CN106454492A - Live pornographic content audit system and method based on delayed transmission - Google Patents
Live pornographic content audit system and method based on delayed transmission Download PDFInfo
- Publication number
- CN106454492A CN106454492A CN201610895718.0A CN201610895718A CN106454492A CN 106454492 A CN106454492 A CN 106454492A CN 201610895718 A CN201610895718 A CN 201610895718A CN 106454492 A CN106454492 A CN 106454492A
- Authority
- CN
- China
- Prior art keywords
- key frame
- live
- video
- frame
- pornograph
- 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
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
- H04N21/433—Content storage operation, e.g. storage operation in response to a pause request, caching operations
- H04N21/4331—Caching operations, e.g. of an advertisement for later insertion during playback
-
- 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
- H04N21/254—Management at additional data server, e.g. shopping server, rights management server
- H04N21/2541—Rights Management
-
- 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
- H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
- H04N21/25866—Management of end-user data
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
Abstract
The invention discloses a live pornographic content audit system and method based on delayed transmission and relates to the field of network technologies. The system comprises an collection module, a sampling module, an analysis module and a content audit system server, wherein the collection module is used for collecting a live video, storing the collected live video into a buffer queue and sending the live video at the preset duration in the buffer queue to a live content server; the sampling module is used for extracting video frames from the buffer queue as key frames in sequence according to the collection time of the live video; the analysis module is used for carrying out image analysis on the key frames and sending the key frames including sensitive characteristics to the content audit system server; a control module stores the received key frames into a database, deletes the key frames judged as non-pornographic contents by audit staff and sends broadcasting stop information of the live video corresponding to the key frames judged as pornographic contents by the audit staff to the live content server. According to the system and the method, the manual cost is reduced, the higher audit accuracy rate is guaranteed and the audit efficiency is improved.
Description
The present invention relates to field of cloud calculation, it is specifically related to a kind of live Pornograph auditing system based on time delay transmission
And method.
Background technology
With the fast development of live industry, increasing people enters live industry initially as main broadcaster, constantly rich
While rich spectators' variation viewing demand, the live content also resulting in main broadcaster's production in live industry is very different, for example, certain
A little main broadcasters, in order to improve popularity, make a desperate move, and propagate obscene Pornograph, not only contrary to law, Er Qie on live platform
Extremely ill effect is caused, therefore, live platform all can introduce substantial amounts of auditor, to straight on live platform in society
Broadcast content to be monitored.But, manual examination and verification efficiency is low, high cost, and the workload of auditor is big, is sometimes difficult to send out in time
Existing Pornograph, with the rapid growth of live content, live platform faces huge pressure in terms of live content examination ＆ verification.
Content of the invention
For defect present in prior art, present invention is primarily targeted at providing a kind of straight based on time delay transmission
Broadcast Pornograph auditing system, another object of the present invention is to providing a kind of live Pornograph examination ＆ verification based on time delay transmission
Method, can reduce the examination ＆ verification link in live video for the live platform personnel put into, reduce cost of labor, guarantee higher
While examination ＆ verification accuracy rate, improve review efficiency.
The present invention provides a kind of live Pornograph auditing system based on time delay transmission, by client and webcast website
Live content server carry out transmission of video, described system includes acquisition module, decimation blocks, analysis module and content auditing
System server；
Acquisition module, it includes buffer queue, and acquisition module is used for gathering live video, and the live video collecting is deposited
Enter buffer queue, and the live video of preset duration in buffer queue is sent to live content server；
Decimation blocks, extract frame of video conduct successively for the acquisition time order according to live video from buffer queue
Key frame, the time interval of adjacent key frame is the default sampling interval；
Analysis module, for constantly obtaining key frame from decimation blocks, carries out graphical analyses to key frame, will include quick
The key frame of sense feature is sent to content auditing system server；
Content auditing system server, it includes control module database, and control module is used for the key frame that will receive
It is stored in data base, auditor is judged to the crucial frame deletion of non-Pornograph, the key frame institute of Pornograph will be judged to
The stopping of corresponding live video is broadcasted information and is sent to live content server.
On the basis of technique scheme, described cache module, decimation blocks and analysis module are all in client.
On the basis of technique scheme, sensitive features are trained from the human body collected in advance exposed sensitive part sample
Obtain；
Described analysis module includes Face Detection unit and characteristic detection unit；
Face Detection unit is used for described key frame smoothing denoising, is gone out including human body skin by skin detection model discrimination
The key frame of skin；
The key frame that characteristic detection unit is used for filtering out Face Detection unit is compared with sensitive features, will include
The key frame of sensitive features is sent to content auditing system server.
On the basis of technique scheme, described analysis module also includes number of people detector unit, and described number of people detection is single
For being compared with human eye feature to the key frame that Face Detection unit filters out, human eye feature is from the number of people collected in advance for unit
In sample, training obtains, and the key frame including human eye feature is marked；
Characteristic detection unit is additionally operable to filter out the key frame including suspicious sensitive features, judges unlabelled described pass
Key frame includes sensitive features；In the described key frame of judge mark, whether human eye feature is identical with the region that sensitive features are located,
If so, judge that this key frame does not include sensitive features；If it is not, judging that this key frame includes sensitive features.
On the basis of technique scheme, skin detection model include human body skin tone testing algorithm, morphological analysis and
Connected domain analysis.
The present invention also provides a kind of live Pornograph checking method based on time delay transmission it is characterised in that including step
Suddenly：
S1. acquisition module collection live video, the live video collecting is stored in buffer queue；
S2a. the live video of preset duration in buffer queue is sent to live content server by acquisition module；
S2b. decimation blocks acquisition time order according to live video from buffer queue extracts frame of video conduct successively
Key frame, the time interval of adjacent key frame is the default sampling interval, enters S3；
S3. analysis module constantly obtains key frame from decimation blocks, carries out graphical analyses to key frame, will include sensitivity
The key frame of feature is sent to content auditing system server；
S4. the key frame of reception is stored in data base by control module, auditor is judged to the key of non-Pornograph
Frame deletion, information is broadcasted in the stopping of the live video being judged to corresponding to the key frame of Pornograph and is sent to live content clothes
Business device.
On the basis of technique scheme, step S2b specifically includes：
S2b.1 decimation blocks obtain initial frame of video and current video frame from the live video buffer queue respectively
Acquisition time, using initial frame of video as last time sample video frame；
S2b.2 calculates the duration of live video, the basis formula of described duration：The collection of duration=current video frame
When m- last time sample video frame acquisition time；
S2b.3 judges that described duration, whether up to or over the default sampling interval, if so, enters S2b.4；If it is not,
Enter S2b.5；
S2b.4 extracts a frame of video as key frame from live video, and the acquisition time of this key frame was sampled with last time
The time interval of the acquisition time of frame of video is the default sampling interval；
S2b.5 obtains the acquisition time of current video frame from the live video buffer queue；
S2b.6 judges whether the acquisition time of current video frame is identical with the acquisition time of last time sample video frame, if so,
Terminate；If it is not, entering S2b.2.
On the basis of technique scheme, described analysis module includes Face Detection unit, characteristic detection unit and people
Head detector unit；Face Detection unit includes skin detection model；Number of people detector unit includes human eye feature, and human eye feature is from pre-
Train in the number of people sample first collected and obtain；
Step S3 includes：
S3.1 obtains a key frame as current key frame from decimation blocks；
By skin detection model, S3.2 Face Detection unit, to current key frame smoothing denoising, judges that current key frame is
No inclusion human body skin, if so, enters S3.3；If it is not, entering S3.6；
Current key frame is compared by S3.3 characteristic detection unit with sensitive features, filters out including suspicious sensitive spy
The key frame levied；Current key frame is compared by number of people detector unit with human eye feature, to the key frame including human eye feature
It is marked, characteristic detection unit judges that unlabelled described key frame includes sensitive features；
In the described key frame of S3.4 characteristic detection unit judge mark, the region at human eye feature and sensitive features place is
No identical, if so, enter S3.8；If it is not, entering S3.5；
Current key frame is sent to content auditing system server by S3.5 characteristic detection unit；
Whether there is the next key frame of current key frame in S3.6 judgment sampling module, if so, enter S3.7；If it is not,
Terminate；
S3.7 key frame as after update using next key frame, enters S3.2；
S3.8 judges that this key frame does not include sensitive features.
On the basis of technique scheme, step S4 includes：
The key frame of reception is stored in data base, the sum of the described key frame of statistics storage by control module；
Judge whether the sum of described key frame reaches default examination ＆ verification threshold value, if so, all key frames are sent to careful
Core personnel；If it is not, ignoring this sum；
Auditor is judged to the crucial frame deletion of non-Pornograph, will be judged to corresponding to the key frame of Pornograph
Live video stopping broadcast information be sent to live content server.
Compared with prior art, advantages of the present invention is as follows：
(1) present invention, by the way of automatic detection combines artificial reinspection, can reduce live platform in live video
The personnel of examination ＆ verification link put into, and reduce cost of labor, mitigate the workload of auditor, and false drop rate is extremely low, is guaranteeing
While higher examination ＆ verification accuracy rate, improve review efficiency.
(2) live video is stored in buffer queue by the present invention, filters out including quick from the live video buffer queue
The key frame of sense feature, meanwhile, live video is sent to live content server, and live content server is according to receiving
Stop broadcast information, do not broadcast the live video corresponding to the key frame being judged to Pornograph, due to the time delay of buffer queue
Effect, can block the propagation of Pornograph in time.
(3) live video is stored in buffer queue by the present invention, due to all entering in client to the extraction of key frame and analysis
OK, only the key frame including sensitive features is sent to content auditing system server, then carries out manual examination and verification, therefore, pole
The earth reduces transmission of video amount and the transmission time between client and content auditing system server, and alleviates content
The load of auditing system server, further increases review efficiency, reduces the cost of live platform.
Brief description
Fig. 1 is the live Pornograph auditing system schematic diagram that the embodiment of the present invention is transmitted based on time delay；
Fig. 2 is the live Pornograph checking method flow chart that the embodiment of the present invention is transmitted based on time delay；
Fig. 3 is the particular flow sheet of step S2b；
Fig. 4 is the particular flow sheet of step S3.
Reference：
Acquisition module 1, decimation blocks 2, analysis module 3, content auditing system server 4, control module 41, data base
42.
Specific embodiment
Term explanation：
Time delay transmits：Frame of video is cached in the server, reaches certain cache-time and send user again to.
Image procossing：By obtaining image, image is carried out with the operation such as feature extraction, eigentransformation, image recognition, from figure
Some useful tolerance, data or information is extracted in picture.
Live Pornograph：Live main broadcaster is carried out on live platform, with the exposed message for main demand of human body, its
Purpose is to provoke the live spectators' libido of initiation, broadly falls into live Pornograph.
Below in conjunction with the accompanying drawings and specific embodiment the present invention is described in further detail.
Shown in Figure 1, the embodiment of the present invention provides a kind of live Pornograph auditing system based on time delay transmission, leads to
The live content server crossing client and webcast website carries out transmission of video, the system include acquisition module 1, decimation blocks 2,
Analysis module 3 and content auditing system server 4.
Acquisition module 1, it includes buffer queue, and acquisition module 1 is used for gathering live video, by the live video collecting
It is stored in buffer queue, and the live video of preset duration in buffer queue is sent to live content server.
Decimation blocks 2, extract frame of video successively for the acquisition time order according to live video from buffer queue and make
For key frame, the time interval of adjacent key frame is the default sampling interval.
Analysis module 3, for constantly obtaining key frame from decimation blocks 2, carries out graphical analyses to key frame, will include
The key frame of sensitive features is sent to content auditing system server 4.
Analysis module 3 includes Face Detection unit and characteristic detection unit.
Face Detection unit is used for key frame smoothing denoising, is gone out including human body skin by skin detection model discrimination
Key frame.
Skin detection model includes human body skin tone testing algorithm, morphological analysis and connected domain analysis.
Human body skin tone testing algorithm mainly divides according to the model that human body skin is in sub-elliptical in YCrCb color space
Cloth, by the statistics to a large amount of colour of skin samples, simulates the parameter meeting colour of skin model of ellipse, when subsequently judging, only needs to judge
Whether the colouring information of this position is in this model of ellipse.
The area of skin color that Preliminary detection goes out is likely to occur empty and discontinuous situation, and area of skin color is not especially complete
Whole, by morphological analysis, be filled with can to these cavities and discontinuously, finally give the mask figure with connection agglomerate
Picture, morphological analysis include corrosion and expansive working.
Connected domain analysis are used for detecting the area of the connection agglomerate in mask image, delete the little connection agglomerate of area, face
Long-pending little connection agglomerate is probably the flase drop that illumination variation is brought, after deletion on subsequent analysis impact less.
The key frame that characteristic detection unit is used for filtering out Face Detection unit is compared with sensitive features, will include
The key frame of sensitive features is sent to content auditing system server 4.
Because general live Pornograph includes the exposed sensitive part of human body, the present invention uses human body skin mask permissible
Exclude most non-pornographic live content, significantly improve the screening efficiency of analysis module 3.
Sensitive features are trained from the human body collected in advance exposed sensitive part sample and are obtained.
Sensitive features can obtain the model of the exposed sensitive part of human body as quick using the training of Adaboost iterative algorithm
Sense feature, in conjunction with human body skin mask, the sensitive features that training is obtained enter line slip contrast in key frame, obtain human body skin
Whether sensitive features are included in skin mask.
Because human breast position and human eye region are easier to produce flase drop, in order to improve the standard of analysis module 3
Really rate, can increase a number of people detector unit, by the detection to head, can exclude the flase drop leading to due to eyes.
Specifically, analysis module 3 also includes number of people detector unit, and number of people detector unit is used for Face Detection unit is screened
The key frame going out and human eye feature are compared, and human eye feature is trained from the number of people sample collected in advance and obtained, and the number of people detects
Unit is marked to the key frame including human eye feature.
Number of people sample in a large amount of live scene is trained using Adaboost iterative algorithm, obtains a number of people
Model, as human eye feature, human eye feature is entered in key frame line slip contrast, whether has the number of people and people in detection key frame
Eye.
Characteristic detection unit is additionally operable to filter out the key frame including suspicious sensitive features, judges unlabelled key frame
Including sensitive features；In the key frame of judge mark, whether human eye feature is identical with the region that sensitive features are located, and if so, judges
This key frame does not include sensitive features；If it is not, judging that this key frame includes sensitive features.
The present invention can filter out key frame including human body skin by Face Detection unit, then passes through feature detection
This key frame and sensitive features are compared by unit, and the key frame including sensitive features is sent to content auditing system service
Device 4, or, the key frame including human body skin is filtered out by Face Detection unit, passes through number of people detector unit and spy respectively
Levy detector unit the key frame including human body skin is screened further, for the key frame only including sensitive features, sentence
This key frame fixed includes sensitive features；For the key frame including human eye feature and sensitive features simultaneously, compare human eye feature and
Whether the region that sensitive features are located is identical, if so, judges that this key frame does not include sensitive features；If it is not, judging this key frame
Including sensitive features.
Cache module, decimation blocks 2 and analysis module 3 are all in client.
Live video is stored in buffer queue by the present invention, filters out including sensitive special from the live video buffer queue
The key frame levied, meanwhile, live video is sent to live content server, and live content server is according to the stopping receiving
Broadcast information, does not broadcast the live video corresponding to the key frame being judged to Pornograph, due to the time-lag action of buffer queue,
The propagation of Pornograph can be blocked in time.
Content auditing system server 4, it includes control module 41 database 42, and control module 41 is used for reception
Key frame is stored in data base 42, the crucial frame deletion that auditor is judged to non-Pornograph, will be judged to Pornograph
The stopping of the live video corresponding to key frame is broadcasted information and is sent to live content server.
Live video is stored in buffer queue by the present invention, due to all carrying out in client to the extraction of key frame and analysis,
Only the key frame including sensitive features is sent to content auditing system server 4, then carries out manual examination and verification, therefore, greatly
Reduce transmission of video amount and transmission time between client and content auditing system server 4, and alleviate content
The load of auditing system server 4, further increases review efficiency, reduces the cost of live platform.
Shown in Figure 2, the embodiment of the present invention provides a kind of live Pornograph checking method based on time delay transmission, bag
Include step：
S1. acquisition module 1 collection live video, the live video collecting is stored in buffer queue.
S2a. the live video of preset duration in buffer queue is sent to live content server by acquisition module 1.
S2b. decimation blocks 2 acquisition time order according to live video from buffer queue extracts frame of video conduct successively
Key frame, the time interval of adjacent key frame is the default sampling interval, enters S3, and step S2a and S2b are carried out simultaneously.
Shown in Figure 3, step S2b specifically includes：
S2b.1 decimation blocks 2 obtain initial frame of video and current video frame from the live video buffer queue respectively
Acquisition time, using initial frame of video as last time sample video frame.
S2b.2 calculates the duration of live video, the basis formula of duration：During the collection of duration=current video frame
The acquisition time of m- last time sample video frame.
S2b.3 judges that duration, whether up to or over the default sampling interval, if so, enters S2b.4；If it is not, entering
S2b.5.
S2b.4 extracts a frame of video as key frame from live video, and the acquisition time of this key frame was sampled with last time
The time interval of the acquisition time of frame of video is the default sampling interval.
S2b.5 obtains the acquisition time of current video frame from the live video buffer queue.
S2b.6 judges whether the acquisition time of current video frame is identical with the acquisition time of last time sample video frame, if so,
Terminate；If it is not, entering S2b.2.
S3. analysis module 3 constantly obtains key frame from decimation blocks 2, carries out graphical analyses to key frame, will include quick
The key frame of sense feature is sent to content auditing system server 4.
Analysis module 3 includes Face Detection unit, characteristic detection unit and number of people detector unit；Face Detection unit includes
Skin detection model；Number of people detector unit includes human eye feature, and human eye feature is trained from the number of people sample collected in advance and obtained.
Skin detection model includes human body skin tone testing algorithm, morphological analysis and connected domain analysis.
Human body skin tone testing algorithm mainly divides according to the model that human body skin is in sub-elliptical in YCrCb color space
Cloth, by the statistics to a large amount of colour of skin samples, simulates the parameter meeting colour of skin model of ellipse, when subsequently judging, only needs to judge
Whether the colouring information of this position is in this model of ellipse.
The area of skin color that Preliminary detection goes out is likely to occur empty and discontinuous situation, and area of skin color is not especially complete
Whole, by morphological analysis, be filled with can to these cavities and discontinuously, finally give the mask figure with connection agglomerate
Picture, morphological analysis include corrosion and expansive working.
Connected domain analysis are used for detecting the area of the connection agglomerate in mask image, delete the little connection agglomerate of area, face
Long-pending little connection agglomerate is probably the flase drop that illumination variation is brought, after deletion on subsequent analysis impact less.
Shown in Figure 4, step S3 includes：
S3.1 obtains a key frame as current key frame from decimation blocks 2.
By skin detection model, S3.2 Face Detection unit, to current key frame smoothing denoising, judges that current key frame is
No inclusion human body skin, if so, enters S3.3；If it is not, entering S3.6.
Current key frame is compared by S3.3 characteristic detection unit with sensitive features, filters out including suspicious sensitive spy
The key frame levied；Current key frame is compared by number of people detector unit with human eye feature, to the key frame including human eye feature
It is marked, characteristic detection unit judges that unlabelled key frame includes sensitive features.
The region that in the key frame of S3.4 characteristic detection unit judge mark, human eye feature and sensitive features are located whether phase
With if so, entrance S3.8；If it is not, entering S3.5.
Current key frame is sent to content auditing system server 4 by S3.5 characteristic detection unit.
Whether there is the next key frame of current key frame in S3.6 judgment sampling module 2, if so, enter S3.7；If it is not,
Terminate.
S3.7 key frame as after update using next key frame, enters S3.2；
S3.8 judges that this key frame does not include sensitive features.
S4. the key frame of reception is stored in data base 42 by control module 41, and auditor is judged to non-Pornograph
Crucial frame deletion, by the stopping of the live video being judged to corresponding to the key frame of Pornograph broadcast information be sent to live interior
Hold server.
Step S4 includes：
The key frame of reception is stored in data base 42, the sum of the key frame of statistics storage by control module 41.
Judge whether the sum of key frame reaches default examination ＆ verification threshold value, if so, all key frames are sent to person approving
Member；If it is not, ignoring this sum.
Auditor is judged to the crucial frame deletion of non-Pornograph, will be judged to corresponding to the key frame of Pornograph
Live video stopping broadcast information be sent to live content server.
Auditor according to circumstances punishes to main broadcaster, gently then alerts, heavy then directly close direct broadcasting room, thoroughly blocks color
The propagation of feelings content.
By the way of the present invention combines artificial reinspection using automatic detection, the examination ＆ verification in live video for the live platform can be reduced
The personnel of link put into, and reduce cost of labor, mitigate the workload of auditor, and false drop rate is extremely low, guarantee higher
Examination ＆ verification accuracy rate while, improve review efficiency, purify Living Network, for spectators provide more preferably, the content of more high-quality.
The present invention is not limited to above-mentioned embodiment, for those skilled in the art, without departing from
On the premise of the principle of the invention, some improvements and modifications can also be made, these improvements and modifications are also considered as the protection of the present invention
Within the scope of.The content not being described in detail in this specification belongs to prior art known to professional and technical personnel in the field.
Claims (9)
1. a kind of live Pornograph auditing system based on time delay transmission, by the live content clothes of client and webcast website
Business device carry out transmission of video it is characterised in that：Described system includes acquisition module, decimation blocks, analysis module and content auditing
System server；
Acquisition module, it includes buffer queue, and acquisition module is used for gathering live video, the live video collecting is stored in slow
Deposit queue, and the live video of preset duration in buffer queue is sent to live content server；
Decimation blocks, extract frame of video as key successively for the acquisition time order according to live video from buffer queue
Frame, the time interval of adjacent key frame is the default sampling interval；
Analysis module, for constantly obtaining key frame from decimation blocks, carries out graphical analyses to key frame, will include sensitive spy
The key frame levied is sent to content auditing system server；
Content auditing system server, it includes control module database, and control module is used for being stored in the key frame of reception
Data base, auditor is judged to the crucial frame deletion of non-Pornograph, will be judged to corresponding to the key frame of Pornograph
Live video stopping broadcast information be sent to live content server.
2. the live Pornograph auditing system based on time delay transmission as claimed in claim 1 it is characterised in that：Described caching
Module, decimation blocks and analysis module are all in client.
3. the live Pornograph auditing system based on time delay transmission as claimed in claim 1 it is characterised in that：Sensitive features
Train from the human body collected in advance exposed sensitive part sample and obtain；
Described analysis module includes Face Detection unit and characteristic detection unit；
Face Detection unit is used for described key frame smoothing denoising, is gone out including human body skin by skin detection model discrimination
Key frame；
The key frame that characteristic detection unit is used for filtering out Face Detection unit is compared with sensitive features, will include sensitivity
The key frame of feature is sent to content auditing system server.
4. the live Pornograph auditing system based on time delay transmission as claimed in claim 3 it is characterised in that：Described analysis
Module also includes number of people detector unit, and described number of people detector unit is used for key frame and the human eye that Face Detection unit is filtered out
Feature is compared, and human eye feature is trained from the number of people sample collected in advance and obtained, and the key frame including human eye feature is entered
Line flag；
Characteristic detection unit is additionally operable to filter out the key frame including suspicious sensitive features, judges unlabelled described key frame
Including sensitive features；In the described key frame of judge mark, whether human eye feature is identical with the region that sensitive features are located, if so,
Judge that this key frame does not include sensitive features；If it is not, judging that this key frame includes sensitive features.
5. as described in claim 3 or 4 based on time delay transmission live Pornograph auditing system it is characterised in that：Skin
Detection model includes human body skin tone testing algorithm, morphological analysis and connected domain analysis.
6. a kind of live Pornograph checking method being transmitted based on time delay based on the arbitrary described system of claim 1 to 5, its
It is characterised by, including step：
S1. acquisition module collection live video, the live video collecting is stored in buffer queue；
S2a. the live video of preset duration in buffer queue is sent to live content server by acquisition module；
S2b. decimation blocks extract frame of video as key successively according to the acquisition time order of live video from buffer queue
Frame, the time interval of adjacent key frame is the default sampling interval, enters S3；
S3. analysis module constantly obtains key frame from decimation blocks, carries out graphical analyses to key frame, will include sensitive features
Key frame be sent to content auditing system server；
S4. the key frame of reception is stored in data base by control module, and the key frame that auditor is judged to non-Pornograph is deleted
Remove, information is broadcasted in the stopping of the live video being judged to corresponding to the key frame of Pornograph and is sent to live content service
Device.
7. the live Pornograph checking method based on time delay transmission as claimed in claim 6 is it is characterised in that step S2b
Specifically include：
S2b.1 decimation blocks obtain initial frame of video and the collection of current video frame from the live video buffer queue respectively
Time, using initial frame of video as last time sample video frame；
S2b.2 calculates the duration of live video, the basis formula of described duration：During the collection of duration=current video frame
The acquisition time of m- last time sample video frame；
S2b.3 judges that described duration, whether up to or over the default sampling interval, if so, enters S2b.4；If it is not, entering
S2b.5；
S2b.4 extracts a frame of video as key frame, the acquisition time of this key frame and last time Sample Video from live video
The time interval of the acquisition time of frame is the default sampling interval；
S2b.5 obtains the acquisition time of current video frame from the live video buffer queue；
S2b.6 judges whether the acquisition time of current video frame is identical with the acquisition time of last time sample video frame, if so, ties
Bundle；If it is not, entering S2b.2.
8. the live Pornograph checking method based on time delay transmission as claimed in claim 6 it is characterised in that：
Described analysis module includes Face Detection unit, characteristic detection unit and number of people detector unit；Face Detection unit includes
Skin detection model；Number of people detector unit includes human eye feature, and human eye feature is trained from the number of people sample collected in advance and obtained；
Step S3 includes：
S3.1 obtains a key frame as current key frame from decimation blocks；
By skin detection model, S3.2 Face Detection unit, to current key frame smoothing denoising, judges whether current key frame wraps
Include human body skin, if so, enter S3.3；If it is not, entering S3.6；
Current key frame is compared by S3.3 characteristic detection unit with sensitive features, filters out including suspicious sensitive features
Key frame；Current key frame is compared by number of people detector unit with human eye feature, and the key frame including human eye feature is carried out
Labelling, characteristic detection unit judges that unlabelled described key frame includes sensitive features；
The region that in the described key frame of S3.4 characteristic detection unit judge mark, human eye feature and sensitive features are located whether phase
With if so, entrance S3.8；If it is not, entering S3.5；
Current key frame is sent to content auditing system server by S3.5 characteristic detection unit；
Whether there is the next key frame of current key frame in S3.6 judgment sampling module, if so, enter S3.7；If it is not, terminating；
S3.7 key frame as after update using next key frame, enters S3.2；
S3.8 judges that this key frame does not include sensitive features.
9. the live Pornograph checking method based on time delay transmission as claimed in claim 6 is it is characterised in that step S4 bag
Include：
The key frame of reception is stored in data base, the sum of the described key frame of statistics storage by control module；
Judge whether the sum of described key frame reaches default examination ＆ verification threshold value, if so, all key frames are sent to person approving
Member；If it is not, ignoring this sum；
Auditor is judged to the crucial frame deletion of non-Pornograph, straight corresponding to the key frame of Pornograph by being judged to
The stopping broadcast information broadcasting video is sent to live content server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610895718.0A CN106454492A (en) | 2016-10-12 | 2016-10-12 | Live pornographic content audit system and method based on delayed transmission |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610895718.0A CN106454492A (en) | 2016-10-12 | 2016-10-12 | Live pornographic content audit system and method based on delayed transmission |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106454492A true CN106454492A (en) | 2017-02-22 |
Family
ID=58173522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610895718.0A Pending CN106454492A (en) | 2016-10-12 | 2016-10-12 | Live pornographic content audit system and method based on delayed transmission |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106454492A (en) |
Cited By (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107181744A (en) * | 2017-05-16 | 2017-09-19 | 北京奇艺世纪科技有限公司 | A kind of Video processing and coding method, processor and encoder |
| CN107197331A (en) * | 2017-05-03 | 2017-09-22 | 北京奇艺世纪科技有限公司 | A kind of method and device of real-time monitoring live content |
| CN107197370A (en) * | 2017-06-22 | 2017-09-22 | 北京密境和风科技有限公司 | The scene detection method and device of a kind of live video |
| CN107277558A (en) * | 2017-06-19 | 2017-10-20 | 网宿科技股份有限公司 | A kind of player client for realizing that live video is synchronous, system and method |
| CN107404656A (en) * | 2017-06-26 | 2017-11-28 | 武汉斗鱼网络科技有限公司 | Live video recommends method, apparatus and server |
| CN107590443A (en) * | 2017-08-23 | 2018-01-16 | 上海交通大学 | Limiter stage live video automatic testing method and system based on the study of depth residual error |
| CN107948713A (en) * | 2017-10-31 | 2018-04-20 | 北京飞利信电子技术有限公司 | A kind of delay live method and system |
| CN108040262A (en) * | 2018-01-25 | 2018-05-15 | 湖南机友科技有限公司 | Live audio and video are reflected yellow method and device in real time |
| CN108055289A (en) * | 2018-01-30 | 2018-05-18 | 深圳市富途网络科技有限公司 | A kind of method and system audited to user-generated content based on internet |
| CN109005425A (en) * | 2018-08-26 | 2018-12-14 | 俞绍富 | Network video monitor and control system |
| CN109492952A (en) * | 2018-12-28 | 2019-03-19 | 广州华多网络科技有限公司 | Audit data processing method, device, electronic equipment and storage medium |
| CN109600622A (en) * | 2018-08-31 | 2019-04-09 | 北京微播视界科技有限公司 | Audio/video information processing method, device and electronic equipment |
| WO2019076187A1 (en) * | 2017-10-16 | 2019-04-25 | 杭州海康威视数字技术股份有限公司 | Video blocking region selection method and apparatus, electronic device, and system |
| CN109783689A (en) * | 2018-12-28 | 2019-05-21 | 广州华多网络科技有限公司 | Information processing method, device and electronic equipment |
| CN109829379A (en) * | 2018-12-28 | 2019-05-31 | 广州华多网络科技有限公司 | Information processing method, device, server and storage medium |
| CN109831696A (en) * | 2018-12-28 | 2019-05-31 | 广州华多网络科技有限公司 | Handle method, apparatus, electronic equipment and the storage medium of violation video content |
| CN109831697A (en) * | 2018-12-28 | 2019-05-31 | 广州华多网络科技有限公司 | The detection method and system of violation handling duration |
| WO2019127951A1 (en) * | 2017-12-29 | 2019-07-04 | 深圳市太赫兹科技创新研究院 | Method, apparatus and system for collecting terahertz time-domain pulse signal |
| CN110113544A (en) * | 2019-05-07 | 2019-08-09 | 上海墨工文化传播有限公司 | A kind of extreme video backstage editing system |
| CN110191356A (en) * | 2019-06-06 | 2019-08-30 | 北京字节跳动网络技术有限公司 | Video reviewing method, device and electronic equipment |
| CN110222523A (en) * | 2019-06-26 | 2019-09-10 | 新华网(北京)科技有限公司 | Detection method, device, system and computer readable storage medium |
| CN110298213A (en) * | 2018-03-22 | 2019-10-01 | 北京深鉴智能科技有限公司 | Video analytic system and method |
| CN110674678A (en) * | 2019-08-07 | 2020-01-10 | 国家计算机网络与信息安全管理中心 | Method and device for identifying sensitive mark in video |
| CN110796053A (en) * | 2019-10-21 | 2020-02-14 | 北京奇艺世纪科技有限公司 | Video detection method and device, electronic equipment and computer readable storage medium |
| CN110830795A (en) * | 2020-01-09 | 2020-02-21 | 武汉精立电子技术有限公司 | Image sensor dead pixel detection method and device and electronic equipment |
| CN110944217A (en) * | 2019-12-12 | 2020-03-31 | 北京达佳互联信息技术有限公司 | Live broadcast information processing method and device, server and storage medium |
| CN111385601A (en) * | 2018-12-29 | 2020-07-07 | 华为技术有限公司 | Video auditing method and system |
| CN111770353A (en) * | 2020-06-24 | 2020-10-13 | 北京字节跳动网络技术有限公司 | Live broadcast monitoring method and device, electronic equipment and storage medium |
| CN108574851B (en) * | 2017-12-11 | 2021-02-05 | 北京金山云网络技术有限公司 | Push method, device, streaming media server and computer readable storage medium |
| CN112347990A (en) * | 2020-11-30 | 2021-02-09 | 重庆空间视创科技有限公司 | Multimode-based intelligent manuscript examining system and method |
| CN112532961A (en) * | 2020-12-04 | 2021-03-19 | 上海影创信息科技有限公司 | Delay detection safety prompting method and system of VR glasses and VR glasses |
| CN112528122A (en) * | 2020-12-07 | 2021-03-19 | 江苏新华烽火数据技术有限公司 | Risk identification method based on image-text and video |
| CN112565655A (en) * | 2020-11-27 | 2021-03-26 | Oppo广东移动通信有限公司 | Video data yellow identification method and device, electronic equipment and storage medium |
| WO2021136125A1 (en) * | 2019-12-31 | 2021-07-08 | 中兴通讯股份有限公司 | Method and system for taking down content, and storage medium |
| CN113315983A (en) * | 2021-05-17 | 2021-08-27 | 唐晓晖 | Live frame transmission system for 5G and 4G network aggregation |
| US20220286716A1 (en) * | 2021-03-08 | 2022-09-08 | Beijing Dajia Internet Information Technology Co., Ltd. | Method and apparatus for live streaming, server, system and storage medium |
| CN116033194A (en) * | 2022-12-15 | 2023-04-28 | 中国联合网络通信集团有限公司 | Illegal content management and control method and system based on mobile terminal |
| CN116527999A (en) * | 2023-05-29 | 2023-08-01 | 国脉通信规划设计有限公司 | 5G wireless video acquisition system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704966A (en) * | 2004-05-28 | 2005-12-07 | 中国科学院计算技术研究所 | Method for detecting pornographic images |
| CN101441717A (en) * | 2007-11-21 | 2009-05-27 | 中国科学院计算技术研究所 | Method and system for detecting eroticism video |
| CN101996314A (en) * | 2009-08-26 | 2011-03-30 | 厦门市美亚柏科信息股份有限公司 | Content-based human body upper part sensitive image identification method and device |
| CN102324025A (en) * | 2011-09-06 | 2012-01-18 | 北京航空航天大学 | Human face detection and tracking method based on Gaussian skin color model and feature analysis |
| CN102377684A (en) * | 2010-08-11 | 2012-03-14 | 腾讯科技（深圳）有限公司 | Method, terminal and system for automatically blocking transmission of pornographic video files |
| CN102567738A (en) * | 2012-01-06 | 2012-07-11 | 华南理工大学 | Rapid detection method for pornographic videos based on Gaussian distribution |
| CN105049934A (en) * | 2015-08-14 | 2015-11-11 | 无锡天脉聚源传媒科技有限公司 | Programmed video review processing method and device |
| CN105915953A (en) * | 2016-06-12 | 2016-08-31 | 腾讯科技（深圳）有限公司 | Method, apparatus and system for discriminating live video |
-
2016
- 2016-10-12 CN CN201610895718.0A patent/CN106454492A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704966A (en) * | 2004-05-28 | 2005-12-07 | 中国科学院计算技术研究所 | Method for detecting pornographic images |
| CN101441717A (en) * | 2007-11-21 | 2009-05-27 | 中国科学院计算技术研究所 | Method and system for detecting eroticism video |
| CN101996314A (en) * | 2009-08-26 | 2011-03-30 | 厦门市美亚柏科信息股份有限公司 | Content-based human body upper part sensitive image identification method and device |
| CN102377684A (en) * | 2010-08-11 | 2012-03-14 | 腾讯科技（深圳）有限公司 | Method, terminal and system for automatically blocking transmission of pornographic video files |
| CN102324025A (en) * | 2011-09-06 | 2012-01-18 | 北京航空航天大学 | Human face detection and tracking method based on Gaussian skin color model and feature analysis |
| CN102567738A (en) * | 2012-01-06 | 2012-07-11 | 华南理工大学 | Rapid detection method for pornographic videos based on Gaussian distribution |
| CN105049934A (en) * | 2015-08-14 | 2015-11-11 | 无锡天脉聚源传媒科技有限公司 | Programmed video review processing method and device |
| CN105915953A (en) * | 2016-06-12 | 2016-08-31 | 腾讯科技（深圳）有限公司 | Method, apparatus and system for discriminating live video |
Cited By (54)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107197331A (en) * | 2017-05-03 | 2017-09-22 | 北京奇艺世纪科技有限公司 | A kind of method and device of real-time monitoring live content |
| CN107197331B (en) * | 2017-05-03 | 2020-01-31 | 北京奇艺世纪科技有限公司 | method and device for monitoring live broadcast content in real time |
| CN107181744A (en) * | 2017-05-16 | 2017-09-19 | 北京奇艺世纪科技有限公司 | A kind of Video processing and coding method, processor and encoder |
| CN107277558A (en) * | 2017-06-19 | 2017-10-20 | 网宿科技股份有限公司 | A kind of player client for realizing that live video is synchronous, system and method |
| CN107197370A (en) * | 2017-06-22 | 2017-09-22 | 北京密境和风科技有限公司 | The scene detection method and device of a kind of live video |
| CN107404656A (en) * | 2017-06-26 | 2017-11-28 | 武汉斗鱼网络科技有限公司 | Live video recommends method, apparatus and server |
| CN107590443A (en) * | 2017-08-23 | 2018-01-16 | 上海交通大学 | Limiter stage live video automatic testing method and system based on the study of depth residual error |
| US11321945B2 (en) | 2017-10-16 | 2022-05-03 | Hangzhou Hikvision Digital Technology Co., Ltd. | Video blocking region selection method and apparatus, electronic device, and system |
| WO2019076187A1 (en) * | 2017-10-16 | 2019-04-25 | 杭州海康威视数字技术股份有限公司 | Video blocking region selection method and apparatus, electronic device, and system |
| CN107948713B (en) * | 2017-10-31 | 2020-10-27 | 北京飞利信电子技术有限公司 | Delayed live broadcasting method and system |
| CN107948713A (en) * | 2017-10-31 | 2018-04-20 | 北京飞利信电子技术有限公司 | A kind of delay live method and system |
| CN108574851B (en) * | 2017-12-11 | 2021-02-05 | 北京金山云网络技术有限公司 | Push method, device, streaming media server and computer readable storage medium |
| WO2019127951A1 (en) * | 2017-12-29 | 2019-07-04 | 深圳市太赫兹科技创新研究院 | Method, apparatus and system for collecting terahertz time-domain pulse signal |
| CN108040262A (en) * | 2018-01-25 | 2018-05-15 | 湖南机友科技有限公司 | Live audio and video are reflected yellow method and device in real time |
| CN108055289A (en) * | 2018-01-30 | 2018-05-18 | 深圳市富途网络科技有限公司 | A kind of method and system audited to user-generated content based on internet |
| CN110298213A (en) * | 2018-03-22 | 2019-10-01 | 北京深鉴智能科技有限公司 | Video analytic system and method |
| CN110298213B (en) * | 2018-03-22 | 2021-07-30 | 赛灵思电子科技(北京)有限公司 | Video analysis system and method |
| CN109005425A (en) * | 2018-08-26 | 2018-12-14 | 俞绍富 | Network video monitor and control system |
| CN109600622A (en) * | 2018-08-31 | 2019-04-09 | 北京微播视界科技有限公司 | Audio/video information processing method, device and electronic equipment |
| CN109600622B (en) * | 2018-08-31 | 2021-04-02 | 北京微播视界科技有限公司 | Audio and video information processing method and device and electronic equipment |
| CN109783689B (en) * | 2018-12-28 | 2021-05-21 | 广州华多网络科技有限公司 | Information processing method and device and electronic equipment |
| CN109492952A (en) * | 2018-12-28 | 2019-03-19 | 广州华多网络科技有限公司 | Audit data processing method, device, electronic equipment and storage medium |
| CN109492952B (en) * | 2018-12-28 | 2022-03-25 | 广州华多网络科技有限公司 | Auditing data processing method and device, electronic equipment and storage medium |
| CN109829379B (en) * | 2018-12-28 | 2021-09-21 | 广州华多网络科技有限公司 | Information processing method, information processing apparatus, server, and storage medium |
| CN109783689A (en) * | 2018-12-28 | 2019-05-21 | 广州华多网络科技有限公司 | Information processing method, device and electronic equipment |
| CN109829379A (en) * | 2018-12-28 | 2019-05-31 | 广州华多网络科技有限公司 | Information processing method, device, server and storage medium |
| CN109831697B (en) * | 2018-12-28 | 2021-05-21 | 广州华多网络科技有限公司 | Method and system for detecting violation processing duration |
| CN109831696A (en) * | 2018-12-28 | 2019-05-31 | 广州华多网络科技有限公司 | Handle method, apparatus, electronic equipment and the storage medium of violation video content |
| CN109831696B (en) * | 2018-12-28 | 2021-03-12 | 广州华多网络科技有限公司 | Method and device for processing illegal video content, electronic equipment and storage medium |
| CN109831697A (en) * | 2018-12-28 | 2019-05-31 | 广州华多网络科技有限公司 | The detection method and system of violation handling duration |
| CN111385601A (en) * | 2018-12-29 | 2020-07-07 | 华为技术有限公司 | Video auditing method and system |
| CN110113544A (en) * | 2019-05-07 | 2019-08-09 | 上海墨工文化传播有限公司 | A kind of extreme video backstage editing system |
| CN110191356A (en) * | 2019-06-06 | 2019-08-30 | 北京字节跳动网络技术有限公司 | Video reviewing method, device and electronic equipment |
| CN110222523B (en) * | 2019-06-26 | 2021-10-15 | 新华网(北京)科技有限公司 | Detection method, device, system and computer readable storage medium |
| CN110222523A (en) * | 2019-06-26 | 2019-09-10 | 新华网(北京)科技有限公司 | Detection method, device, system and computer readable storage medium |
| CN110674678A (en) * | 2019-08-07 | 2020-01-10 | 国家计算机网络与信息安全管理中心 | Method and device for identifying sensitive mark in video |
| CN110796053A (en) * | 2019-10-21 | 2020-02-14 | 北京奇艺世纪科技有限公司 | Video detection method and device, electronic equipment and computer readable storage medium |
| CN110796053B (en) * | 2019-10-21 | 2022-07-29 | 北京奇艺世纪科技有限公司 | Video detection method and device, electronic equipment and computer readable storage medium |
| CN110944217A (en) * | 2019-12-12 | 2020-03-31 | 北京达佳互联信息技术有限公司 | Live broadcast information processing method and device, server and storage medium |
| WO2021136125A1 (en) * | 2019-12-31 | 2021-07-08 | 中兴通讯股份有限公司 | Method and system for taking down content, and storage medium |
| CN110830795A (en) * | 2020-01-09 | 2020-02-21 | 武汉精立电子技术有限公司 | Image sensor dead pixel detection method and device and electronic equipment |
| CN111770353A (en) * | 2020-06-24 | 2020-10-13 | 北京字节跳动网络技术有限公司 | Live broadcast monitoring method and device, electronic equipment and storage medium |
| CN112565655A (en) * | 2020-11-27 | 2021-03-26 | Oppo广东移动通信有限公司 | Video data yellow identification method and device, electronic equipment and storage medium |
| CN112347990B (en) * | 2020-11-30 | 2024-02-02 | 重庆空间视创科技有限公司 | Multi-mode-based intelligent manuscript examining system and method |
| CN112347990A (en) * | 2020-11-30 | 2021-02-09 | 重庆空间视创科技有限公司 | Multimode-based intelligent manuscript examining system and method |
| CN112532961A (en) * | 2020-12-04 | 2021-03-19 | 上海影创信息科技有限公司 | Delay detection safety prompting method and system of VR glasses and VR glasses |
| CN112528122A (en) * | 2020-12-07 | 2021-03-19 | 江苏新华烽火数据技术有限公司 | Risk identification method based on image-text and video |
| US20220286716A1 (en) * | 2021-03-08 | 2022-09-08 | Beijing Dajia Internet Information Technology Co., Ltd. | Method and apparatus for live streaming, server, system and storage medium |
| US11601690B2 (en) * | 2021-03-08 | 2023-03-07 | Beijing Dajia Internet Information Technology Co., Ltd. | Method and apparatus for live streaming, server, system and storage medium |
| CN113315983B (en) * | 2021-05-17 | 2022-11-11 | 唐晓晖 | Live frame signal transmission system for 5G and 4G network aggregation |
| CN113315983A (en) * | 2021-05-17 | 2021-08-27 | 唐晓晖 | Live frame transmission system for 5G and 4G network aggregation |
| CN116033194A (en) * | 2022-12-15 | 2023-04-28 | 中国联合网络通信集团有限公司 | Illegal content management and control method and system based on mobile terminal |
| CN116527999A (en) * | 2023-05-29 | 2023-08-01 | 国脉通信规划设计有限公司 | 5G wireless video acquisition system |
| CN116527999B (en) * | 2023-05-29 | 2023-09-12 | 国脉通信规划设计有限公司 | 5G wireless video acquisition system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106454492A (en) | Live pornographic content audit system and method based on delayed transmission | |
| CN106998444B (en) | Big data face monitoring system | |
| CN108170813A (en) | A kind of method and its system of full media content intelligent checks | |
| CN104504377B (en) | A kind of passenger on public transport degree of crowding identifying system and method | |
| CN106934326A (en) | Method, system and equipment for safety inspection | |
| CN105631427A (en) | Suspicious personnel detection method and system | |
| CN106485217B (en) | A kind of method and system of identification visit scenic spot stream of people saturation degree | |
| CN108364374A (en) | Face access control device based on deep learning and method | |
| CN107087211A (en) | A kind of anchor shots detection method and device | |
| CN104240349A (en) | Method for quickly confirming real name identity in important place and human image and identity comparison safety inspection system | |
| CN106529373A (en) | Human face recognition monitoring alarm method and system | |
| CN109302586A (en) | A kind of structuring face snap camera and corresponding video monitoring system | |
| CN107205137A (en) | One kind monitoring recognition methods and device | |
| CN106780875A (en) | A kind of recognition of face gate inhibition integrated apparatus and its recognition methods | |
| CN104123565A (en) | Identity card authentication and holder identity authentication method based on multimodal identification | |
| CN105225420B (en) | Air traffic controller's fatigue detection method based on principal component analysis, device and system | |
| CN110493574B (en) | Security monitoring visualization system based on streaming media and AI technology | |
| CN107690051A (en) | One kind alarm video recording method and device | |
| CN107358190A (en) | A kind of image key area management method and device | |
| CN108012117A (en) | A kind of city based on background modeling is set up a stall detection method | |
| CN113038375A (en) | Method and system for sensing and positioning hidden camera | |
| CN108174198B (en) | Video image quality diagnosis analysis detection device and application system | |
| CN111242096B (en) | People number gradient-based people group distinguishing method | |
| CN111931567A (en) | Human body recognition method and device, electronic equipment and storage medium | |
| CN109275045A (en) | Mobile terminal encrypted video ad traffic recognition methods based on DFI |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20170222 |