CN104410799B - A kind of careful method of distributed skill - Google Patents
A kind of careful method of distributed skill Download PDFInfo
- Publication number
- CN104410799B CN104410799B CN201410811306.5A CN201410811306A CN104410799B CN 104410799 B CN104410799 B CN 104410799B CN 201410811306 A CN201410811306 A CN 201410811306A CN 104410799 B CN104410799 B CN 104410799B
- Authority
- CN
- China
- Prior art keywords
- skill
- file
- sub
- segmentation
- video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The present invention relates to a kind of method that distributed skill is examined, described method includes：Receive skill and examine file；File cutting；Skill is segmented to examine；Report merges；" the step of file cutting " includes following sub-step：It is determined that segmentation number；Find the careful most long duration of video skill；Find audio skill and examine most long duration；Be converted to the duration of frame unit；Obtain and overlap length；Determine section length；File fragmentation.Method of the present invention fully utilizes the GPU computing capabilitys of local resource by way of task is segmented.In the case where taking into full account the characteristics of skill is examined and requiring, the overlapping region of segmentation is set up, it is ensured that segmentation does not influence skill to examine quality.Compared with traditional skill examines method, the method of the invention takes full advantage of local computing resource, unserviceable computing capability in conventional method is incorporated into systems approach, the interests of single server resource in conventional method are expanded into local more resources, tasks carrying purpose rapidly and efficiently is reached.
Description
Technical field
The present invention relates to a kind of method that distributed skill is examined, the method for electronic digit processing is a kind of to digital video text
Part is tested the method for processing.
Background technology
With the increase of TV station's portfolio, what the careful efficiency of biometrics program skill can not often meet program to broadcast will
Ask, the particularly technological audit of coded format and video quality is unqualified in program making, it is necessary to expend substantial amounts of hardware resource
Returned when reforming or changing, program completes modification and carries out technological audit again, and current skill, which examines efficiency, can not ensure program
It is efficient to be examined by skill, so as to influence the standby progress broadcast of program, cause program can not be broadcasted in the defined time.
The content of the invention
In order to overcome problem of the prior art, the present invention proposes a kind of careful method of distributed skill.Described method makes
With the mode that distribution skill is careful, the examination burst of time-consuming more coded format and video quality is examined that multithreading is examined,
Improve the careful operating efficiency of skill.
The object of the present invention is achieved like this：A kind of careful method of distributed skill, the system bag that described method is used
Include：The skill in server or server cluster or Cloud Server is set to examine file reception unit, segmenting unit, provided with multithreading
The skill document examination member of video content detecting and alarm and code/decode format detecting and alarm, the report with technical result queue merge single
The step of member, described method, is as follows：
Receive the step of skill examines file：The source file that file reception unit reception requires technological audit is examined for skill；
The step of file cutting：Cutting is carried out to source file for segmenting unit, multiple segmental files are formed；
It is segmented the careful step of skill：Coded format examination and video matter are carried out to each segmental file respectively for skill document examination member
Amount is examined, forms the examining report of each segmentation；
The step of report merges：For reporting that combining unit is merged to each examining report being segmented, form whole
Skill examines the examining report of file；
Described " the step of file cutting ", includes following sub-step：
It is determined that the sub-step of segmentation number：For determining segmentation number according to dicing strategy, partition strategy includes pressing file
The idle number that length is segmented or examines node according to skill determines segmentation number；
Find the sub-step of the careful most long duration of video skill：For traveling through the duration threshold value ginseng that all video skills are examined in item
Number, obtains video skill and examines maximum duration；
Find the sub-step that audio skill examines most long duration：For traveling through the duration threshold parameter that all audio skills are examined in item,
Obtain audio skill and examine maximum duration；
Be converted to the sub-step of the duration of frame unit：For according to video formats, audio duration in seconds to be turned
It is changed to the duration of frame unit；
Obtain the sub-step for overlapping length：For taking video skill to examine in maximum duration and audio duration in units of frame
Maximum is used as the coincidence length in units of frame；
Determine the sub-step of section length：For according to segmentation number, coincidence length, video file total length, it is determined that often
The length of individual segmentation, its in and out point is in units of frame；
The sub-step of file fragmentation：Skill is examined into file fragmentation for the length according to segmentation.
Further, described " the careful step of segmentation skill " includes following sub-step：
The sub-step of decoding：Segmental file is decoded for the position according in and out point, regarding for no compression is decoded as
Frequency evidence；
The careful sub-step of parallel skill：For skill examine node start different threads call respectively video content detecting and alarm and
Code/decode format detecting and alarm is detected to the video/audio without compression.
Further, described " report combining step " includes following sub-step：
Store the sub-step of first testing result：Testing result in for first to be segmented adds skill and concluded fruit team
Row；
The sub-step of the next segmentation of sequence detection：For traveling through the first testing result in next segmentation, if skill is examined
Out points of the out points interval comprising the testing result of the testing result of last in result queue are interval, then continue to scan on this
Next testing result in section, until the out points area of last testing result in queue is examined in its out points interval not comprising skill
Between, form final Ji Shen result queues；
Judge it is interval whether intersecting sub-step：, will most if "Yes" for judging whether former and later two intervals intersect
Whole skill examine last testing result interval in queue be modified to the two it is interval and；If "No" by the testing result and
The testing result of the careful item of all skills is added in final Ji Shen result queues below in the segmentation；
Judge whether to have incorporated the sub-step of all segmentation results：For being added by all testing results of segmentation
After final Ji Shen result queues, judge whether it is to have incorporated all segmentation results, returned to if "No" " under sequence detection
The sub-step of one segmentation ", terminates to merge if "Yes", exports examining report.
The beneficial effect comprise that：Method of the present invention is by the GPU computing capabilitys of local resource by appointing
The mode of business segmentation is fully utilized.In the case where taking into full account the characteristics of skill is examined and requiring, the coincidence area of segmentation is set up
Domain, it is ensured that segmentation does not influence skill to examine quality.Compared with traditional skill examines method, the method for the invention takes full advantage of local
Computing resource, unserviceable computing capability in conventional method is incorporated into systems approach, by single server in conventional method
The interests of resource expand to local more resources, reach tasks carrying purpose rapidly and efficiently.
Brief description of the drawings
The invention will be further described with reference to the accompanying drawings and examples.
Fig. 1 is the system schematic used described in the methods described of embodiments of the invention one；
Fig. 2 is the flow chart of the methods described of embodiments of the invention one；
Fig. 3 is segment logic schematic diagram in the methods described of embodiments of the invention one；
Fig. 4 is the methods described of embodiments of the invention three " report combining step " schematic flow sheet.
Embodiment
Embodiment one：
The present embodiment is a kind of careful method of distributed skill, and the system that described method is used includes：It is arranged on server
Or the skill in server cluster or Cloud Server examines file reception unit, segmenting unit, the video content detection provided with multithreading
Skill document examination member, the report combining unit with Ji Shen result queues of engine and code/decode format detecting and alarm, as shown in Figure 1.
The system used described in the present embodiment methods described can be realized with flexible mode, can be achieved as comprising multiple skills
A set of server of node cluster is examined, each software module can also be deployed in cloud computing platform and realize that the automatic skill of cloud is examined.
If server can be deployed in user's building environment such as TV station, Network TV Station；If the automatic skill of cloud is examined, then pass through
Private clound or public cloud mode provide skill and examine service, can be deployed in the privately owned cloud environment of user, or be deployed to public cloud
On platform.System used in the present embodiment methods described possesses very big retractility, and the automatic skill of burst cloud examines node can be from
Several nodes can make full use of cloud resource to hundreds of nodes, improve the careful efficiency of skill.
Careful report content=file format detection+code/decode format the detection of complete distributed skill+audio-visual content detection
Result.Wherein, file format detection is because efficiency is very high（The treatment effeciency of double to strong E5-2600 platforms be 10 speeds with
On）, along with file format detection also has the problem of file is redirected, therefore examined without distributed skill, only carry out whole section of detection.
Code/decode format is detected and audio-visual content detection performs segment processing.
Method described in the present embodiment comprises the following steps that flow chart is as shown in Figure 2：
Receive the step of skill examines file：The source file that file reception unit reception requires technological audit is examined for skill.Source document
The system that part is examined by request skill is sent, and the careful system of these request skills can be the system or making video text for storing video file
The system of part, or by the various video files of the transmission such as satellite or network.
The step of file cutting：Cutting is carried out to source file for segmenting unit, multiple segmental files are formed.File cutting
The characteristics of skill is examined must be taken into full account, missing inspection is prevented.
The segmentation principle of file cutting is to examine parameter according to skill to carry out most rational segmentation, is both segmented in the absence of redundancy,
Missing inspection will not be produced, to ensure that the careful high efficiency of parallel skill and skill conclude the correctness of fruit.File cutting will be according to video content
Require to carry out cutting of both detection and code/decode format detection.Video content detecting and alarm is according to set section plan first
Slightly know the number of section, the idle number that dicing strategy includes being cut into slices by file designated length or examining node according to skill is true
Fixed section.
Because many detections are that front and rear related, such as Jing Yin detection is to want Jing Yin perdurabgility to exceed certain threshold values
It can just alert, it is necessary to consider overlay region to ensure that burst logical process is accurate, video content detecting and alarm can calculate each
The in and out point information of segmentation.
The length requirement of overlay region mainly considers the duration threshold parameter that video skill is examined in project, these duration threshold values, bag
Include：Continuous how many frames are detected as the frozen frame inspection that the time threshold values of the black field detection of black field, continuous how many frame are detected as frozen frame
Uncorrelated frame within the time threshold values of survey, continuous how many frame can be regarded as time threshold values, the continuous how many frame of the folder frame detection of folder frame
It is detected as the time threshold values of the exceeded detections of the exceeded YUV of YUV, continuous how many frame and is detected as the exceeded inspection of the exceeded brightness of brightness
Time threshold values of survey etc..Duration threshold value in also having consideration audio skill careful simultaneously, including：Continuously how long it is detected as Jing Yin
The time threshold values of Jing Yin detection, the continuous time threshold values for how long being detected as the anti-phase detection of the anti-phase audio of audio, even
Continue and how long be detected as the time threshold values of the exceeded exceeded detection of real peak value of real peak value, continuously how long be detected as ringing
Exceeded time threshold values of the exceeded detection of loudness of degree etc..Because audio duration threshold value is all using the second as unit, and during video
Long threshold value be all using frame as unit, in order to look after video skill examine, therefore, also by the unit conversion using the second as duration into
Using frame as duration unit, to contrast.Again by most long video duration threshold parameter and most long audio duration threshold parameter
Contrast, takes the length of its this maximum effect overlay region.
" the step of file cutting " includes following sub-step：
It is determined that the sub-step of segmentation number：For determining segmentation number according to dicing strategy, partition strategy includes pressing file
The idle number that length is segmented or examines node according to skill determines segmentation number.
Find the sub-step of the careful most long duration of video skill：For traveling through the duration threshold value ginseng that all video skills are examined in item
Number, obtains video skill and examines maximum duration.
Find the sub-step that audio skill examines most long duration：For traveling through the duration threshold parameter that all audio skills are examined in item,
Obtain audio skill and examine maximum duration.
Be converted to the sub-step of the duration of frame unit：For according to video formats, audio duration in seconds to be turned
It is changed to the duration of frame unit.
Obtain the sub-step for overlapping length：For taking video skill to examine in maximum duration and audio duration in units of frame
Maximum is used as the coincidence length in units of frame.
Determine the sub-step of section length：For according to segmentation number, coincidence length, video file total length, it is determined that often
The length of individual segmentation, its in and out point is in units of frame.
The sub-step of file fragmentation：Skill is examined into file fragmentation for the length according to segmentation.The content of segmentation includes two
Point：Code/decode format is detected（Segmental file encoding and decoding are detected）+ audio-visual content is detected（Segmental file audio-visual content is detected）.
On the one hand the segmentation of audio-visual content detection needs to know the number of section according to partition strategy, and partition strategy includes
It is segmented by file designated length or determines segments according to the idle number of segmented node.On the other hand need consider due to
Many detections are that front and rear related, such as Jing Yin detection is to want Jing Yin perdurabgility to exceed certain threshold values can just alert, point
Section bad will result in of logical process is failed to judge, and the segment logic shown in design drawing 3 is needed for this, emphasis has sought to overlay region.
Method is the duration parameters that all skills of statistics examine item, takes maximum duration as the minimum of segmentation and overlaps length, according to segmentation parameter
Segmentation division is completed with length is overlapped（See above several step process）.
Code/decode format detection segmentation, it is necessary to audio-visual content detect segmentation complete on the basis of, be to MPEG2
IBP GOP align.
It is segmented the careful step of skill：Coded format examination and video matter are carried out to each segmental file respectively for skill document examination member
Amount is examined, forms the examining report of each segmentation.Fragmentation technique skill first examines node according to the position of out points to being detected text
Part is decoded, and is decoded as the video/audio of no compression.Skill examines the different threads of node startup and calls video content detection to draw respectively
Hold up and the video/audio without compression is detected with code/decode format detecting and alarm.
The step of report merges：For reporting that combining unit is merged to each examining report being segmented, form whole
Skill examines the examining report of file., it is necessary to carry out duplicate removal and merging treatment to each testing result in order during merging.Using multi-thread
All skills are examined detection by journey technology, and every section of testing result is traveled through in order, form examining report.
Distributed skill examine splicing merge process be exactly in order to each testing result carry out duplicate removal merging treatment mistake
Journey.The testing result of each detection section is got, and is counted.Duplicate removal is reported afterwards, and final report (FR) merges.
In addition, when distributed skill is examined, frame detection will not be reprocessed for overlay region, be can further improve efficiency, also subtracted
Duplicate removal work when frame testing result merges is lacked.
Distributed skill described in the present embodiment, which is examined, can realize many calculate node parallel processings, according to calculate node number, base
Originally linear increment can be reached.The automatic skill for not performing distributed treatment examines task and performs careful of the automatic skill of distributed treatment
It is completely the same that the obtained automatic skill of being engaged in examines report, it is ensured that data it is objective and accurate, the exchange for being also beneficial to data message is shared
And statistical analysis.
Embodiment two：
Originally it is the improvement of embodiment one, is refinement of the embodiment one on " the careful step of segmentation skill ".Described in the present embodiment
" segmentation skill careful step " include following sub-step：
The sub-step of decoding：Segmental file is decoded for the position according in and out point, regarding for no compression is decoded as
Frequency evidence.Because file has been segmented into, this step can carry out decoding simultaneously using multiple threads to each segmentation, accelerate decoding
Speed.
The careful sub-step of parallel skill：For skill examine node start different threads call respectively video content detecting and alarm and
Code/decode format detecting and alarm is detected to the video/audio without compression.Skill examines each frame that material is detected by analysis
Video and voice data, and the data parsed and relevant criterion are subjected to calculating contrast, so that it is determined that being in tested material
It is no and to think problematic local marking point record containing various video and audio problems, used for artificial reinspection.
Skill, which examines content, to be included：Audio-visual content is detected and code/decode format detection.It can be called for video content detection
The main contents that multiple threads are detected simultaneously include:Black field, colour bar, blue bottom, green bottom, frozen frame, station symbol, Y/U/V are exceeded, R/G/
B ultrasound mark, width type, flower screen detection etc..Mainly include in terms of audio:Jing Yin, distorsion, VU, level, stereo phase, loudness are exceeded
Deng detection.In addition, examining the detection range of parameter by adjusting skill, the strategy of detection is adjusted, program file can be accurately positioned and deposited
Regard, audio issue.
Embodiment three：
The present embodiment is the improvement of above-described embodiment, refinement of the above-described embodiment on " report combining step ", such as Fig. 4 institutes
Show." report combining step " described in the present embodiment includes following sub-step：
Store the sub-step of first testing result：Testing result in for first to be segmented adds skill and concluded fruit team
Row.The data meeting dynamic change of Ji Shen result queues, becomes final skill after the completion of merging and concludes during merging
Fruit queue.
The sub-step of the next segmentation of sequence detection：For traveling through the first testing result in next segmentation, if skill is examined
Out points of the out points interval comprising the testing result of the testing result of last in result queue are interval, then continue to scan on this
Next testing result in section, until the out points area of last testing result in queue is examined in its out points interval not comprising skill
Between, form final Ji Shen result queues.This sub-step is report duplicate removal, duplicate removal principle：Front and rear two sections of skill concludes the entering out of fruit
Point is interval to whether there is inclusion relation, if it is present deleting subinterval.
Judge it is interval whether intersecting sub-step：, will most if "Yes" for judging whether former and later two intervals intersect
Whole skill examine last testing result interval in queue be modified to the two it is interval and, if "No" by the testing result and
The testing result of the careful item of all skills is added in final Ji Shen result queues below in the segmentation.This sub-step is closed for report
And, combination principle：The conclude out points interval of fruit of front and rear two sections of skill does not include relation with the presence or absence of intersecting, if it is present
By two interval merging.
Judge whether to have incorporated the sub-step of all segmentation results：For being added by all testing results of segmentation
After final Ji Shen result queues, judge whether it is to have incorporated all segmentation results, returned to if "No" " under sequence detection
The sub-step of one segmentation ", terminates to merge if "Yes", exports examining report.
Complete skill examines file Codec detection knot complete after report content=file format test result+merging
Complete file audio-visual content testing result after really+merging., please while feeding back to after the completion of the careful report of skill in write into Databasce
The initiator for asking skill careful.
Finally it should be noted that being merely illustrative of the technical solution of the present invention and unrestricted above, although with reference to preferable cloth
Scheme is put the present invention is described in detail, it will be understood by those within the art that, can be to technology of the invention
Scheme（The calculating of such as section length, the calculating of overlap length, sequencing of step etc.）Modify or equivalent substitution,
Without departing from the spirit and scope of technical solution of the present invention.
Claims (3)
1. a kind of careful method of distributed skill, the system that described method is used includes：It is arranged on server or server cluster
Or the skill in Cloud Server examines file reception unit, segmenting unit, the video content detecting and alarm provided with multithreading and encoding and decoding
The step of skill document examination member, the report combining unit with technical result queue of format detection engine, described method, is as follows：
Receive the step of skill examines file：The source file that file reception unit reception requires technological audit is examined for skill；
The step of file cutting：Cutting is carried out to source file for segmenting unit, multiple segmental files are formed；
It is segmented the careful step of skill：Coded format examination is carried out to each segmental file respectively for skill document examination member and video quality is examined
Look into, form the examining report of each segmentation；
The step of report merges：For reporting that combining unit is merged to each examining report being segmented, form whole skill and examine
The examining report of file；
Characterized in that, described " the step of file cutting " includes following sub-step：
It is determined that the sub-step of segmentation number：For determining segmentation number according to dicing strategy, partition strategy includes pressing file size
The idle number for being segmented or examining node according to skill determines segmentation number；
Find the sub-step of the careful most long duration of video skill：For traveling through the duration threshold parameter that all video skills are examined in item, obtain
Maximum duration is examined to video skill；
Find the sub-step that audio skill examines most long duration：For traveling through the duration threshold parameter that all audio skills are examined in item, obtain
Audio skill examines maximum duration；
Be converted to the sub-step of the duration of frame unit：For according to video formats, audio duration in seconds to be converted to
The duration of frame unit；
Obtain the sub-step for overlapping length：For the maximum for taking video skill to examine in maximum duration and audio duration in units of frame
Value is used as the coincidence length in units of frame；
Determine the sub-step of section length：For according to segmentation number, coincidence length, video file total length, it is determined that each point
The length of section, its in and out point is in units of frame；
The sub-step of file fragmentation：Skill is examined into file fragmentation for the length according to segmentation.
2. according to the method described in claim 1, it is characterised in that described " the careful step of segmentation skill " includes following sub-step
Suddenly：
The sub-step of decoding：Segmental file is decoded for the position according in and out point, the video counts of no compression are decoded as
According to；
The careful sub-step of parallel skill：Examine the different threads of node startup for skill calls video content detecting and alarm and volume to solve respectively
Code format detection engine is detected to the video/audio without compression.
3. method according to claim 1 or 2, it is characterised in that described " report combining step " includes following son
Step：
Store the sub-step of first testing result：Testing result in for first to be segmented adds Ji Shen result queues；
The sub-step of the next segmentation of sequence detection：For traveling through the first testing result in next segmentation, the fruit if skill concludes
Out points of the out points interval comprising the testing result of the testing result of last in queue are interval, then continue to scan in the section
Next testing result, until its out points interval is interval not comprising the out points that skill examines last testing result in queue,
Form final Ji Shen result queues；
Judge it is interval whether intersecting sub-step：For judging whether former and later two intervals intersect, by final skill if "Yes"
Examine last testing result interval in queue be modified to the two it is interval and；By the testing result and this point if "No"
The testing result of the careful item of all skills is added in final Ji Shen result queues below in section；
Judge whether to have incorporated the sub-step of all segmentation results：For being added finally by all testing results of segmentation
After Ji Shen result queues, judge whether it is to have incorporated all segmentation results, " next point of sequence detection is returned to if "No"
The sub-step of section ", terminates to merge if "Yes", exports examining report.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410811306.5A CN104410799B (en) | 2014-12-24 | 2014-12-24 | A kind of careful method of distributed skill |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410811306.5A CN104410799B (en) | 2014-12-24 | 2014-12-24 | A kind of careful method of distributed skill |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104410799A CN104410799A (en) | 2015-03-11 |
| CN104410799B true CN104410799B (en) | 2017-10-31 |
Family
ID=52648386
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410811306.5A Active CN104410799B (en) | 2014-12-24 | 2014-12-24 | A kind of careful method of distributed skill |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104410799B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104837033B (en) * | 2015-05-20 | 2018-09-25 | 腾讯科技（北京）有限公司 | A kind of information processing method and server |
| CN105488100A (en) * | 2015-11-18 | 2016-04-13 | 国信司南(北京)地理信息技术有限公司 | Efficient detection and discovery system for secret-associated geographic data in non secret-associated environment |
| CN105578203B (en) * | 2016-01-15 | 2018-10-19 | 北京赛特斯信息科技股份有限公司 | Video file quality detecting system and method |
| CN113225623B (en) * | 2021-04-30 | 2023-01-31 | 四川云从天府人工智能科技有限公司 | Adaptive video file accelerated parsing method, device, equipment and storage medium |
| CN116489421A (en) * | 2023-05-08 | 2023-07-25 | 湖南芒果无际科技有限公司 | Video AI intelligent detection method and system based on file slicing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102006499A (en) * | 2010-12-10 | 2011-04-06 | 北京中科大洋科技发展股份有限公司 | Method for detecting video and audio quality of digital television program file |
| CN103873852A (en) * | 2012-12-11 | 2014-06-18 | 上海文广互动电视有限公司 | Multi-mode parallel video quality fault detection method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110271307A1 (en) * | 2009-12-18 | 2011-11-03 | Tektronix International Sales Gmbh | Video data stream evaluation systems and methods |
-
2014
- 2014-12-24 CN CN201410811306.5A patent/CN104410799B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102006499A (en) * | 2010-12-10 | 2011-04-06 | 北京中科大洋科技发展股份有限公司 | Method for detecting video and audio quality of digital television program file |
| CN103873852A (en) * | 2012-12-11 | 2014-06-18 | 上海文广互动电视有限公司 | Multi-mode parallel video quality fault detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104410799A (en) | 2015-03-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104410799B (en) | A kind of careful method of distributed skill | |
| JP6937988B2 (en) | Dynamic video overlay | |
| CN102006499B (en) | Method for detecting video and audio quality of digital television program file | |
| CN107977922B (en) | Image analysis method, device and system | |
| CN103873852A (en) | Multi-mode parallel video quality fault detection method and device | |
| US20110052136A1 (en) | Pattern-based monitoring of media synchronization | |
| CN111641869B (en) | Video split mirror method, video split mirror device, electronic equipment and computer readable storage medium | |
| US8358842B2 (en) | Electronic device with function of separating panels of digital comic strip and method thereof | |
| CN104506853A (en) | System and method for technical reviewing of audios/videos | |
| US8675498B2 (en) | System and method to provide aggregated alarm indication signals | |
| CN103176996A (en) | Image search method based on image feature information and image search engine server based on image feature information | |
| WO2019019767A1 (en) | Client identity information processing method and apparatus, storage medium and computer device | |
| CN108710918B (en) | Fusion method and device for multi-mode information of live video | |
| CN108134746B (en) | Method and device for processing rail transit data | |
| CN109426438B (en) | Real-time big data mirror image storage method and device | |
| CN111741328B (en) | Video analysis method, electronic device, storage medium and system | |
| CN109040784A (en) | Commercial detection method and device | |
| CN106789209B (en) | Exception handling method and device | |
| CN110941949B (en) | Data table rendering control method and device, computer equipment and storage medium | |
| WO2018177018A1 (en) | Multi-tuner set-top box production detection method, system and detection device | |
| JP2012204878A (en) | Video data stream evaluation method and system | |
| CN112312208A (en) | Multimedia information processing method and device, storage medium and electronic equipment | |
| CN115953715B (en) | Video detection method, device, equipment and storage medium | |
| CN112346874A (en) | Abnormal volume processing method and device based on cloud platform | |
| CN107979778B (en) | Video analysis method, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |