CN110493615A - A kind of live video monitoring method and electronic equipment - Google Patents
A kind of live video monitoring method and electronic equipment Download PDFInfo
- Publication number
- CN110493615A CN110493615A CN201810462293.3A CN201810462293A CN110493615A CN 110493615 A CN110493615 A CN 110493615A CN 201810462293 A CN201810462293 A CN 201810462293A CN 110493615 A CN110493615 A CN 110493615A
- Authority
- CN
- China
- Prior art keywords
- live video
- image data
- similarity
- stored image
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000006870 function Effects 0.000 claims description 14
- 238000012880 independent component analysis Methods 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 230000035945 sensitivity Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 208000001613 Gambling Diseases 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 230000005291 magnetic effect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000007850 degeneration Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- 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
-
- 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
- G06V20/48—Matching video sequences
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
- H04N21/2407—Monitoring of transmitted content, e.g. distribution time, number of downloads
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/266—Channel or content management, e.g. generation and management of keys and entitlement messages in a conditional access system, merging a VOD unicast channel into a multicast channel
Abstract
Present disclose provides a kind of live video monitoring method and electronic equipment, method includes: to be decoded to live video, obtains live video voice data and live video image data；Live video voice data is identified, whether to determine in live video voice data including sensitive content；Live video image data is compared with pre-stored image data, to determine the similarity of live video image data and pre-stored image data；When live video voice data includes sensitive content, and/or, the similarity of live video image data and pre-stored image data is greater than threshold value, is monitored to live video.
Description
Technical field
This disclosure relates to a kind of live video monitoring method and electronic equipment.
Background technique
It is increasing with live streaming, it is possible that some relate to live content that is yellow, relating to sudden and violent, related to gambling activities, relate to political affairs etc., need
Unsound direct broadcasting room is closed in time, is purified Internet environment, and a large amount of direct broadcasting room such as game, live teaching broadcast etc. do not need real prison
Control, thus propose monitoring emphasis room concept, monitoring personnel in monitor video, can by mark key monitoring room, from
In the direct broadcasting room of magnanimity, select it is possible that relate to it is yellow, a small amount of direct broadcasting room such as relate to sudden and violent, related to gambling activities, relate to political affairs, and then to these emphasis rooms
Between be monitored, improve working efficiency, purify live content.
But above-mentioned the method that live video is monitored is had the following deficiencies:
Human cost needed for artificial monitoring not only is larger, but also when direct broadcasting room quantity is more, each direct broadcasting room it is straight
The monitoring precision for broadcasting video is lower, so that the being found property of the live video of flame is lower, good can not realize
Prevent the propagation of network flame.
Summary of the invention
The disclosure combines the identification based on voice and based on video image, reaches better recognition effect.It is obtaining
When taking live video, the voice data and net cast screenshot between net cast are respectively obtained, is judged in the audio, video data
Audio data whether be related to sensitive content, and itself and pre-stored image similarity are calculated, if being related to sensitive content or two figures
It is then emphasis room this direct broadcasting room label as similarity is higher, if not being related to sensitive content or two image similarities are low,
For common direct broadcasting room, monitoring personnel carries out normal direct broadcasting room to discover whether that there are emphasis rooms with machine monitoring, and if so,
Emphasis room is added, key monitoring finally is carried out to all emphasis rooms.
An aspect of this disclosure provides a kind of live video monitoring method, comprising: is decoded, obtains to live video
To live video voice data and live video image data；The live video voice data is identified, to determine
Whether state in live video voice data includes sensitive content；The live video image data and pre-stored image data are carried out
Compare, with the similarity of the determination live video image data and pre-stored image data；When the live video voice data
Including sensitive content, and/or, the similarity of the live video image data and pre-stored image data is greater than threshold value, to described
Live video is monitored.
Optionally, before being identified to the live video voice data, comprising: use the complete set of adaptive noise
Empirical mode decomposition method decomposes the live video voice data, obtains at least one intrinsic mode function；Using
Independent component analysis method handles at least one described intrinsic mode function, isolates at least one voice signal point
Amount；At least one described voice signal components are reconstructed, the live video voice data enhanced.
Optionally, the live video image data is compared with pre-stored image data, with the determination live streaming view
The similarity of frequency image data and pre-stored image data, comprising: calculate separately the live video image data and pre-stored image
The characteristic similarity of textural characteristics similarity and Riemann manifold between data is flowed according to the textural characteristics similarity and Riemann
The characteristic similarity of shape determines the comprehensive similarity of the live video image data and pre-stored image data.
Optionally, the live video image data is compared with pre-stored image data, with the determination live streaming view
The similarity of frequency image data and pre-stored image data, comprising: obtain the first texture of the live video image data Q respectively
Second textural characteristics of feature and pre-stored image data I；Calculate the similarity of first textural characteristics and the second textural characteristics
Stexture:
Wherein,For the texture feature vector of first textural characteristics,It is special for the texture of second textural characteristics
Vector is levied, n is feature vector number；Calculate the feature of the Riemann manifold of the live video image data and pre-stored image data
Similarity Smanifold；Determine the comprehensive similarity S (Q, I) of the live video image data Q and pre-stored image data I:
S (Q, I)=ωZSmanifold(Q,I)+ωtStexture(Q,I)
Wherein, ωZAnd ωtIt is two adjustable weights, and meets ωZ+ωt=1.
Optionally, when the live video voice data includes sensitive content, and/or, the live video image data
It is greater than threshold value with the similarity of pre-stored image data, first level monitoring is carried out to the live video；When the live video
Voice data does not include sensitive content, also, the live video image data and the similarity of pre-stored image data are less than etc.
In threshold value, second level monitoring is carried out to the live video.
Another aspect of the disclosure provides a kind of electronic equipment, comprising: processor；Memory is stored with calculating
Machine executable program, the program by the processor when being executed, so that the processor executes: carrying out to live video
Decoding, obtains live video voice data and live video image data；The live video voice data is identified, with
Determine in the live video voice data whether include sensitive content；By the live video image data and pre-stored image number
According to being compared, with the similarity of the determination live video image data and pre-stored image data；When the live video language
Sound data include sensitive content, and/or, the similarity of the live video image data and pre-stored image data is greater than threshold value,
The live video is monitored.
Optionally, it before the processor identifies the live video voice data, also executes: using adaptive
The complete set empirical mode decomposition method of noise decomposes the live video voice data, obtains at least one eigen mode
State function；At least one described intrinsic mode function is handled using independent component analysis method, isolates at least one
Voice signal components；At least one described voice signal components are reconstructed, the live video voice data enhanced.
Optionally, the live video image data is compared by the processor with pre-stored image data, with determination
The similarity of the live video image data and pre-stored image data, comprising: calculate separately the live video image data
The characteristic similarity of textural characteristics similarity and Riemann manifold between pre-stored image data, it is similar according to the textural characteristics
Degree and the characteristic similarity of Riemann manifold determine the comprehensive similarity of the live video image data and pre-stored image data.
Optionally, the live video image data is compared by the processor with pre-stored image data, with determination
The similarity of the live video image data and pre-stored image data, comprising: obtain the live video image data Q respectively
The first textural characteristics and pre-stored image data I the second textural characteristics；It calculates first textural characteristics and the second texture is special
The similarity S of signtexture:
Wherein,For the texture feature vector of first textural characteristics,It is special for the texture of second textural characteristics
Vector is levied, n is feature vector number；Calculate the feature of the Riemann manifold of the live video image data and pre-stored image data
Similarity Smanifold；Determine the comprehensive similarity S (Q, I) of the live video image data Q and pre-stored image data I:
S (Q, I)=ωZSmanifold(Q,I)+ωtStexture(Q,I)
Wherein, ωZAnd ωtIt is two adjustable weights, and meets ωZ+ωt=1.
Optionally, the processor executes: when the live video voice data includes sensitive content, and/or, it is described straight
The similarity for broadcasting video image data and pre-stored image data is greater than threshold value, carries out first level monitoring to the live video；
When the live video voice data does not include sensitive content, also, the live video image data and pre-stored image data
Similarity be less than or equal to threshold value, to the live video carry out second level monitoring.
Detailed description of the invention
In order to which the disclosure and its advantage is more fully understood, referring now to being described below in conjunction with attached drawing, in which:
Fig. 1 diagrammatically illustrates the flow chart of live video monitoring method according to an embodiment of the present disclosure.
Fig. 2 diagrammatically illustrates the block diagram of live video monitoring system according to an embodiment of the present disclosure.
Fig. 3 diagrammatically illustrates the block diagram of the electronic equipment according to the disclosure.
Specific embodiment
According in conjunction with attached drawing to the described in detail below of disclosure exemplary embodiment, other aspects, the advantage of the disclosure
Those skilled in the art will become obvious with prominent features.
In the disclosure, term " includes " and " containing " and its derivative mean including rather than limit；Term "or" is packet
Containing property, mean and/or.
In the present specification, following various embodiments for describing disclosure principle only illustrate, should not be with any
Mode is construed to limitation scope of disclosure.Referring to attached drawing the comprehensive understanding described below that is used to help by claim and its equivalent
The exemplary embodiment for the disclosure that object limits.Described below includes a variety of details to help to understand, but these details are answered
Think to be only exemplary.Therefore, it will be appreciated by those of ordinary skill in the art that without departing substantially from the scope of the present disclosure and spirit
In the case where, embodiment described herein can be made various changes and modifications.In addition, for clarity and brevity,
The description of known function and structure is omitted.In addition, running through attached drawing, same reference numbers are used for identity function and operation.
Fig. 1 diagrammatically illustrates the flow chart of live video monitoring method according to an embodiment of the present disclosure.
As shown in Figure 1, method specifically includes following operation:
S1 obtains live video.
In above-mentioned steps, the video being broadcast live can be obtained in real time.Such as in numerous portals live streaming platform, pipe
Reason person can constantly intercept the video flowing of each direct broadcasting room in the caching on backstage.Flow is larger or inadequate buffer space
In the case where, can also temporally threshold extraction direct broadcasting room live video.It can also be directed to the difference of viewing number, regulation is not
With the acquisition dynamics of live video.Above it is the schematical video acquiring method of the disclosure, but is also not necessarily limited to this.
S2 is decoded live video, obtains live video voice data.
It is to separate the audio data in live video by decoded mode, what is obtained is straight in above-mentioned steps
Broadcasting video speech data is the audio data during live streaming, such as can be the sound of main broadcaster, the background sound during being broadcast live
Deng.
S3 is decoded live video, obtains live video image data.
It is to separate the image data in live video by decoded mode, what is obtained is straight in above-mentioned steps
Broadcasting video image data is the image data during live streaming, such as can be the portrait of main broadcaster, the scene figure in game live streaming
As etc..
Whether S4 identifies live video voice data, to determine in live video voice data including in sensitivity
Hold.
In above-mentioned steps, using the complete set empirical mode decomposition method (Complementary of adaptive noise
EnsembleEmpirical Mode Decomposition with Adaptive Noise, CEEMDAN) to live video language
Sound data are decomposed, at least one intrinsic mode function is obtained, and can effectively be reduced and be calculated cost and modal overlap is overcome to ask
Topic；Then, using independent component analysis method (independent component analysis, ICA) at least one sheet
Sign mode function is handled, at least one effective voice signal components is isolated；Finally, at least one voice signal point
Amount is reconstructed, the live video voice data enhanced.
Live video image data is compared by S5 with pre-stored image data, with determine live video image data with
The similarity of pre-stored image data.
In above-mentioned steps, pre-stored image data be server in it is pre-stored include violation content image, lead to
It crosses the live video image data that will be isolated to be compared with pre-stored image data, whether to determine live video image data
Including violation content.
Specifically, the first textural characteristics of live video image data Q and the second line of pre-stored image data I are obtained respectively
Feature is managed, since texture is narrow band signal, and different texture generally has different centre frequency and bandwidth, therefore is utilized respectively
Small wave converting method extracts the textural characteristics of two images to be compared, and filter will input texture image I (x, y) and shearlet
Small echo carries out convolution, and image texture characteristic can be obtained；
Calculate the similarity S of the first textural characteristics and the second textural characteristicstexture:
Wherein,For the texture feature vector of the first textural characteristics,For the texture feature vector of the second textural characteristics, n
For feature vector number；
Calculate the characteristic similarity of the Riemann manifold of live video image data and pre-stored image data
Smanifold；In this step, the characteristic similarity for calculating the Riemann manifold of two image datas is substantially to compare
Neighborhood (x, y the ∈ R of same position pixel (x, y)2), wherein R2For image area, the embodiment of the present disclosure is using Euclidean distance ratio
It is whether similar compared with the neighborhood of point (x, y):
Wherein, gtIt is a given window function, and is the Gaussian function of variable t, h is defined in R2On seat
Scale value；
The image similarity in Riemann manifold defined above compares formula, is used to calculate the similarity between image.
The comparison formula needs to solve one with x, and y is as follows as degeneration PDE, the PDE equation of variable:
Wherein D is Euclidean distance, ΔxD is differential of the D in the direction x, ΔyFor D in the differential in the direction y, Tr is matrix pair
The quadratic sum of linea angulata, DxyIndicate D to the partial derivative of xy；
Determine the comprehensive similarity S (Q, I) of live video image data Q and pre-stored image data I:
S (Q, I)=ωZSmanifold(Q,I)+ωtStexture(Q,I)
Wherein, ωZAnd ωtIt is two adjustable weights, and meets ωZ+ωt=1.
In addition, the value of S (Q, I) is smaller in above-mentioned steps, then it is higher to be considered as similarity.ωZAnd ωtSelection be foundation
What the contribution degree of corresponding component part determined, using weight-coefficient compromise is based on, according to feedback information appropriate adjustment range formula
In weight coefficient achieve the purpose that Optimized Matching result, effect calculates accuracy than a kind of image similarity is used alone
It is high.
S6 judges that live video voice data includes sensitive content, and/or, live video image data and pre-stored image
Whether the similarity of data is greater than threshold value.
In above-mentioned steps, if the live video voice data identified include relate to it is yellow, relate to sudden and violent, related to gambling activities, relate to political affairs etc. it is quick
Feel content, is then considered as the live video and is accused of in violation of rules and regulations, and/or, if the phase of live video image data and pre-stored image data
It is greater than threshold value like degree, then is considered as the live video and is accused of in violation of rules and regulations.
The direct broadcasting room of the live video is included in emphasis room if S6 determines to set up by S7.In this step, substantially
It is to classify to the direct broadcasting room of entire live streaming platform, determines that the biggish direct broadcasting room of violation possibility and violation possibility are smaller
Direct broadcasting room, specifically, if S6 judgement set up, it is determined that the direct broadcasting room of the live video be the biggish live streaming of violation possibility
Between, it is on the contrary, it is determined that the direct broadcasting room of the live video is the lesser direct broadcasting room of violation possibility.Then it is transferred to S8.
S8 carries out first level monitoring to live video.In this step, to the violation possibility determined in step S7
Biggish direct broadcasting room carries out first level monitoring.Specifically, can take monitoring personnel manually to the live video of the direct broadcasting room into
Row is checked, to determine whether there is unlawful practice.
S9, if S6 determines invalid, i.e. live video voice data does not include sensitive content, also, live video figure
As whether data and the similarity of pre-stored image data are less than or equal to threshold value, then it is assumed that the direct broadcasting room where the live video is separated
The lesser direct broadcasting room of possibility is advised, and is included in normal room.Then it is transferred to S10.
S10 carries out second level monitoring to it for the lesser direct broadcasting room of violation possibility determined in S9.Specifically
Ground can enable monitoring personnel to carry out the direct broadcasting room of random inspection with the direct broadcasting room of these such as normal rooms of system random inspection
Monitoring.
Fig. 2 diagrammatically illustrates the block diagram of live video monitoring system according to an embodiment of the present disclosure.
As shown in Fig. 2, live video monitoring system 200 includes Video decoding module 210, speech recognition module 220, image
Identification module 230 and video monitoring module 240.The system 200 can execute the method described above with reference to Fig. 1, straight to realize
Broadcast video monitoring.
Specifically, in identification procedure, Video decoding module 210 is decoded live video, obtains live streaming view
Frequency voice data and live video image data；Speech recognition module 220 identifies live video voice data, with determination
It whether include sensitive content in live video voice data；Picture recognition module 230 is by live video image data and prestores figure
As data are compared, to determine that the similarity of live video image data and pre-stored image data, video monitoring module 240 are used
It include sensitive content in working as live video voice data, and/or, the similarity of live video image data and pre-stored image data
Greater than threshold value, the live video is monitored.
Fig. 3 diagrammatically illustrates the block diagram of the electronic equipment according to the disclosure.
As shown in figure 3, electronic equipment 300 includes processor 310, computer readable storage medium 320.The robot 300
The method described above with reference to Fig. 1 can be executed, to realize that live video monitors.
Specifically, processor 310 for example may include general purpose microprocessor, instruction set processor and/or related chip group
And/or special microprocessor (for example, specific integrated circuit (ASIC)), etc..Processor 310 can also include using for caching
The onboard storage device on way.Processor 310 can be for executing the method flow according to the embodiment of the present disclosure for referring to Fig. 1 description
Different movements single treatment units either multiple processing units.
Computer readable storage medium 320, such as can be times can include, store, transmitting, propagating or transmitting instruction
Meaning medium.For example, readable storage medium storing program for executing can include but is not limited to electricity, magnetic, optical, electromagnetic, infrared or semiconductor system, device,
Device or propagation medium.The specific example of readable storage medium storing program for executing includes: magnetic memory apparatus, such as tape or hard disk (HDD)；Optical storage
Device, such as CD (CD-ROM)；Memory, such as random access memory (RAM) or flash memory；And/or wire/wireless communication chain
Road.
Computer readable storage medium 320 may include computer program 321, which may include generation
Code/computer executable instructions retouch the execution of processor 310 for example above in conjunction with Fig. 1
The method flow stated and its any deformation.
Computer program 521 can be configured to have the computer program code for example including computer program module.Example
Such as, in the exemplary embodiment, the code in computer program 321 may include one or more program modules, for example including
321A, module 321B ....It should be noted that the division mode and number of module are not fixation, those skilled in the art can
To be combined according to the actual situation using suitable program module or program module, when these program modules are combined by processor 310
When execution, processor 310 is executed for example above in conjunction with method flow described in Fig. 1 and its any deformation.
Although the disclosure, art technology has shown and described referring to the certain exemplary embodiments of the disclosure
Personnel it should be understood that in the case where the spirit and scope of the present disclosure limited without departing substantially from the following claims and their equivalents,
A variety of changes in form and details can be carried out to the disclosure.Therefore, the scope of the present disclosure should not necessarily be limited by above-described embodiment,
But should be not only determined by appended claims, also it is defined by the equivalent of appended claims.
Claims (10)
1. a kind of live video monitoring method, comprising:
Live video is decoded, live video voice data and live video image data are obtained；
The live video voice data is identified, whether to include in sensitivity in the determination live video voice data
Hold；
The live video image data is compared with pre-stored image data, with the determination live video image data with
The similarity of pre-stored image data；
When the live video voice data includes sensitive content, and/or, the live video image data and pre-stored image number
According to similarity be greater than threshold value, the live video is monitored.
2. live video monitoring method according to claim 1, before being identified to the live video voice data,
Include:
The live video voice data is decomposed using adaptive noise complete set empirical mode decomposition method, is obtained
At least one intrinsic mode function；
At least one described intrinsic mode function is handled using independent component analysis method, isolates at least one voice
Signal component；
At least one described voice signal components are reconstructed, the live video voice data enhanced.
3. live video monitoring method according to claim 1, by the live video image data and pre-stored image number
According to being compared, with the similarity of the determination live video image data and pre-stored image data, comprising:
Calculate separately the textural characteristics similarity and Riemann manifold between the live video image data and pre-stored image data
Characteristic similarity, the live video image is determined according to the characteristic similarity of the textural characteristics similarity and Riemann manifold
The comprehensive similarity of data and pre-stored image data.
4. live video monitoring method according to claim 3, by the live video image data and pre-stored image number
According to being compared, with the similarity of the determination live video image data and pre-stored image data, comprising:
The first textural characteristics of the live video image data Q and the second textural characteristics of pre-stored image data I are obtained respectively；
Calculate the similarity S of first textural characteristics and the second textural characteristicstexture:
Wherein,For the texture feature vector of first textural characteristics,For second textural characteristics textural characteristics to
Amount, n are feature vector number；
Calculate the characteristic similarity S of the Riemann manifold of the live video image data and pre-stored image datamanifold；
Determine the comprehensive similarity S (Q, I) of the live video image data Q and pre-stored image data I:
S (Q, I)=ωZSmanifold(Q,I)+ωtStexture(Q,I)
Wherein, ωZAnd ωtIt is two adjustable weights, and meets ωZ+ωt=1.
5. live video monitoring method according to claim 1, wherein when the live video voice data includes sensitivity
Content, and/or, the similarity of the live video image data and pre-stored image data is greater than threshold value, to the live video
Carry out first level monitoring；
When the live video voice data does not include sensitive content, also, the live video image data and pre-stored image
The similarity of data is less than or equal to threshold value, carries out second level monitoring to the live video.
6. a kind of electronic equipment, comprising:
Processor；
Memory is stored with computer executable program, and the program by the processor when being executed, so that the place
Device is managed to execute:
Live video is decoded, live video voice data and live video image data are obtained；
The live video voice data is identified, whether to include in sensitivity in the determination live video voice data
Hold；
The live video image data is compared with pre-stored image data, with the determination live video image data with
The similarity of pre-stored image data；
When the live video voice data includes sensitive content, and/or, the live video image data and pre-stored image number
According to similarity be greater than threshold value, the live video is monitored.
7. electronic equipment according to claim 6, the processor carries out identifying it to the live video voice data
Before, also execute:
The live video voice data is decomposed using adaptive noise complete set empirical mode decomposition method, is obtained
At least one intrinsic mode function；
At least one described intrinsic mode function is handled using independent component analysis method, isolates at least one voice
Signal component；
At least one described voice signal components are reconstructed, the live video voice data enhanced.
8. electronic equipment according to claim 6, the processor is by the live video image data and pre-stored image
Data are compared, with the similarity of the determination live video image data and pre-stored image data, comprising:
Calculate separately the textural characteristics similarity and Riemann manifold between the live video image data and pre-stored image data
Characteristic similarity, the live video image is determined according to the characteristic similarity of the textural characteristics similarity and Riemann manifold
The comprehensive similarity of data and pre-stored image data.
9. electronic equipment according to claim 8, the processor is by the live video image data and pre-stored image
Data are compared, with the similarity of the determination live video image data and pre-stored image data, comprising:
The first textural characteristics of the live video image data Q and the second textural characteristics of pre-stored image data I are obtained respectively；
Calculate the similarity S of first textural characteristics and the second textural characteristicstexture:
Wherein,For the texture feature vector of first textural characteristics,For second textural characteristics textural characteristics to
Amount, n are feature vector number；
Calculate the characteristic similarity S of the Riemann manifold of the live video image data and pre-stored image datamanifold；
Determine the comprehensive similarity S (Q, I) of the live video image data Q and pre-stored image data I:
S (Q, I)=ωZSmanifold(Q,I)+ωtStextur(Q,I)
Wherein, ωZAnd ωtIt is two adjustable weights, and meets ωZ+ωt=1.
10. electronic equipment according to claim 6, wherein the processor executes:
When the live video voice data includes sensitive content, and/or, the live video image data and pre-stored image number
According to similarity be greater than threshold value, to the live video carry out first level monitoring；
When the live video voice data does not include sensitive content, also, the live video image data and pre-stored image
The similarity of data is less than or equal to threshold value, carries out second level monitoring to the live video.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810462293.3A CN110493615A (en) | 2018-05-15 | 2018-05-15 | A kind of live video monitoring method and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810462293.3A CN110493615A (en) | 2018-05-15 | 2018-05-15 | A kind of live video monitoring method and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110493615A true CN110493615A (en) | 2019-11-22 |
Family
ID=68545217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810462293.3A Pending CN110493615A (en) | 2018-05-15 | 2018-05-15 | A kind of live video monitoring method and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110493615A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111372091A (en) * | 2020-02-26 | 2020-07-03 | 广州趣丸网络科技有限公司 | Live content risk information control method and system |
| CN111460907A (en) * | 2020-03-05 | 2020-07-28 | 浙江大华技术股份有限公司 | Malicious behavior identification method, system and storage medium |
| CN112217841A (en) * | 2020-12-09 | 2021-01-12 | 平安国际智慧城市科技股份有限公司 | Live broadcast room management method and device, computer equipment and storage medium |
| CN112235601A (en) * | 2020-10-13 | 2021-01-15 | 中国联合网络通信集团有限公司 | Live broadcast behavior correction method, terminal, edge server and computer equipment |
| CN112800919A (en) * | 2021-01-21 | 2021-05-14 | 百度在线网络技术（北京）有限公司 | Method, device and equipment for detecting target type video and storage medium |
| CN115002500A (en) * | 2022-06-15 | 2022-09-02 | 北京搜房科技发展有限公司 | Live broadcast management method and device, storage medium and electronic equipment |
| CN115623239A (en) * | 2022-10-21 | 2023-01-17 | 宁波理查德文化创意有限公司 | Personalized live broadcast control method based on use habit |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106101819A (en) * | 2016-06-21 | 2016-11-09 | 武汉斗鱼网络科技有限公司 | A kind of live video sensitive content filter method based on speech recognition and device |
| CN107027047A (en) * | 2017-05-16 | 2017-08-08 | 武汉斗鱼网络科技有限公司 | Net cast monitoring method, device and electronic equipment |
| CN107181979A (en) * | 2017-03-30 | 2017-09-19 | 武汉斗鱼网络科技有限公司 | A kind of network direct broadcasting monitoring method and device |
| EP3236660A1 (en) * | 2016-04-22 | 2017-10-25 | Beijing Xiaomi Mobile Software Co., Ltd. | Video processing method and device |
| CN107465657A (en) * | 2017-06-22 | 2017-12-12 | 武汉斗鱼网络科技有限公司 | Live video monitoring method, storage medium, electronic equipment and system |
-
2018
- 2018-05-15 CN CN201810462293.3A patent/CN110493615A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3236660A1 (en) * | 2016-04-22 | 2017-10-25 | Beijing Xiaomi Mobile Software Co., Ltd. | Video processing method and device |
| CN106101819A (en) * | 2016-06-21 | 2016-11-09 | 武汉斗鱼网络科技有限公司 | A kind of live video sensitive content filter method based on speech recognition and device |
| CN107181979A (en) * | 2017-03-30 | 2017-09-19 | 武汉斗鱼网络科技有限公司 | A kind of network direct broadcasting monitoring method and device |
| CN107027047A (en) * | 2017-05-16 | 2017-08-08 | 武汉斗鱼网络科技有限公司 | Net cast monitoring method, device and electronic equipment |
| CN107465657A (en) * | 2017-06-22 | 2017-12-12 | 武汉斗鱼网络科技有限公司 | Live video monitoring method, storage medium, electronic equipment and system |
Non-Patent Citations (1)
| Title |
|---|
| 司凯威: "《数字传媒研究》", 31 December 2016 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111372091A (en) * | 2020-02-26 | 2020-07-03 | 广州趣丸网络科技有限公司 | Live content risk information control method and system |
| CN111460907A (en) * | 2020-03-05 | 2020-07-28 | 浙江大华技术股份有限公司 | Malicious behavior identification method, system and storage medium |
| CN112235601A (en) * | 2020-10-13 | 2021-01-15 | 中国联合网络通信集团有限公司 | Live broadcast behavior correction method, terminal, edge server and computer equipment |
| CN112217841A (en) * | 2020-12-09 | 2021-01-12 | 平安国际智慧城市科技股份有限公司 | Live broadcast room management method and device, computer equipment and storage medium |
| CN112800919A (en) * | 2021-01-21 | 2021-05-14 | 百度在线网络技术（北京）有限公司 | Method, device and equipment for detecting target type video and storage medium |
| CN115002500A (en) * | 2022-06-15 | 2022-09-02 | 北京搜房科技发展有限公司 | Live broadcast management method and device, storage medium and electronic equipment |
| CN115002500B (en) * | 2022-06-15 | 2024-02-13 | 北京搜房科技发展有限公司 | Live broadcast management method and device, storage medium and electronic equipment |
| CN115623239A (en) * | 2022-10-21 | 2023-01-17 | 宁波理查德文化创意有限公司 | Personalized live broadcast control method based on use habit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110493615A (en) | A kind of live video monitoring method and electronic equipment | |
| CN110555390B (en) | Pedestrian re-identification method, device and medium based on semi-supervised training mode | |
| CN108875732B (en) | Model training and instance segmentation method, device and system and storage medium | |
| CN114913565B (en) | Face image detection method, model training method, device and storage medium | |
| CN108573243A (en) | A kind of comparison method of the low quality face based on depth convolutional neural networks | |
| CN108197669B (en) | Feature training method and device of convolutional neural network | |
| CN109766925B (en) | Feature fusion method and device, electronic equipment and storage medium | |
| CN112200041B (en) | Video motion recognition method and device, storage medium and electronic equipment | |
| TWI769787B (en) | Target tracking method and apparatus, storage medium | |
| CN110428399A (en) | Method, apparatus, equipment and storage medium for detection image | |
| CN113936302B (en) | Training method and device for pedestrian re-recognition model, computing equipment and storage medium | |
| CN114936979B (en) | Model training method, image denoising method, device, equipment and storage medium | |
| CN113408566A (en) | Target detection method and related equipment | |
| CN110489659A (en) | Data matching method and device | |
| CN112037800A (en) | Voiceprint nuclear model training method and device, medium and electronic equipment | |
| WO2023093244A1 (en) | Model training method and apparatus, device, medium and program product | |
| CN111178503A (en) | Mobile terminal-oriented decentralized target detection model training method and system | |
| CN111339869A (en) | Face recognition method, face recognition device, computer readable storage medium and equipment | |
| CN111128196A (en) | Method and device for identifying anchor based on voiceprint characteristics | |
| CN105069473B (en) | More exemplary weights packet learning methods towards online uncertain image recognition | |
| CN113763420A (en) | Target tracking method, system, storage medium and terminal equipment | |
| CN116503399B (en) | Insulator pollution flashover detection method based on YOLO-AFPS | |
| CN110532890B (en) | Distributed twin convolutional neural network pedestrian re-identification method based on cloud end, edge end and equipment end | |
| CN110163489B (en) | Method for evaluating rehabilitation exercise effect | |
| CN116546304A (en) | Parameter configuration method, device, equipment, storage medium and product |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191122 |