The content of the invention
In order to solve the problems, such as video location in the prior art, the embodiments of the invention provide a kind of media content positioningMethod and system, it is special in video by concentrating each local video operator of the mode of positioning video accurately to obtainThe location information of frame of video is determined, so as to carry out a variety of applications.
The embodiments of the invention provide a kind of media content positioning method, including：
Central server and a plurality of carrier servers obtain media source file；
The central server is analyzed the media source file, obtain based on programme content include locating frame andLocation information including real-time time；
The location information is sent to the matching unit of a plurality of carrier servers by the central server；
The matching unit of the carrier server is carried out according to the location information to the media source file of the acquisitionMatching, obtains the accurate signature identification of media source file, including：A frame of video is determined in media source file using real-time timeScope, then found in the frame of video scope with the locating frame identical frame of video, it is final determine only one with it is describedLocating frame identical frame of video is as signature identification；
The carrier server is cut the media source file according to the signature identification, generates piece of videoSection, and store corresponding label；When receiving when reviewing instruction of user terminal transmission, review instruction according to described and search out accordinglyLabel, according to the label to the user terminal play video segment.
A further aspect of a kind of described media content positioning method according to embodiments of the present invention, the programContent includes programm name, character features, or scene characteristic；The locating frame is that a certain frame in the media source file regardsFrequency image.
Another further aspect of a kind of described media content positioning method according to embodiments of the present invention, is directed toThe location information of live broadcast stream media also includes real-time time and/or timer reference of programs, for the location information of file Streaming MediaAlso include File Percent and/or voice data.
Another further aspect of a kind of described media content positioning method according to embodiments of the present invention, in describedCentral server also generates the label related to location information, programme content is recorded in the label, and the label is sent to againThe matching unit of several carrier servers.
Another further aspect of a kind of described media content positioning method according to embodiments of the present invention, is being obtainedAlso include after the accurate signature identification of media source file, the carrier server determines according to the signature identification and labelThe programme content of signature identification, described in the frame of video insertion related information where signature identification, when being played to the frame of videoThe related information is sent to user terminal.
Another further aspect of a kind of described media content positioning method according to embodiments of the present invention, the fortuneThe matching unit for seeking business's server is directed to the live broadcast stream media recorded using the real-time time and/or program clock in location informationBenchmark carries out Primary Location to the media source file, recycles the locating frame to match video in the section of Primary LocationFrame, find with the locating frame identical frame of video as signature identification；
Or the matching unit of the carrier server uses the file percentage in location information for file Streaming MediaThan and/or voice data carry out Primary Location, recycle the locating frame to match frame of video in the section of Primary Location, findWith the locating frame identical frame of video as signature identification.
The embodiment of the present invention additionally provides a kind of system of media content positioning, including：
Central server, a plurality of carrier servers, the matching unit of carrier server；
The central server and a plurality of carrier servers obtain media source file；
The central server is analyzed the media source file, obtains existing including locating frame based on programme contentInterior location information, the location information is sent to the matching unit of a plurality of carrier servers；
The matching unit of the carrier server is carried out according to the location information to the media source file of the acquisitionMatching, obtains accurate signature identification, including：A frame of video scope is determined in media source file using real-time time, thenFound in the frame of video scope identical with the locating frame with the locating frame identical frame of video, final determination only oneFrame of video as signature identification；
The carrier server also includes video segment cutter unit, memory cell, search unit, transmitting element；InstituteState video segment cutter unit to be cut the media source file according to the signature identification, generate video segment；It is describedMemory cell, for label to be carried out into binding storage with corresponding video segment；The search unit, for being sent out according to user terminalThe label reviewed instruction and find corresponding video fragment sent；The transmitting element, for according to the label to the user terminalPlay video segment.
A further aspect of described a kind of media content positioning system according to embodiments of the present invention, the centerServer also includes tag unit, generates the label related to location information, programme content is recorded in the label, genuinely convinced in describedThe label is sent to the matching unit of a plurality of carrier servers by business device.
Another further aspect of described a kind of media content positioning system according to embodiments of the present invention, the fortuneBattalion's business's server also includes determining unit, for determining the programme content of signature identification according to the signature identification and label；InsertEnter unit, in the frame of video insertion related information description where signature identification；Transmitting element, when being played to the frame of videoThe related information is sent to user terminal.
The embodiment of the present invention realizes media fragmentation by the transmission based on video content location information and media frame is closedJoin content information, can be widely applied to DTV review and Internet video in carry out for insertion advertisement, the industry such as presentation of informationBusiness.
Embodiment
Following description can make any those skilled in the art using the present invention.Provided in specific embodiment and applicationDescription information it is merely illustrative.The various extensions and combination of embodiment as described herein are aobvious for those skilled in the artAnd be clear to, without departing from the spirit and scope of the present invention, the rule that the present invention defines may apply to itIn his embodiment and application.Therefore, the present invention is not limited solely to shown embodiment, and the present invention covers and principle illustrated herein and spyLevy consistent maximum magnitude.
Following detailed description provides in the form of flow chart, logic module and other symbol manipulations are expressed, Ke YiPerformed in computer system.One program, computer perform step, logical block, process etc., are envisioned for obtaining herein desiredThe one or more steps of result or the self-consistent sequence of instruction.These steps are the physical operations to physical quantity.These physicsAmount includes electricity, magnetic or radio signal, and they are stored, transmitted in computer systems, combined, compared and other behaviourMake.These signals can be bit, numerical value, element, symbol, character, condition, numeral etc..Each step can be by hardware, softPart, firmware or combinations thereof perform.
Fig. 1 gives the flow chart of media content positioning method a kind of according to one embodiment of present invention.
Including step 101, central server and a plurality of carrier servers obtain media source file.
Programming source can be DVB-C either DVB-S or other modes.
Step 102, the central server is analyzed the media source file, obtains including based on programme contentLocation information including locating frame.
The programme content can include programm name, character features, or scene characteristic etc..
The locating frame is a certain frame video image in the media source file.
Being directed to the location information of live broadcast stream media also includes real-time time and/or timer reference of programs (PCR), for textThe location information of part Streaming Media also includes File Percent and/or voice data.
Step 103, the location information is sent to the matchings of a plurality of carrier servers and set by the central serverIt is standby.
The location information is fabricated to after xml document and is sent to a plurality of carrier services by the central serverThe matching unit of device, wherein, the operator includes digital TV network carrier server or internet video provides fortuneSeek business's server.
The central server also generates the label related to location information, records programme content in the label, and by instituteState the matching unit that label is sent to a plurality of carrier servers.
Step 104, source of media of the matching unit of the carrier server according to the location information to the acquisitionFile is matched, and obtains the accurate signature identification of media source file.
Also include after above-mentioned steps 104, the carrier server is according to the signature identification by the source of mediaFile is cut, and generates video segment, and store corresponding label；When receive user terminal transmission when reviewing instruction, rootReview instruction according to described and search out corresponding label, video segment is played to the user terminal according to the label.
Also include after above-mentioned steps 104, the carrier server determines special according to the signature identification and labelLevy mark programme content, where signature identification frame of video insertion related information description, when being played to the frame of video toUser terminal sends the related information.
The matching unit of the carrier server for record live broadcast stream media use location information in it is real-time whenBetween and/or timer reference of programs to the media source file carry out Primary Location, recycle the locating frame in Primary LocationFrame of video is matched in section, is found with the locating frame identical frame of video as signature identification.
The matching unit of the carrier server for file Streaming Media use location information in File Percent and/Or voice data carries out Primary Location, recycles the locating frame to match frame of video in the section of Primary Location, finds and instituteLocating frame identical frame of video is stated as signature identification.
, can be by the unified accurate positioning video section destination locations of mode by the above embodiments, will not be due toThe difference of video playback is caused in the geographical position of each place operator, network delay；And can be in video frequency program it is appropriateMoment adds related information, such as advertisement or recommended information etc., improves the convenience of user profile acquisition；By based on regardingThe transmission of frequency content location information realizes media fragmentation and associates content information with media frame, can be widely applied to DTVReview and Internet video is directed to advertising lamp business.
It is illustrated in figure 2 a kind of system construction drawing of media content positioning of the embodiment of the present invention.
Including central server 201, a plurality of carrier servers 202, the matching unit 203 of carrier server.
The central server 201 and a plurality of carrier servers 202 obtain media source file；
The central server 201 is analyzed the media source file, obtains including positioning based on programme contentLocation information including frame, the location information is sent to the matching unit 203 of a plurality of carrier servers.
The matching unit 203 of the carrier server enters according to the location information to the media source file of the acquisitionRow matching, obtains accurate signature identification.
The central server 201 also includes tag unit 2011, the generation label related to location information, in the labelProgramme content is recorded, the label is sent to the matching unit of a plurality of carrier servers by the central server 201203。
The carrier server 202 also includes video segment cutter unit 2021, memory cell 2022, search unit2023, transmitting element 2024；The video segment cutter unit 2021 enters the media source file according to the signature identificationRow cutting, generates video segment；The memory cell 2022, for label to be carried out into binding storage with corresponding video segment；The search unit 2023, for the label reviewed instruction and find corresponding video fragment sent according to user terminal；The transmissionUnit 2024, for playing video segment to the user terminal according to the label.
The carrier server 202 also includes determining unit 2025, for according to the signature identification and label determinationThe programme content of signature identification；Unit 2026 is inserted, in the frame of video insertion related information description where signature identification；HairUnit 2024 is sent, the related information is sent to user terminal when being played to the frame of video.
The carrier server 202 can individually only include above-mentioned transmitting element 2024, determining unit 2025 and insertionUnit 2026.
, can be by the unified accurate positioning video section destination locations of mode by the above embodiments, will not be due toThe difference of video playback is caused in the geographical position of each place operator, network delay；And the present invention realizes media fragmentationContent information is associated with media frame, can be widely applied to that DTV is reviewed and Internet video is directed to advertising lamp business.
It is illustrated in figure 3 the flow chart that media content positioning of the embodiment of the present invention is applied to review business.
Include in the present embodiment, positioned at Pekinese's carrier server 1, the carrier server 2 positioned at Shanghai, be located atThe time delay that Pekinese's carrier server 1 receives live media source file is 30 seconds, and the carrier server 2 positioned at Shanghai connectsThe time delay for receiving live media source file is 45 seconds, and carrier server 1 and carrier server 2 and central server 3 are simultaneouslyLive media source file is received by digital video broadcast for cable (DVB-C) or DTV satellite broadcasting (DVB-S), it is describedLive media source file is recorded on carrier server 1 and carrier server 2, for reviewing user terminal in businessReview.In conventional technology user terminal review be using the time in electronic program guides (EPG) as play starting point, but byCertain delay is respectively provided with when carrier server 1 and carrier server 2 are recording program, so being sent back to user terminalVideo frequency program corresponding to seeing includes time delay, i.e. and the video frequency program sent to user always postpones 30 seconds or 45 seconds,User terminal can not accurately review the beginning of video frequency program.
Solve the above problems by the following means in embodiments of the present invention：
Step 301, central server 3 is carried out by carrying out video analysis to live media source file, such as to TV play AVideo analysis, using the frame of video at the beginning and end of TV play A as locating frame, and record current real-time time, exampleSuch as the time started 8：30, or the timer reference of programs recorded when starting is 8:30:20, wherein because each second has 25 framesFrame of video, a wide range of time when real-time time is record location frame, timer reference of programs are the timestamp of media source file, shouldTimestamp records the time range smaller than real-time time, can only all indicate to regard in real-time time or timer reference of programsThe scope of frequency frame, specifically which frame of video can not be accurately indicated, in this example location information only include real-time time andLocating frame, in other examples location information can also include timer reference of programs.
Central server 3 can also be it is determined that record television play A title information be as should during locating frame in this stepThe label of location information.
Step 302, the location information including locating frame, real-time time and label is made as xml by the central server 3File, the carrier server 2 positioned at Pekinese's carrier server 1 and positioned at Shanghai is sent to by satellite.
Step 303, the carrier server 2 positioned at Pekinese's carrier server 1 and positioned at Shanghai is utilized respectively describedLocation information matches to live media source file, finds accurate signature identification.
Wherein, approximate range is determined in media source file using real-time time, real-time time is 8 in this example:30,1 minute frame of video scope is exactly determined, then searching and the locating frame identical frame of video in the frame of video of 1 minute,It is final to determine only one with the locating frame identical frame of video as signature identification.It can be avoided by method soComplex calculation and the operation of locating frame are matched in the frame of video of magnanimity, mitigates system burden, and shorten match time.
The signature identification in media source file can be obtained by above-mentioned mode carrier server, the spy can be utilizedSign represents to carry out a variety of operations, to meet the needs of different user.
Step 304, positioned at Beijing carrier server 1 and positioned at Shanghai carrier server 2 according to the feature markKnowledge is cut to the media source file, generation TV play A video segment, and records mark corresponding with the video segmentLabel, that is, record the programm name of the video segment：TV play A.
Wherein, the location information not only includes frame of video when TV play A starts, and can also terminate including TV play AWhen frame of video, the signature identification in the media source file also just identifies TV play beginning and end, cutWhen can be cut according to the signature identification of TV play beginning and end.
Step 305, receive user terminal transmission reviews instruction, and this is reviewed instruction and wants to review program including at least userThe numbering of title or program, it is in this example the title TV play A of program.
Step 306, according to it is described review instructions match described in label, be sent to user after searching out corresponding video segmentEnd.
Although in this example positioned at Beijing carrier server 1 and positioned at Shanghai carrier server 2 in record tv play AWhen there is delay, if finding the programme content reviewed according to programm name, the programm name corresponds to standard in the epgTime, and the time of the standard is not actually the initial time (because delay 30 seconds) for recording the program, therefore to user30 seconds and the TV play A to be reviewed incoherent other programme contents can be sent when sending the program.
By above-mentioned method, locating frame and label can accurate positioning television play A beginning, no matter operator postponeHow much, if using frame of video as location information, it can accurately find the starting point for the TV play A to be reviewed.
It is illustrated in figure 4 the flow chart that media content positioning of the embodiment of the present invention is applied to content identification.
Include in the present embodiment, internet video server 1, central server 2, the media source file in the present embodimentFor the static program file of file Streaming Media, static program file is had been positioned in carrier server 1 and central server 2.
Step 401, central server 2 is analyzed static program file, will appear from regarding for particular persons or thingsFrequency frame records the percentage for currently belonging to whole static program file as locating frame, such as 5%, or current soundFrequency information, the File Percent can be used for the position for substantially determining locating frame, can also be saved using audio-frequency information in static statePositioned in mesh file, it is hereby achieved that the position of locating frame.The positioning of the File Percent and audio-frequency information can only refer toShow the scope of frame of video, specifically which frame of video can not be accurately indicated, by locating frame determine occur particular persons orThe position of person's things, only includes File Percent in the present embodiment.
Central server 2 is also it is determined that record the information of particular persons or things during locating frame as should in this stepThe label of location information.
Step 402, the location information including locating frame, File Percent and label is made as by the central server 2Xml document, internet video server 1 is sent to by satellite.
Step 403, the internet video server 1 using the location information to live media source file carry outMatch somebody with somebody, find accurate signature identification.
Wherein, approximate range is determined in media source file using File Percent, File Percent in this example is5%, frame of video scope is determined in 5% position of the media source file, is then found and locating frame phase in this section of frame of videoSame frame of video, it is final to determine only one with the locating frame identical frame of video as signature identification.
Step 404, internet video server 1 determines the programme content of signature identification according to label, is identified in this featureThe frame of video insertion related information description at place.
For example, locating frame is to occur performer A frame of video in media source file in this example, label is the surname of the performer AName, the related information of insertion are described as the brief introduction of the performer A.
Step 405, the order triggered when the frame of video where receiving user terminal and being played to this feature mark, it is described mutualNetworked video server 1 sends the brief introduction of the performer A to user terminal.
, can be by the unified accurate positioning video section destination locations of mode by the above embodiments, will not be due toThe difference of video playback is caused in the geographical position of each place operator, network delay；And can be in video frequency program it is appropriateMoment adds related information, such as advertisement or recommended information etc., improves the convenience of user profile acquisition；By based on regardingThe transmission of frequency content location information realizes media fragmentation and associates content information with media frame, can be widely applied to DTVReview and Internet video is directed to advertising lamp business.
The present invention can realize in any suitable form, including hardware, software, firmware or their any combination.ThisInvention can according to circumstances selectively part be realized, for example, software performing in one or more data processors andDigital signal processor.The element and component of this paper each embodiment can be fitted with any physically, functionally, in logicWhen mode realize.In fact, One function can in separate unit, in one group of unit or be used as other functional unitsA part realize.Therefore, the system and method can both be realized in separate unit, can also be physically and functionallyIt is distributed between different unit and processor.
Technical staff in the related art will recognize that embodiments of the invention have many possible modifications and groupClose, although form is slightly different, still using identical fundamental mechanism and method.For purposes of explanation, it is described above to reference toSeveral specific embodiments.However, above-mentioned illustrative discussion is not intended to the precise forms that exhaustive or limitation is invented herein.BeforeShown in text, many modifications and variations are possible.Selected and described embodiment, to explain the principle and in fact of the present inventionBorder is applied, to enable repairing for application-specific of the those skilled in the art best using of the invention and each embodimentChange, deform.