include but is not limited to：USB flash disk, read-only storage (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disc or CD etc. is various can be with the medium of store program codes.
Embodiment 7
The embodiment of the present invention additionally provides a kind of storage medium.Alternatively, in the present embodiment, above-mentioned storage medium can For preserving the program code performed by a kind of method for processing video frequency of above-described embodiment.
Alternatively, in the present embodiment, above-mentioned storage medium can be located in multiple network equipments of computer network At least one network equipment.
Alternatively, in the present embodiment, storage medium is arranged to the program code that storage is used to perform following steps：
The first step, in multimedia file playing process, obtains the Figure Characteristics value and the first wish feature of spectators user Value, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value is used to identify user The wish of viewing certain content is wished in preset time period；
Second step, calculates the multimedia file according to the Figure Characteristics value and the first wish characteristic value and includes in specific The probability of appearance；
3rd step, judges whether the probability exceedes preset value, if so, then carrying out feature inspection to the multimedia file Survey；
4th step, according to feature testing result judge the multimedia file whether be certain content multimedia file；
5th step, is handled the multimedia file according to the judging result.
Alternatively, storage medium is also configured to the program code that storage is used to perform following steps：Work as multimedia file When being the pornographic video of program request, the broadcasting of order video is exited；When multimedia file is live pornographic video, closes and play Between the net cast of the video.
Embodiment 8
The embodiment of the present invention also provides a kind of terminal, which can be in terminal group Any one computer terminal.Alternatively, in the present embodiment, above computer terminal can also replace with mobile terminal Deng terminal device.
Alternatively, in the present embodiment, above computer terminal can be located in multiple network equipments of computer network At least one network equipment.
Alternatively, Fig. 8 is the structure diagram of terminal according to embodiments of the present invention.As shown in figure 8, the computer Terminal A can include：One or more (one is only shown in figure) processor 801, memory 803 and transmitting devices 805.
Wherein, memory 803 can be used for storage software program and module, as the Behavior-based control in the embodiment of the present invention is special The multimedia file recognition methods of sign and the corresponding programmed instruction/module of device, processor 801 are stored in memory by operation Software program and module in 803, so as to perform various functions application and data processing, that is, realize above-mentioned multimedia text Part recognition methods.Memory 803 may include high speed random access memory, can also include nonvolatile memory, such as one or Multiple magnetic storage devices, flash memory or other non-volatile solid state memories.In some instances, memory 803 can be into one Step includes that relative to the remotely located memory of processor 801, these remote memories network connection to computer can be passed through Terminal A.The example of above-mentioned network includes but not limited to internet, intranet, LAN, mobile radio communication and combinations thereof.
Above-mentioned transmitting device 805 is used to data are received or sent via a network.Above-mentioned network instantiation It may include cable network and wireless network.In an example, transmitting device 805 includes a network adapter, it can pass through Cable is connected with other network equipments with router so as to be communicated with internet or LAN.In an example, pass Defeated device 805 is radio-frequency module, it is used to wirelessly be communicated with internet.
Wherein, specifically, memory 803 is used for information, the Yi Jiying for storing deliberate action condition and default access user Use program.
Processor 801 can call the information and application program that memory 803 stores by transmitting device, following to perform Step：
The first step, in multimedia file playing process, obtains the Figure Characteristics value and the first wish feature of spectators user Value, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value is used to identify user The wish of viewing certain content is wished in preset time period；
Second step, calculates the multimedia file according to the Figure Characteristics value and the first wish characteristic value and includes in specific The probability of appearance；
3rd step, judges whether the probability exceedes preset value, if so, then carrying out feature inspection to the multimedia file Survey；
4th step, according to feature testing result judge the multimedia file whether be certain content multimedia file.
Specific example in the present embodiment may be referred to the example described in above-described embodiment 1 and embodiment 2, herein not Repeat again.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
If the integrated unit in above-described embodiment is realized in the form of SFU software functional unit and is used as independent product Sale or in use, the storage medium that above computer can be read can be stored in.Based on such understanding, skill of the invention The part or all or part of the technical solution that art scheme substantially in other words contributes the prior art can be with soft The form of part product embodies, which is stored in storage medium, including some instructions are used so that one Platform or multiple stage computers equipment (can be personal computer, server or network equipment etc.) perform each embodiment institute of the present invention State all or part of step of method.
In the above embodiment of the present invention, the description to each embodiment all emphasizes particularly on different fields, and does not have in some embodiment The part of detailed description, may refer to the associated description of other embodiment.
In several embodiments provided herein, it should be understood that disclosed client, can be by others side Formula is realized.Wherein, device embodiment described above is only schematical, such as the division of the unit, is only one Kind of division of logic function, can there is an other dividing mode when actually realizing, for example, multiple units or component can combine or Another system is desirably integrated into, or some features can be ignored, or do not perform.It is another, it is shown or discussed it is mutual it Between coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some interfaces, unit or module Connect, can be electrical or other forms.
The unit illustrated as separating component may or may not be physically separate, be shown as unit The component shown may or may not be physical location, you can with positioned at a place, or can also be distributed to multiple In network unit.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can also That unit is individually physically present, can also two or more units integrate in a unit.Above-mentioned integrated list Member can both be realized in the form of hardware, can also be realized in the form of SFU software functional unit.

Claims (23)

  1. A kind of 1. multimedia file recognition methods of Behavior-based control feature, it is characterised in that including：
    In multimedia file playing process, the Figure Characteristics value and the first wish characteristic value of spectators user, the portrait are obtained Characteristic value is used to identify hobby of the user to certain content, and the first wish characteristic value is used to identify user in preset time period The interior wish for wishing to watch certain content；
    The probability of certain content is included according to the Figure Characteristics value and the first wish characteristic value calculating multimedia file；
    Judge whether the probability exceedes preset value, if so, then carrying out feature detection to the multimedia file；
    According to feature testing result judge the multimedia file whether be certain content multimedia file.
  2. It is 2. according to the method described in claim 1, it is characterized in that, described according to the Figure Characteristics value and the first wish feature Value calculates the probability that the multimedia file includes certain content, including：
    The second wish characteristic value of each user is determined according to the Figure Characteristics value and the first wish characteristic value；
    The probability of certain content is included according to the second wish characteristic value of all users calculating multimedia file.
  3. It is 3. according to the method described in claim 2, it is characterized in that, described according to the Figure Characteristics value and the first wish feature Value determines the second wish characteristic value of each user, including：
    Sum to the Figure Characteristics value and the first wish characteristic value, obtain the second wish characteristic value, alternatively,
    According to the weight set in advance for Figure Characteristics value and the first wish characteristic value, to the Figure Characteristics value and the first wish Characteristic value is weighted averagely, obtains the second wish characteristic value.
  4. 4. according to the method described in claim 2, it is characterized in that, described calculate according to the second wish characteristic value of all users The multimedia file includes the probability of certain content, including：
    The second wish characteristic value of each user is compared with default threshold value respectively；
    The second wish characteristic value is calculated more than the number of users of threshold value and the ratio of total number of users amount, obtains the multimedia File includes the probability of certain content.
  5. 5. according to the method described in claim 1, it is characterized in that, obtaining the Figure Characteristics value and the first wish of spectators user Before characteristic value, further include：
    The behavioral data of user is analyzed, determines the Figure Characteristics value and the first wish characteristic value of user.
  6. 6. according to the method described in claim 5, it is characterized in that, the behavioral data of the analysis user, determines the picture of user As characteristic value, including：
    The behavioral data of user is obtained, the behavioral data includes the first behavioral data, clear for browsing certain content related text Look at certain content picture concerned the second behavioral data, access the third line of certain content correlation forum for data and specific interior It is data to hold the fourth line chatted in relevant chat group；
    Judge that first behavioral data, the second behavioral data, the third line are data and fourth line is whether data are sky respectively, Correspondence obtains the first judging result, the second judging result, the 3rd judging result and the 4th judging result；
    According to the first weight of first judging result set in advance, the second weight of second judging result, described 3rd weight of the 3rd judging result and the 4th weight of the 4th judging result, sentence first judging result, second Disconnected result, the 3rd judging result and the 4th judging result are allocated integration, obtain the behavioural characteristic value of the user.
  7. 7. according to the method described in claim 6, it is characterized in that, the behavioral data of the analysis user, determines the of user One wish characteristic value, including：
    Behavioral data of the user within nearest a period of time is obtained, the behavioral data includes browsing certain content related text At the first time, browse the second time of certain content picture concerned, access the 3rd time of certain content correlation forum and in spy Determine the 4th time chatted in the relevant chat group of content；
    First weight is assigned for the first time, second time assigns second weight, the 3rd time Assign the 3rd weight, the 4th time assigns the 4th weight, during to the first time, the second time, the 3rd Between and the 4th time be weighted average, obtain the wish characteristic value of user.
  8. 8. according to the method described in claim 1, it is characterized in that, the multimedia file is video；
    It is described to the multimedia file carry out feature detection, including：
    To the image of video constant duration extraction preset number；
    Feature detection is carried out to every image, judges whether described image includes special characteristic, the feature detection includes sensitivity Location detection and skin pixel detection.
  9. 9. according to the method described in claim 8, it is characterized in that, described judge the multimedia text according to feature testing result Part whether be certain content multimedia file, including：
    When judging that the picture number comprising special characteristic is more than threshold value P set in advance, judge the video to be specific interior Hold video, otherwise judge the video for normal video；Or
    The picture number comprising special characteristic for determining to judge and the total number of images purpose extracted for the video detection Ratio, when definite ratio is more than threshold value Q, judges that the video for certain content video, otherwise judges the video for just Normal video.
  10. 10. according to the method described in any one in claim 1-9, it is characterised in that the certain content is Pornograph, The video is order video or live video.
  11. A kind of 11. multimedia file processing method, it is characterised in that including：
    In multimedia file playing process, the Figure Characteristics value and the first wish characteristic value of spectators user, the portrait are obtained Characteristic value is used to identify hobby of the user to certain content, and the first wish characteristic value is used to identify user in preset time period The interior wish for wishing to watch certain content；
    The probability of certain content is included according to the Figure Characteristics value and the first wish characteristic value calculating multimedia file；
    Judge whether the probability exceedes preset value, if so, then carrying out feature detection to the multimedia file；
    According to feature testing result judge the multimedia file whether be certain content multimedia file；
    The multimedia file is handled according to the judging result.
  12. 12. according to the method for claim 11, it is characterised in that the multimedia file is order video or live regards Frequently, the certain content is Pornograph；
    It is described processing is carried out to the multimedia file according to the judging result to include：If multimedia file is the pornographic of program request Video, then exit the broadcasting of order video；If multimedia file is live pornographic video, the video for playing the video is closed Direct broadcasting room.
  13. A kind of 13. multimedia file identification device of Behavior-based control feature, it is characterised in that including：
    Acquiring unit, the Figure Characteristics value and the first wish in multimedia file playing process, obtaining spectators user are special Value indicative, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value, which is used to identify, to be used Wish to watch the wish of certain content in preset time period in family；
    Computing unit, for calculating the multimedia file comprising specific according to the Figure Characteristics value and the first wish characteristic value The probability of content；
    Detection unit, for judging whether the probability exceedes preset value, if so, then carrying out feature inspection to the multimedia file Survey；
    Determination unit, for according to feature testing result judge the multimedia file whether be certain content multimedia text Part.
  14. 14. device according to claim 13, it is characterised in that the computing unit includes：
    First computation subunit, for determining the second meaning of each user according to the Figure Characteristics value and the first wish characteristic value It is willing to characteristic value；
    Second computation subunit, for calculating the multimedia file comprising specific according to the second wish characteristic value of all users The probability of content.
  15. 15. device according to claim 14, it is characterised in that first computation subunit includes：
    First computing module, for summing to the Figure Characteristics value and the first wish characteristic value, obtains second meaning It is willing to characteristic value；
    Second computing module, the weight set in advance for Figure Characteristics value and the first wish characteristic value for basis, to the picture It is averaged as characteristic value and the first wish characteristic value are weighted, obtains the second wish characteristic value.
  16. 16. device according to claim 14, it is characterised in that second computation subunit includes：
    Comparing module, for the second wish characteristic value of each user to be compared with default threshold value respectively；
    Probability evaluation entity, for calculating the second wish characteristic value more than the number of users of threshold value and the ratio of total number of users amount Value, obtains the probability that the multimedia file includes certain content.
  17. 17. device according to claim 13, it is characterised in that described device further includes：
    Pretreatment unit, for analyzing the behavioral data of user, determines the Figure Characteristics value and the first wish characteristic value of user.
  18. 18. device according to claim 17, it is characterised in that the pretreatment unit includes：
    First processing subelement, is used for：The behavioral data of user is obtained, the behavioral data includes browsing the related text of certain content This first behavioral data, the second behavioral data for browsing certain content picture concerned, access certain content correlation forum the Three behavioral datas and the fourth line chatted in the relevant chat group of certain content are data；The first behavior number is judged respectively According to, the second behavioral data, the third line are data and fourth line is whether data are empty, correspondence obtains the first judging result, second Judging result, the 3rd judging result and the 4th judging result；According to the first weight of first judging result set in advance, The 4th of second weight of second judging result, the 3rd weight of the 3rd judging result and the 4th judging result Weight, integration is allocated to first judging result, the second judging result, the 3rd judging result and the 4th judging result, Obtain the behavioural characteristic value of the user；
    Second processing subelement, is used for:Behavioral data of the user within nearest a period of time is obtained, the behavioral data includes clear Look at certain content related text first time, browse the second time of certain content picture concerned, to access certain content related 3rd time of forum and the 4th time chatted in the relevant chat group of certain content；For described in first time imparting First weight, second time assign second weight, the 3rd time assigns the 3rd weight, it is described 4th when Between assign the 4th weight, the first time, the second time, the 3rd time and the 4th time are weighted average, obtained To the wish characteristic value of user.
  19. 19. device according to claim 13, it is characterised in that the multimedia file is video；
    The detection unit includes：
    Subelement is extracted, for the image to video constant duration extraction preset number；
    Detection sub-unit, for carrying out feature detection to every image, judges whether described image includes special characteristic, the spy Sign detection includes sensitive part detection and skin pixel detection.
  20. 20. device according to claim 19, it is characterised in that the determination unit includes：
    First determination subelement, for when judging that the picture number comprising special characteristic is more than threshold value P set in advance, sentencing The video that breaks is certain content video, otherwise judges the video for normal video；
    Second determination subelement, the picture number comprising special characteristic for determining to judge are extracted with being directed to the video detection Obtained total number of images purpose ratio, when definite ratio is more than threshold value Q, judges the video for certain content video, no Then judge the video for normal video.
  21. 21. according to the device described in any one in claim 13-20, it is characterised in that the certain content is in pornographic Hold, the video is order video or live video.
  22. 22. a kind of multimedia file processing unit, it is characterised in that described device includes：
    Acquiring unit, the Figure Characteristics value and the first wish in multimedia file playing process, obtaining spectators user are special Value indicative, the Figure Characteristics value are used to identify hobby of the user to certain content, and the first wish characteristic value, which is used to identify, to be used Wish to watch the wish of certain content in preset time period in family；
    Computing unit, for calculating the multimedia file comprising specific according to the Figure Characteristics value and the first wish characteristic value The probability of content；
    Detection unit, for judging whether the probability exceedes preset value, if so, then carrying out feature inspection to the multimedia file Survey；
    Determination unit, for according to feature testing result judge the multimedia file whether be certain content multimedia text Part；
    Processing unit, for being handled according to the judging result the multimedia file.
  23. 23. device according to claim 22, it is characterised in that the multimedia file is order video or live regards Frequently, the certain content is Pornograph；
    The processing unit, specifically for when definite multimedia file is the pornographic video of program request, exiting broadcasting for order video Put；When definite multimedia file is live pornographic video, close between the net cast for playing the video.
CN201610929276.7A 2016-10-13 2016-10-31 Behavior feature-based multimedia file identification method, processing method and device Active CN108024148B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201610929276.7A CN108024148B (en) 2016-10-31 2016-10-31 Behavior feature-based multimedia file identification method, processing method and device
PCT/CN2017/104275 WO2018068664A1 (en) 2016-10-13 2017-09-29 Network information identification method and device
US16/026,786 US10805255B2 (en) 2016-10-13 2018-07-03 Network information identification method and apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610929276.7A CN108024148B (en) 2016-10-31 2016-10-31 Behavior feature-based multimedia file identification method, processing method and device

Publications (2)

Publication Number Publication Date
CN108024148A true CN108024148A (en) 2018-05-11
CN108024148B CN108024148B (en) 2020-02-28

Family

ID=62070429

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610929276.7A Active CN108024148B (en) 2016-10-13 2016-10-31 Behavior feature-based multimedia file identification method, processing method and device

Country Status (1)

Country Link
CN (1) CN108024148B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108834171A (en) * 2018-07-27 2018-11-16 新华三大数据技术有限公司 Portrait method and device
CN112291622A (en) * 2020-10-30 2021-01-29 中国建设银行股份有限公司 Method and device for determining favorite internet surfing time period of user
CN112395556A (en) * 2020-09-30 2021-02-23 广州市百果园网络科技有限公司 Abnormal user detection model training method, abnormal user auditing method and device

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1387338A (en) * 2001-03-29 2002-12-25 松下电器产业株式会社 Data reproducing device and method
CN1640137A (en) * 2002-02-28 2005-07-13 皇家飞利浦电子股份有限公司 Use of transcript information to find key audio/video segments
CN101646050A (en) * 2009-09-09 2010-02-10 中国电信股份有限公司 Text annotation method and system, playing method and system of video files
CN102999588A (en) * 2012-11-15 2013-03-27 广州华多网络科技有限公司 Method and system for recommending multimedia applications
US20140258322A1 (en) * 2013-03-06 2014-09-11 Electronics And Telecommunications Research Institute Semantic-based search system and search method thereof
CN105335422A (en) * 2014-08-06 2016-02-17 阿里巴巴集团控股有限公司 Public opinion information warning method and apparatus

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1387338A (en) * 2001-03-29 2002-12-25 松下电器产业株式会社 Data reproducing device and method
CN1640137A (en) * 2002-02-28 2005-07-13 皇家飞利浦电子股份有限公司 Use of transcript information to find key audio/video segments
CN101646050A (en) * 2009-09-09 2010-02-10 中国电信股份有限公司 Text annotation method and system, playing method and system of video files
CN102999588A (en) * 2012-11-15 2013-03-27 广州华多网络科技有限公司 Method and system for recommending multimedia applications
US20140258322A1 (en) * 2013-03-06 2014-09-11 Electronics And Telecommunications Research Institute Semantic-based search system and search method thereof
CN105335422A (en) * 2014-08-06 2016-02-17 阿里巴巴集团控股有限公司 Public opinion information warning method and apparatus

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108834171A (en) * 2018-07-27 2018-11-16 新华三大数据技术有限公司 Portrait method and device
CN108834171B (en) * 2018-07-27 2021-09-17 新华三大数据技术有限公司 Image method and device
CN112395556A (en) * 2020-09-30 2021-02-23 广州市百果园网络科技有限公司 Abnormal user detection model training method, abnormal user auditing method and device
CN112291622A (en) * 2020-10-30 2021-01-29 中国建设银行股份有限公司 Method and device for determining favorite internet surfing time period of user

Also Published As

Publication number Publication date
CN108024148B (en) 2020-02-28

Similar Documents

Publication Publication Date Title
CN110166827B (en) Video clip determination method and device, storage medium and electronic device
CN111212303B (en) Video recommendation method, server and computer-readable storage medium
CN109784274B (en) Method for identifying trailing and related product
CN105302911B (en) A kind of data screening engine method for building up and data screening engine
US10805255B2 (en) Network information identification method and apparatus
CN110532866A (en) Video data detection method, device, computer equipment and storage medium
CN108304452B (en) Article processing method and device and storage medium
CN111291596B (en) Early warning method and device based on face recognition
CN110866236A (en) Private picture display method, device, terminal and storage medium
CN111432206A (en) Video definition processing method and device based on artificial intelligence and electronic equipment
CN109548691A (en) A kind of pet recognition methods, device, medium and electronic equipment
CN108024148A (en) The multimedia file recognition methods of Behavior-based control feature, processing method and processing device
CN105956051A (en) Information finding method, device and system
CN113139415A (en) Video key frame extraction method, computer device and storage medium
Fan et al. A novel approach for privacy-preserving video sharing
CN111354013A (en) Target detection method and device, equipment and storage medium
WO2015102711A2 (en) A method and system of enforcing privacy policies for mobile sensory devices
CN108804893A (en) A kind of control method, device and server based on recognition of face
CN112528140A (en) Information recommendation method, device, equipment, system and storage medium
CN115983873B (en) User data analysis management system and method based on big data
CN116261009B (en) Video detection method, device, equipment and medium for intelligently converting video audience
CN112668357A (en) Monitoring method and device
JP7070665B2 (en) Information processing equipment, control methods, and programs
CN115376054A (en) Target detection method, device, equipment and storage medium
CN109885760A (en) Information source tracing method and system based on user interest

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
GR01 Patent grant
GR01 Patent grant