CN105052161B - The system and method for real-time television purposes of commercial detection - Google Patents
The system and method for real-time television purposes of commercial detection Download PDFInfo
- Publication number
- CN105052161B CN105052161B CN201480015936.XA CN201480015936A CN105052161B CN 105052161 B CN105052161 B CN 105052161B CN 201480015936 A CN201480015936 A CN 201480015936A CN 105052161 B CN105052161 B CN 105052161B
- Authority
- CN
- China
- Prior art keywords
- frame
- candidate segment
- coherent
- candidate
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 91
- 238000001514 detection method Methods 0.000 title claims description 16
- 238000012360 testing method Methods 0.000 claims abstract description 53
- 230000001427 coherent effect Effects 0.000 claims description 81
- 238000013500 data storage Methods 0.000 claims description 74
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 10
- 230000001052 transient effect Effects 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000012790 confirmation Methods 0.000 claims 1
- 230000008569 process Effects 0.000 abstract description 33
- 238000012545 processing Methods 0.000 abstract description 14
- 238000011897 real-time detection Methods 0.000 abstract description 2
- 230000000875 corresponding effect Effects 0.000 description 17
- 238000004590 computer program Methods 0.000 description 13
- 239000012634 fragment Substances 0.000 description 12
- 230000000694 effects Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 235000008331 Pinus X rigitaeda Nutrition 0.000 description 2
- 235000011613 Pinus brutia Nutrition 0.000 description 2
- 241000018646 Pinus brutia Species 0.000 description 2
- 238000000429 assembly Methods 0.000 description 2
- 230000000712 assembly Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000013515 script Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000012840 feeding operation Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000000338 in vitro Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000001387 multinomial test Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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]
- H04N21/20—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
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B27/00—Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
- G11B27/02—Editing, e.g. varying the order of information signals recorded on, or reproduced from, record carriers
- G11B27/031—Electronic editing of digitised analogue information signals, e.g. audio or video signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04H—BROADCAST COMMUNICATION
- H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
- H04H60/35—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users
- H04H60/37—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users for identifying segments of broadcast information, e.g. scenes or extracting programme ID
- H04H60/375—Commercial
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04H—BROADCAST COMMUNICATION
- H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
- H04H60/56—Arrangements characterised by components specially adapted for monitoring, identification or recognition covered by groups H04H60/29-H04H60/54
- H04H60/59—Arrangements characterised by components specially adapted for monitoring, identification or recognition covered by groups H04H60/29-H04H60/54 of video
-
- 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
- H04N21/2668—Creating a channel for a dedicated end-user group, e.g. insertion of targeted commercials based on end-user profiles
-
- 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
-
- 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]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/81—Monomedia components thereof
- H04N21/812—Monomedia components thereof involving advertisement 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]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/835—Generation of protective data, e.g. certificates
- H04N21/8358—Generation of protective data, e.g. certificates involving watermark
Abstract
The system and method that there is described herein a kind of commercial message for occurring in one or more broadcast video streams, the automation real-time detection of public service bulletin or similar television program segment long, duplicate in short-term and processing.It is utilized a kind of by identifying the process for having the specific discrete video segment long in short-term other than other attributes to identify possible commercial advertisement segment.Seeming that the video clip with these features is likely to is considered as that television commercial, publicity or public service are announced and be stored in the television advertising database separated with main television content database.Incoming video from multiple tv programme sources is processed into fingerprint and is placed in main television content database.Meanwhile occurrence of the test incoming program in television advertising database.If it find that occurrence, then remove the television advertising speculated from main contents database.
Description
Inventor
Ze Funiu Meier (Zeev Neumeier)
Brian Reed (Brian Reed)
Priority claim
It is submitting and on November 6th, 2013 is by U.S. Patent number 8,595,781 that the application constitutes on May 27th, 2010
Entitled " the METHODS FOR IDENTIFYING VIDEO SEGMENTS AND DISPLAYING CONTEXTUAL of publication
TARGETED CONTENT ON A CONNECTED TELEVISION is (for identifying video clip and having connected TV
The method for showing contextual target content) " U.S. Patent Application No. 12/788,721 part continuation application, that application is
It is required that entitled " the SYSTEM FOR PROCESSING CONTENT INFORMATION IN A submitted on May 29th, 2009
The U.S. Provisional Patent Application No. of TELEVIDEO SIGNAL (system for handling the content information in television video frequency signal) "
The non-provisional application of 61/182,334 equity, and be the entitled " CONTEXTUAL for requiring on December 29th, 2009 to submit
TARGETING BASED ON DATA RECEIVED FROM A TELEVISION SYSTEM (is based on receiving from television system
Data contextual target determine) " U.S. Provisional Patent Application No. 61/290,714 equity non-provisional application；This
Application further constitutes entitled " the METHODS FOR DISPLAYING CONTEXTUALLY submitted on May 27th, 2010
TARGETED CONTENT ON A CONNECTED TELEVISION is (based on context fixed for showing on having connected TV
For the method for the content of target) " U.S. Patent Application No. 12/788,748 part continuation application；The further structure of the application
At entitled " the METHODS FOR INDENTIFYING VIDEO SEGMENTS AND submitted on November 25th, 2013
DISPLAYING CONTEXTUALLY TARGETED CONTENT ON A CONNECTED TELEVISION (is regarded for identification
Frequency segment and the method that contextual target content is shown on having connected TV) " U.S. Patent Application No. 14/089,003
Part continuation application；The application further constitutes entitled " the SYSTEMS AND METHODS FOR submitted on March 17th, 2014
ADDRESSING A MEDIA DATABASE USING DISTANCE ASSOCIATIVE HASHING using distance (for being closed
Connection property hashing is to the system and method for media database addressing) " the part of U.S. Patent Application No. 14/217,039 continue
Application；The application further constitutes entitled " the SYSTEMS AND METHODS FOR submitted on March 17th, 2014
IDENTIFYING VIDEO SEGMENTS FOR DISPLAYING CONTEXTUALLY RELEVANT CONTENT (is used for
Mark video clip is so as to the system and method that show contextual content) " U.S. Patent Application No. 14/217,075
Part continuation application；The application further constitutes entitled " the SYSTEMS AND METHODS FOR submitted on March 17th, 2014
The U.S. Patent Application No. of ON-SCREEN GRAPHICS DETECTION (system of test pattern and side on screen) "
14/217,375 part continuation application；The application further constitutes entitled " the SYSTEMS AND submitted on March 17th, 2014
METHODS FOR IMPROVING SERVER AND CLIENT PERFORMANCE IN FINGERPRINT ACR
The U.S. Patent application of SYSTEMS (system and method for improving server and client side's performance in fingerprint ACR system) "
Number 14/217,425 part continuation application；The application further constitutes the entitled " SYSTEMS submitted on March 17th, 2014
AND METHODS FOR MULTI-BROADCAST DIFFERENTIATION (system distinguished for multi-broadcast and side
Method) " U.S. Patent Application No. 14/217,435 part continuation application；And the application further constitutes March 15 in 2013
Entitled " the SYSTEMS AND METHODS FOR IDENTIFYING VIDEO SEGMENTS BEING that day submits
DISPLAYED ON REMOTELY LOCATED TELEVISIONS is (for identifying the video shown on being remotely located TV
The system and method for segment) " U.S. Patent Application No. 61/791,578 part continuation application；Aforementioned application is either current
The application of the either equity for being entitled to the submission date of current copending application of co-pending.
Invention field
Present invention relates in general to image recognitions, and more particularly relate to using automation content identification data library
Real-time television purposes of commercial detection system and method.
Background
The automatic detection of television commercial is the very interested field of different supporters for a long time.Many consumption
Person wishes that always seeking certain mode keeps entire commercial advertisement mute or even skip, and is especially being recorded for time shift sight
In the program seen.Have been proposed for this needs many solutions and all consumer device grade operation and
It is related to once identifying a commercial advertisement when commercial advertisement occurs in the broadcast video stream watched or recorded.
Other using this automation contents recognizing technology be need to verify specific TV commercial advertisement actually by
Contract provision those of is broadcasted side in each in many channels, time slot and market.In addition, other companies are to which company
Playing that statistics is investigated and collected in which advertisement in which market.Notebook data can for company oneself use or
For reselling.Meanwhile the investigation commercial advertisement of other companies comes for the more effective advertisement of client development.The analysis of these types is usual
It is completed by market grade and needs to monitor a large amount of channels and market.
Since half a century in past, this commercial advertisement verifying or collection procedure are during broadcast time according to plan by the mankind
Manually perform, or visually search (the F.F., rewinding etc.) tape broadcasted in the early time or other records to execute.Can such as it recognize
Know, waits commercial advertisement to play (broadcast), sets up recording equipment to record broadcast, and/or search for the record of broadcasted content
It can be time-consuming, laborious and expensive task to verify commercial content broadcasting.Accordingly, there exist a kind of unsatisfied needs:
Optimize automatic detection and the identification procedure of the business information in live video stream.
Because almost the consumer of each market segment can be used now while transmit the wired of hundreds of program channels
System and satellite system, which commercial advertisement is broadcasted in which market at any given time for mark, verifying and tracking
Ability propose extraordinary calculating challenge.The state of the art is principally dedicated to always mark view single (site) equipment
When (such as television set) is presented commercial advertisement.It is known to those skilled in the art, allow them to while real-time monitoring is hundreds of
A video flowing will calculate power and carrying cost simultaneously, and commercially reasonably extension (scaling) method is still unsatisfied need
It wants.
It summarizes
In certain embodiments, example relevant to the automation real-time television purposes of commercial detection in content identification data library is used
Property method may include: to absorb at least one audiovisual streams, including at least will be associated with one or more frames of the audiovisual streams
At least some of data storage is into a program content data memory；Determine that a candidate segment, the candidate segment include coming
At least one or more successive frame of at least one audiovisual streams absorbed from this；And the candidate segment determined by testing is one
At least one occurrence in a known advertisement data storage, and if test returns to the known advertisement to the candidate segment
At least one occurrence in data storage, then at least will at least some of data associated with the candidate segment from the program
It is removed in content data memory.
In certain embodiments, example relevant to the automation real-time television purposes of commercial detection in content identification data library is used
Property method may further include: test determined by the candidate segment at least one of the program content data memory
Occurrence, and if test returns at least one occurrence in the program content data memory to the candidate segment,
At least at least some of data associated with the candidate segment are removed and will be with this from the program content data memory
The associated at least some of data of candidate segment are stored in the known advertisement data storage.
In certain embodiments, determine that (candidate segment includes the view that at least one is absorbed from this to a candidate segment
Listen at least one or more successive frame of stream) it may include: one received to a frame of the stream of audiovisual data absorbed
Or multiple instructions；The one or more instruction to frame is analyzed, including at least determining that the one or more indicates whether can
It can be a start frame of a candidate segment；If it is determined that a start frame, then analyze received after the start frame
One or more frames, including at least determine whether one of one or more frame received after the start frame may be one
End frame；And if it is determined that an end frame, then establish (i) start frame, (ii) after the start frame and in the knot
The received one or more frame and (iii) end frame are as the candidate segment before beam frame.
In certain embodiments, analyze one or more of frame is indicated (including at least determine this or more
It is a indicate whether may be a candidate segment a start frame) and analyze received one or more after the start frame
A frame (including at least determining whether a frame received after the start frame may be an end frame) may include: determination
Whether one frame is monochrome frame or with one or more one with the and then substantially different fingerprint value of previous frame
At least one of in a frame, and if a frame is a monochrome frame or has one or more with and then previous frame base
In sheet in a frame of different fingerprint values at least one of, then establish the frame as in a start frame or an end frame
At least one of.
In certain embodiments, determine that (candidate segment includes the view that at least one is absorbed from this to a candidate segment
Listen at least one or more successive frame of stream) it may include: one received to a frame of the stream of audiovisual data absorbed
Or multiple instructions；By with the associated data of a nearest coherent intake frame sequence of received frame end and with deposit before
Storage is compared in the coherent associated data of frame sequence of one or more of the program content data memory；And at least
The comparison data are based in part on to provide to the one or more consecutive frames being stored in front of in the program content data memory
One or more instructions of sequence.
In certain embodiments, determine that (candidate segment includes the view that at least one is absorbed from this to a candidate segment
Listen at least one or more successive frame of stream) it may further include in the coherent frame sequence of the indicated one or more
Each frame sequence, determine whether the coherent frame sequence is an advertisement.In certain embodiments, determining the coherent frame sequence is
It is no be an advertisement may include: in number of seconds length and frame count associated with the frame sequence that links up at least one of carry out
Assessment, and if number of seconds length or frame count are substantially similar to a standard advertisement length, establish the coherent frame sequence
As a candidate segment.In certain embodiments, in number of seconds length and frame count associated with the frame sequence that links up
At least one carries out assessment and establishes if number of seconds length or frame count are substantially similar to a standard advertisement length to be somebody's turn to do
Coherent frame sequence may further include as a candidate segment: adjustment at least one threshold value associated with matching is adjusted
The return of whole at least one threshold value with low amount suspects that item is associated；It is based at least partially at least one adjusted
Threshold value matches the coherent frame sequence of the one or more that the candidate segment is previously stored in the data storage；And
And if being based at least partially at least one threshold value adjusted, which is previously stored in the data storage
The one or more consecutive frames sequences match, it is determined that the starting point and end point of the candidate segment.
In certain embodiments, at least one threshold value (at least one threshold value adjusted associated with matching is adjusted
Suspect that item is associated with the return of low amount) may include: adjust it is associated with a path tracing algorithm at least one
Radius, at least one radius adjusted suspect that item is associated with the return of low amount.In certain embodiments, adjustment with
Match at least one associated threshold value (at least one threshold value adjusted suspects that item is associated with the return of low amount)
It may include: adjustment and at least one duration relevant to a path tracing algorithm associated at least one case (bin), institute
At least one radius of adjustment suspects that item is associated with the return of low amount.
In certain embodiments, it determines the starting point of the candidate segment and end point may include: that adjustment is related to matching
At least one threshold value of connection, at least one threshold value adjusted suspect that item is associated with the return of comparatively high amts；It will be with the time
Associated one or more this or more for indicating to be previously stored in the data storage of each frame in selected episode
The associated one or more instructions of a respective frame in a coherent frame sequence are compared, this is relatively based at least partially on
At least one threshold value adjusted；And this is based at least partially on relatively to test each frame in the candidate segment therewith
Before be stored in inconsistency between the respective frame that the one or more in the data storage links up in frame sequence.
In certain embodiments, it is based at least partially on each frame relatively tested in the candidate segment and deposits before
The inconsistency stored up between a respective frame in the coherent frame sequence of the one or more in the data storage can wrap
If including the coherent frame sequence of the one or more that any frame in the candidate segment is previously stored in the data storage
In do not have a corresponding occurrence, then by the frame of none corresponding occurrence in the candidate segment from being stored in this before
It is removed in the coherent frame sequence of the one or more in data storage.In certain embodiments, it is based at least partially on the ratio
Each frame relatively to test in the candidate segment is previously stored in the one or more consecutive frames sequence in the data storage
If the inconsistency between a respective frame in column may include this be stored in the data storage before or
Any frame in multiple coherent frame sequences does not have a corresponding occurrence in the candidate segment, then will be stored in the number before
It is added to the candidate segment according to the frame in the coherent frame sequence of the one or more in memory.In certain embodiments, until
The each frame relatively to test in the candidate segment that is at least partly based on this be previously stored in the data storage this one
The inconsistency between a respective frame in a or multiple coherent frame sequences may include: to be based at least partially on this relatively to come
It tests in the coherent frame sequence of the one or more that each frame in the candidate segment is previously stored in the data storage
A respective frame between inconsistency；And any inconsistency by test discovery is remedied, this, which is remedied, can operate use
In the boundary of mark candidate item.
In certain embodiments, at least one audiovisual streams is absorbed (including at least by one or more frames with the audiovisual streams
Associated at least some of data storage is into a program content data memory) and test the identified candidate segment
At least one occurrence in a known advertisement data storage and if test to the candidate segment to return to this known
At least one occurrence in advertisement data store then at least will at least some of data associated with the candidate segment from this
Removing in program content data memory may include: at least two audiovisual streams of intake, including at least will be at least two view
Listen the associated at least some of data storage of one or more frames of at least one audiovisual streams in stream to a programme content number
According in memory, which includes to referring to at least one of the audiovisual streams of the one or more frame from wherein receiving
Show, and tests identified at least one occurrence of the candidate segment in a known advertisement data storage, and
If test returns at least one occurrence in the known advertisement data storage to the candidate segment, at least will be with the time
The associated at least some of data of selected episode are removed from the program content data memory.
In certain embodiments, determine that (candidate segment includes the view that at least one is absorbed from this to a candidate segment
Listen at least one or more successive frame of stream) may include be based at least partially on a candidate segment at least one duration it is true
The candidate segment is recognized with a Low threshold matching；And be based at least partially on one or more fingerprints of the candidate segment with
The matching of one known advertisement confirms that the candidate segment has a high threshold matching.In certain embodiments, a time is determined
Selected episode (at least one or more successive frame that the candidate segment includes at least one audiovisual streams absorbed from this) can wrap
Including determination can be advertisement, commercial advertisement, public service bulletin, publicity, at least part of advertising time, advertisement pod at least
The segment of at least one in a part or the TV programme that will be presented publicity.In certain embodiments, a candidate is determined
Segment (candidate segment include from this at least one absorb audiovisual streams at least one or more successive frame) may include
Determine a candidate segment, the candidate segment include from this at least one audiovisual streams absorbed at least one or more is continuous
Video frame.
In certain embodiments, example relevant to the automation real-time television purposes of commercial detection in content identification data library is used
Property computer program product may include at least one non-transient computer-readable media, and at least one the non-transient calculating
Machine readable medium may include: one or more instructions for absorbing at least one audiovisual streams, including at least will be with the audiovisual
The associated at least some of data storage of one or more frames of stream is into a program content data memory；For determining one
One or more instructions of a candidate segment, the candidate segment include from this at least one audiovisual streams absorbed at least one
Or multiple successive frames；And for a candidate segment determined by testing in a known advertisement data storage at least
One occurrence and if test returns at least one occurrence in the known advertisement data storage to the candidate segment
Then at least remove at least some of data associated with the candidate segment from the program content data memory one or
Multiple instruction.
In certain embodiments, example relevant to the automation real-time television purposes of commercial detection in content identification data library is used
Property system may include the circuit for being configured for absorbing at least one audiovisual streams, including at least by with the one of the audiovisual streams
Or multiple associated at least some of data storages of frame are into a program content data memory；It is configured for determining one
The circuit of a candidate segment, the candidate segment include the audiovisual streams from least one intake at least one or more is continuous
Frame；And it is configured for testing at least one of an identified candidate segment in a known advertisement data storage
A occurrence and if at least one occurrence for being returned in the known advertisement data storage to the candidate segment of test
The circuit that at least at least some of data associated with the candidate segment are removed from the program content data memory.
In addition to the foregoing, the text (for example, attached drawing and/or detailed description) and/or attached drawing such as present disclosure
Different other methods, system and/or program product embodiment are illustrated and described in teaching.
Foregoing teachings are the simplification, summary and omissions for summarizing and therefore necessarily include details；Therefore, skill in the art
Art personnel will be recognized that general introduction is merely illustrative and is not intended to be limited in any way.Device described herein and/or
Other of process and/or other themes aspect, embodiment, feature and advantage will become obvious in set forth herein teach.
Brief Description Of Drawings
Referring to the following drawings, certain embodiments of the present invention has been described more particularly below.
Fig. 1 illustrates the typical advertising time (or advertisement pod (ad pod)) in time scale, and being highlighted can be with
How commercial message is intercutted.If the duration of commercial advertisement is fallen into certain parameters (104 Low threshold) can with first it is detected that
Commercial advertisement, and if the fingerprint matching of some existing commercial advertisement in segment and database 135 is suspected, with high confidence
Degree is identified.
How the various components that Fig. 2 illustrates system interact, to make the independent database of TV programme and commercial message
It can be established, safeguarded and be utilized in a manner of the reliable real-time identification that can be realized advertisement information.
Fig. 3 illustrates flow chart, and how interrelated the sequence for summarizing each step of this method is with them.
Fig. 4 illustrates prior art flowchart.
Fig. 5 illustrates operating process, which indicates and the real-time television advertisement using automation content identification data library
Detect relevant exemplary operations.
Fig. 6 illustrates the alternate embodiment of the operating process of Fig. 5.
Fig. 7 illustrates the alternate embodiment of the operating process of Fig. 5.
Fig. 8 illustrates the alternate embodiment of the operating process of Fig. 5.
Fig. 9 illustrates the alternate embodiment of the operating process of Fig. 5.
Figure 10 illustrates the alternate embodiment of the operating process of Fig. 5.
Figure 11 illustrates the alternate embodiment of the operating process of Fig. 5.
Figure 12 illustrates illustrative computer program product.
Figure 13 illustrates system relevant to using the real-time television purposes of commercial detection in content identification data library is automated.
It is described in detail
It there is described herein a kind of commercial message for occurring in one or more broadcast video streams, public service bulletin
Or the system and method for the automation real-time detection and processing of similar television program segment long, duplicate in short-term.One kind is utilized
Possible commercial film is identified by identifying with the specific discrete video segment long in short-term other than other attributes
The process of section.Seem that the video clip with these features is likely to be considered as television commercial, publicity or public service
It announces and is stored in the television advertising database separated with main television content database.From multiple tv programme sources
Incoming video is processed into fingerprint and is placed in main television content database.Meanwhile the test incoming program is in electricity
Depending on the occurrence in advertising database.If it find that occurrence, then move the television advertising speculated from main contents database
It removes.If not finding occurrence, video is passed to according to individual television advertising data library test.If in television advertising number
According to occurrence is found in library, then video clip is removed from main television content database.This high efficiency method enables a system to
The television advertising material of known frequent repetition and the high redundancy across many channels is removed.It is this to detect and out of main TV
The process for holding removal television advertising in database reduces the wrong report number of matches of major television content and improves video
The system performance of match system.
Present invention relates in general to video frequency signal processings, and more particularly relate to handle multiple broadcast video signals
It flows to identify, extract and analyze the technology of commercial advertisement or its certain types of video content, the video content of these types
It shares certain features or has and the matched signature of contents known.For realizing the method system by computer server
A series of software process of upper operation form.This server includes microprocessor, database, input equipment and output equipment,
Wherein, the database includes being stored in fixed memory or other digital storage systems and being held by the microprocessor
Capable computer-readable instruction.
One of the problem of video fingerprint data library of maintenance automation content recognition (ACR) system, is wide from live telecast
The content for broadcasting intake includes a high proportion of television commercial.For its essence, commercial advertisement is repeated and the content is used
Many example Fast Fillings of the same content refer to (matching) database.Fig. 1 is illustrated with the advertising time 109 (in electricity
Depending on being referred to as " advertisement pod (ad pod) " in industry) typical programme channel 106, in addition to other materials, when advertisement
Between channel generally include the segment of many different lengths: television advertising, program promotional public service bulletin and will presented.
The duration of typical television advertising 107 is 30 seconds.The duration of typical TV programme publicity is 15 seconds.The overall length of advertisement pod 139
Total duration variation of the degree usually from 60 seconds to three minute.Advertisement pod and other non-program materials occupy average TV programme hour
More than 33%.
This large amount of duplicate contents can cause ACR system to generate a large amount of wrong report matchings or can't find at all on the contrary
Matching.In one case, this is because during same TV programme the repetition of commercial advertisement and many different TV frequencies
The repetition of the specific commercial advertisement in many different TV programme on road.Therefore, when unknown video source includes television commerce
When advertisement fingerprint (clue), the database with high proportion television commercial will generate and the business in entire database
The proportional many matchings of the repetition percentage of advertisement, to generate useless result in duplicate commercial advertisement duration.
There are problems when however, whether duplicate two fingerprints in determining reference database repeating in this way, because they
Ignored minuent is answered to match with content matching or because of their expressions.In order to alleviate this problem, invention describes one
The algorithm of kind detection commercial advertisement and other frequent duplicate video clips (program promotional that will such as present).This algorithm
Length when basic premise is commercial advertisement is repeated many times on given television channel and repeats across multiple television channels.
Once video clip is marked as commercial advertisement, it will be stripped and be stored in individually from primary database
In television commerce database.Other than clearing up main TV program database, this will also be able to detection particular commercial without
Will their independent loads to the useful feature in main advertising database.
Television advertising detection process works as follows: as seen in Fig. 2, TV programme are fed through 202 uploads, processing simultaneously
It is output in main matching system database 204 and is also routed to television commerce video detector 203.The present invention is monitored
Each television channel, with television commercial detector of the invention example.
Television commercial detector is in the input feeding operation usually positioned with central server device of the invention
Hold search process.This with it is of the invention user household to have connected the process that runs in TV similar, wherein the invention with
Fixed every time interval sample number reads pel spacing from one or more screen positions.As domestic applications, the mistake
Journey further to the pel spacing execute digital processing with prepare the spacing be input to will attempt by the picture element matrix with
Contents known database carries out matched content matching system.
Different from the application of the invention run on the TV of enabling, commercial detector content search process continues
Search each single item possibility matching is until reach matching threshold and that matching aggregation process stopping.This is searched for regular content
The difference is that in general, the content search for returning to multiple occurrences be considered as it is invalid, because of the segment of regular length
It is not repeated in the body interior of TV programme usually.This advertisement search process continuous operation and check come from each TV Festival
Mesh source, be fed to the video of system of the invention.
For each of returning to possible occurrence from the above process, following procedure is asynchronously triggered, above-mentioned algorithm is allowed
It continues searching while following procedure executes in three steps:
Step 1: the length of each possible occurrence of assessment, the length should be longer than given threshold value, for example, 5 seconds, and
And it is less than another threshold value, it may is that 60 seconds.In Fig. 1, this is both the duration as shown in 109 and Low threshold matches
101.In fact, according to such as method disclosed in United States Patent (USP) 8,585,781, this is only segment in predetermined threshold
Difference between the first match point around 104 and the timestamp of the last one match point.
Step 2: matching if possible is in the acceptable parameter of previous item test (step 1), then it again passes by interior
Hold search, but test request more refines, wherein execute same search as before, but need the point (fingerprint) of two samples with
More close match in 105.Assess the test as a result, and if with the matched point (fingerprint) of this tightened up search hundred
Divide than being less than given threshold value, then possible matching is cancelled.
Step 3: the test before fitting through if possible, then it is by another content search again, but configures wide
Pine is much, that is, search is as before, but each point of two samples is allowed to be spaced further apart, perhaps in time or distance
On at least one.The result of this test helps the boundary for identifying suspected television commercial and is evaluated not lack
Point (that is, all the points in two samples all match) and there is beginning in the time threshold for the commercial advertisement initially suspected
(that is, if initially possible matching is 15 seconds long, the result of loose content search cannot be 30 for boundary and end boundary
It is second long etc.).
Referring to Fig.1 2, if unknown video clip sample passes through all of above three tests, then it is assumed that it is TV quotient
Industry advertisement, publicity, board service bulletin, and the information for defining the video clip is sent to video clip processing by 206
Device 207.If already existing television commercial, publicity or board service are announced in the sample and advertising database 205
Matching, then new example is removed from main TV program database 204 and is not needed described by video clip processor 207
The advertisement matched is again placed in advertising database.If the sample is matched with main content of TV program, TV Festival
That segment (unknown fragment is matched in period) of purpose is considered as advertisement, and enters the new program and TV number of system
It is all removed according to two examples of matching advertisement already existing in library 204, and on the position at them that new business is wide
Announcement is input in advertising database 205.
It is presented in Fig. 3 and summarizes the flow chart of this method.Each television video stream or " channel " are received simultaneously by system 301
And it is initially tested according to existing known commercial advertising database 303.If there is matching 304, then by it from TV Festival
308 are removed in mesh database.If do not matched, test 307 is carried out to it according to known TV program database 308.If
It is matched, then video clip matches, then it is removed 310 from TV program database 310 and be added to television commercial
Database 311.If it is matched, then it is assumed that it is not associated clip 312 and is ignored.Fig. 4 illustrates prior art process
Figure is for comparing.
Fig. 5 illustrates operating process 500, which indicates wide with the real-time television using automation content identification data library
It accuses and detects relevant exemplary operations.In Fig. 5 and in each exemplary following figure for including operating process, it will be arrived about Fig. 1
The above-mentioned example of Fig. 3 and/or discussion and explaination are provided about other examples and context.However, it should be understood that can be in many
These operating processes are executed in other environment and context and/or in the modified version of Fig. 1 to Fig. 3.Although moreover, with institute
Each operation stream is presented in the sequence of explanation, it should be appreciated that can sequentially be executed with other different from illustrated sequence
Each operation, or may be performed simultaneously these operations.
After a start operation, operating process 500 is moved to operation 502.Operation 502, which depicts, absorbs at least one audiovisual
Stream, including at least depositing at least some of data storage associated with one or more frames of the audiovisual streams to program content data
In reservoir.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, the incoming audiovisual comprising program and ad content
The satellite that data flow can be passed up to by data flow from national network broadcasting channel center is transmitted to ground.It can also be from
Other sources (for example, by transmitting from other sources to ground) receive, are received by optical fiber connection, use conventional VHF, UHF or microwave
Antenna receives, receives data flow by data network (such as internet).Data associated with data flow can store
It can be referred to as in the media data storage of main matching system database.
By its data associated with data flow can store the operation in media data storage may include with
Operation described in lower father's patent application: U.S. Patent Application No. 12/788,721 (being United States Patent (USP) 8,595,781 now)
“METHODS FOR IDENTIFYING VIDEO SEGMENTS AND DISPLAYING CONTEXTUAL TARGETED
CONTENT ON A CONNECTED TELEVISION is (for identifying video clip and showing up and down on having connected TV
The method of literary object content) " (patent of ' 781)；And/or with the related U.S. patent application number 14/ applying submitting simultaneously immediately
217,039“SYSTEMS AND METHODS FOR ADDRESSING A MEDIA DATABASE USING DISTANCE
ASSOCIATIVE HASHING (for using the system and method apart from relevance hashing to media database addressing) " and/
Or " the SYSTEMS AND METHODS FOR IDENTIFYING VIDEO SEGMENTS FOR of related application number 14/217,075
(it is based on context fixed to show for identifying video clip by DISPLAYING CONTEXTUALLY RELEVANT CONTENT
For the system and method for the content of target) " (" related application ").
Next, operation 504, which depicts, determines a candidate segment, which includes from least one intake
At least one of audiovisual streams or multiple successive frames.For example, being taken the photograph as shown in Fig. 1 to Fig. 3 and/or about described in it
After taking stream of audiovisual data, possible occurrence is selected perhaps by the content search process carried out to incoming feeding.It may
Occurrence can indicate just to be already present on media content in similar for it or identical coherent frame fragment
Any coherent frame fragment absorbed in any channel feeding in database.Possible occurrence can indicate relevant advertisement or
Another relevant short-movie section, as public service is announced.It may then pass through three step process and assess possible occurrence, the mistake
Journey may include assessing the length of possible occurrence, requiring the possibility occurrence with existing frame fragment using tightened up matching
Fingerprint spend closely and require to assess using less stringent matching close with the fingerprint of the possibility occurrence of existing frame
Degree, to help to identify the boundary of possible occurrence.Possible occurrence is returned as candidate segment.
Next, operation 506 depict test determined by candidate segment in known advertisement data storage at least
One occurrence, and if test returns to the matching of at least one of the known advertisement data storage to the candidate segment
, then at least at least some of data associated with the candidate segment are removed from program content data memory.For example, such as
Shown in Fig. 1 to Fig. 3 and/or about described in it, if data associated with candidate segment with and be stored in advertisement number
According to the associated Data Matching of known advertisement in library, then data associated with the candidate segment being ingested are out of media
Hold in database and is removed.Then, which can continue to end operation.
Fig. 5 also illustrates the alternate embodiment of example operational flow 500.Fig. 5 illustrates example embodiment, wherein operation stream
Journey 500 may include at least one additional operations 508.
Operation 508 illustrates candidate segment determined by test and matches at least one of program content data memory
, and if test returns at least one occurrence in the program content data memory to the candidate segment, at least
At least some of data associated with the candidate segment are removed and will be with the candidate from the program content data memory
The associated at least some of data of segment are stored in the known advertisement data storage.For example, as shown in Fig. 1 to Fig. 3 and/
Or about described in it, if operation 506 the result is that by not knowing that under a cloud is the time of advertisement before checking database
Selected episode, then according to media content database inspection data associated with the candidate segment.If also in media content data
Discovery suspection is the candidate segment of advertisement in library, then data quilt associated with the segment found in media content database
It removes.Further, data associated with the segment are placed in known advertisement data storage.Then, the operation stream
Journey can continue to end operation.
Fig. 6 illustrates the alternate embodiment of the example operational flow 500 of Fig. 5.Fig. 6 illustrates example embodiment, wherein grasping
Making 502 may include at least one additional operations.Additional operations may include operation 602, operation 604, operation 606 and/or behaviour
Make 608.
Operation 602 is illustrated to receive and be indicated the one or more of the frame in the stream of audiovisual data absorbed.For example, as schemed
1, to shown in Fig. 3 and/or about described in it, selects data associated with the specific frame from incoming data flow.
Further, operation 604 illustrates analysis and indicates the one or more of frame, including at least determines this or more
It is a indicate whether may be candidate segment start frame.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, it can
It will indicate that specific frame may be the specific label of the beginning of commercial advertisement to check that data associated with specific frame whether there is
Name.
Further, operation 606 illustrates if it is determined that start frame, then analysis is received one after the start frame
Or multiple frames, including at least determine whether one of one or more frame received after the start frame may be end frame.
For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, if detecting that signature shows that the frame may be commercial advertisement
First frame, then the process starts to analyze after the start frame claimed received successive frames to check appointing in these successive frames
What frame whether may include specific signature.
Further, operation 608 illustrates if it is determined that end frame, then establish (i) start frame, (ii) after the starting
The received one or more frame and (iii) end frame are as the candidate segment after frame and before the end frame.Example
Such as, as shown in Fig. 1 to Fig. 3 and/or about described in it, when claimed start frame and end frame has been determined, those
Frame and all intermediate frames to link together can be candidate segment.
Further, Fig. 6 illustrates the further alternative embodiment of operation 604.Operation 604 may include that at least one is attached
Add operation 610.
Operation 610 illustrate determine frame whether be a monochrome frame or have it is one or more with it is and then previous
In one frame of the substantially different fingerprint value of frame at least one of, and if a frame be a monochrome frame or have one
Or at least one multiple with an and then frame of the substantially different fingerprint value of previous frame, then the frame is established as one
At least one of in a start frame or an end frame.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, table
It include that frame is monochromatic (for example, black, pure white) that bright frame, which may be the signature of start frame, because commercial advertisement is often with example
As one or more completely black frames start.Alternatively, the start frame of commercial advertisement can be indicated by " scene changes ".Change speech
It, the program before two successive commercial advertisements or commercial advertisement will likely include ones which different scenes.When frame content with it is previous
When the content of frame is dramatically different, scene changes can be indicated, this can indicate conversion or quotient from program to commercial advertisement
Conversion between industry advertisement.Detect between two frames it is dramatically different can be at least partly by the patent of ' 781 and/or related Shen
Please disclosed in one or more operations complete.Notably, scene changes do not determine commercial advertisement individually, and only
It is only that this frame may indicate start frame.Other operations described herein can help to continue to assess by the head with scene changes
Whether the segment of frame identification is actually commercial advertisement, thus, even if scene changes indicate in program from a scene to another
The variation of one scene, rather than the variation from program to commercial advertisement, test later will make using the field as commercial advertisement
The segment that scape variation starts is cancelled.
The same signature described in operation 610 can also be applied to the knot for determining whether frame can be in operation 606
Beam frame.
Fig. 7 illustrates the alternate embodiment of example operational flow 500.Fig. 7 illustrates example embodiment, wherein operating 504
It may include at least one additional operations.Additional operations may include operation 702, operation 704, operation 706 and/or operation 708.
Operation 702 is illustrated to receive and be indicated the one or more of the frame in the stream of audiovisual data absorbed.For example, as schemed
1, to shown in Fig. 3 and/or about described in it, selects data associated with the specific frame from incoming data flow.
Further, operation 704 illustrate by with a coherent intake frame sequence phase of nearest received frame end
Associated data and be previously stored in one or more of the program content data memory link up frame sequence it is associated
Data are compared.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, according to media content data memory
To be ingested and checked with the associated data of the coherent frame fragment of specific frame end to determine that the data are deposited
Whether have in reservoir can with just absorb and with any coherent frame fragment of the fragment match of specific frame end.
Further, operation 706, which illustrates, is based at least partially on the comparison data and provides to being stored in the program before
One or more instructions of the coherent frame sequence of one or more of content data memory.For example, as shown in Fig. 1 to Fig. 3
And/or about described in it, matching operation 704 can return related to one or more at least partly matched coherent segments
The data of connection.With the coherent frame fragment of one or more of media content database of the coherent frame fragment just absorbed
Match or part matching can indicate that the coherent frame fragment just absorbed may be commercial advertisement.Thus, it is also possible in media
Hold at least partly matched one or more coherent frame fragments in database.Importantly, matching operation is not for performance reason
Search media content database lucky matches frame by frame and pixel-by-pixel to obtain.But from media content database return and just
At least partly relevant the suspected matching of the segment of intake, perhaps by disclosed in the patent of ' 781 and/or related application
Operation is matched and/or is returned.
Further, operation 708 is illustrated for each frame sequence in the coherent frame sequence of the indicated one or more
Column determine whether the coherent frame sequence is advertisement.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, one or
Multinomial test can be applied to coherent frame sequence, which is the advertisement suspected or claimed in this stage, using this
It is one or more to test to confirm that the coherent frame sequence is suspecting for advertisement.
Further, Fig. 7 illustrates the further alternative embodiment of operation 708.Operation 708 may include that at least one is attached
Add operation 710.
Operation 710, which illustrates, carries out at least one in number of seconds length and frame count associated with the frame sequence that links up
Assessment, and if number of seconds length or frame count are substantially similar to a standard advertisement length, establish the coherent frame sequence
As a candidate segment.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, check indicate to be claimed it is wide
The length for the coherent frame sequence accused.If length close to common Ad Length (for example, 30 seconds, 15 seconds, 60 seconds or other often
The Ad Length seen), then the coherent frame sequence is still considered suspected advertisement.If length is kept off common
Ad Length (for example, 52 seconds), then the coherent frame sequence is considered being unlikely to be advertisement and/or is no longer regarded as advertisement.
Alternatively, if the quantity of quantity common frame close in advertisement of the frame in coherent frame sequence, the coherent frame sequence is still
So it is considered suspected advertisement.The quantity of common frame can be such as 30 second time 29.997 frames per second, or
Substantially 900 frames of person.The length tested can be precisely the duration or quantity of common frame in advertisement, or can be close to wide
The duration or quantity (for example, in two seconds or three seconds, or in 60-90 frame) of common frame in announcement.For out of media
Hold database return candidate item matching algorithm can return do not have lucky identical coherent length candidate item (for example, they
Can be coherent segment start or at the end of the frame that lacks, or may include coherent segment start or end at it is additional
Frame, may is that due to it is called may include the matching process of such as path tracing and/or PPLEB).Operation 710 can be with table
Show at least part of the step 1 disclosed elsewhere.
Fig. 8 illustrates operating process 800, which indicates to continue the example of the example operational flow 500 of Fig. 5
Operation.Operating process 800 may include operation 802, operation 804 and/or operation 806.
Operation 802 illustrates adjustment at least one threshold value associated with matching, at least one threshold value adjusted and
The return of low amount suspects that item is associated.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, and again will
The segment that links up with the associated data of the advertisement suspected and with the one or more returned from media content database is associated
The matchings of data obtain one or more possible matchings.Use loose matching tolerance from phase with for performance purpose
Matched initial operation is retrieved to big media content database (to be draped over one's shoulders perhaps by via ' 781 patents and/or related application
The operation of dew) in contrast, can by the way that additional matches are completed in the advertisement claimed compared with the search result of relatively small amount,
The additional matches (are perhaps draped over one's shoulders again by ' 781 patents and/or related application using what tightened up matching tolerance was completed
The operation of dew).It can be associated with computationally more dense operation using the matching of tightened up tolerance；However, when being used for
When being compared to lesser amount of segment, additional calculations are born in the acceptable limit.
Further, operation 804, which illustrates, is based at least partially at least one threshold value for being adjusted for the candidate segment
The coherent frame sequence of the one or more being previously stored in the data storage is matched.For example, as in Fig. 1 to Fig. 3
It is shown and/or about described in it, data associated with the advertisement claimed and data associated with possible matching
Matching operation be to be completed with the tightened up tolerance established in operation 802.Operation 802 and 804 can indicate herein
In at least part of step 2 for Anywhere disclosing.It is based at least partially on the tightened up matching of tolerance, will can be returned before
One or more being removed from the occurrence retrieved in the possible occurrence returned.
Further, if operation 806 illustrates at least one threshold value for being based at least partially on and being adjusted, the candidate segment
One or more consecutive frames sequences match being previously stored in the data storage, it is determined that the starting of the candidate segment
Point and end point.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, again according to the advertisement inspection claimed
Any matching retrieved before from media content database.It is wide that the effect of this additional examination can be the claimed business of mark
The boundary of announcement is (that is, since the commercial advertisement claimed or terminating to reduce any extra frame or add back from institute's sound
The beginning of the commercial advertisement of title terminates missing frame, both by with the occurrence retrieved before from media content database
Comparison carry out).Operation 806 can be at least partly related to the step 3 Anywhere disclosed herein.
Fig. 8 further illustrates example embodiment, wherein operation 802 may include at least one additional operations.Additional behaviour
Work may include operation 808 and/or operation 810.
Operation 808 illustrates associated with a path tracing algorithm at least one radius of adjustment, adjusted this extremely
A few radius suspects that item is associated with the return of low amount.For example, as shown in Fig. 1 to Fig. 3 and/or about described by it
, can by adjusting with PPLEB search and/or the associated radius of path tracing algorithm, perhaps by the patent of ' 781 and/or
Operation disclosed in related application is so that tolerance is tightened up.
Further, operation 810 illustrate adjustment to and at least one associated case of a path tracing algorithm it is related
At least one duration, the return of at least one radius adjusted and low amount suspects that item is associated.For example, such as Fig. 1
It, can be by adjusting associated with PPLEB search and/or path tracing algorithm to shown in Fig. 3 and/or about described in it
When major radius, perhaps by operation disclosed in the patent of ' 781 and/or related application so that tolerance is tightened up.The duration can
With for example with token from time-based discount branch mailbox (time discount binning) scheduled time case removal before the time
Amount is related.The duration can for example describe associated with one or more tokens of time-based discount branch mailbox scheduled time case
The time is broadcast live.
Fig. 8 further illustrates example embodiment, wherein operation 806 may include at least one additional operations.Additional behaviour
Work may include operation 812, operation 814 and/or operation 816.
Operation 812 illustrates adjustment at least one threshold value associated with matching, at least one threshold value adjusted and
The return of comparatively high amts suspects that item is associated.For example, being operated as shown in Fig. 1 to Fig. 3 and/or about described in it
802 tolerances for being adjusted tightened up are relaxed pine for additional matches, due to the tightened up tolerance matching by operation 804
Retrieved occurrence is removed, which may include advertisement that will be claimed and retrieve from media content database
Less matching is matched.
Further, operation 814 is illustrated one or more instructions associated with each frame in the candidate segment
A respective frame associated one be previously stored in the coherent frame sequence of the one or more in the data storage
Or multiple instructions are compared, this is relatively based at least partially at least one threshold value adjusted.For example, such as Fig. 1 to Fig. 3
Shown in and/or about described in it, the looser tolerance of use is (for example, described in the patent of ' 781 or related application
Radius, duration, other business etc.) compare claimed advertisement and remaining occurrence again.
Further, operation 816, which illustrates, is based at least partially on this relatively to test each frame in the candidate segment
It is previously stored in inconsistent between a respective frame in the coherent frame sequence of the one or more in the data storage
Property.For example, can be spaced more as shown in Fig. 1 to Fig. 3 and/or about described in it in each point of wherein two samples
Claimed advertisement can be more accurately determined in this test process far (for example, perhaps in time or apart from upper)
Boundary.
Fig. 9 illustrates the further alternate embodiment of operation 816, which may include at least one additional operations.
Operation 816 may include operation 902, operation 904, operation 906 and/or operation 908.
If operation 902 illustrate that any frame in the candidate segment is previously stored in the data storage this one
Do not have a corresponding occurrence in a or multiple coherent frame sequences, then by none corresponding occurrence in the candidate segment
Frame from the one or more being stored in the data storage before link up frame sequence in remove.For example, such as Fig. 1 to Fig. 3
Shown in and/or about described in it, can remove and perhaps exist at the beginning or end for the advertisement claimed or perhaps
It is the centre for the advertisement claimed, early there is no the excessive frame of corresponding occurrence from the segment retrieved in media content database.
If operation 904 illustrates in the coherent frame sequence of the one or more being stored in the data storage before
Any frame does not have a corresponding occurrence, then this will be stored in front of in the data storage in the candidate segment
Or the frame in multiple coherent frame sequences is added to the candidate segment.For example, being retouched as shown in Fig. 1 to Fig. 3 and/or about it
State, from it is being found in the segment that media content database is retrieved, there is no the frame of corresponding occurrence in the advertisement claimed
The segment that may be added to the advertisement for indicating claimed, perhaps at the beginning or end for the advertisement claimed and/or in institute
The centre for the advertisement claimed.
Each frame that operation 906 is based at least partially on this relatively to test in the candidate segment is previously stored in the number
According to the inconsistency between a respective frame in the coherent frame sequence of the one or more in memory.For example, as Fig. 1 is arrived
Retrieved shown in Fig. 3 and/or about the advertisement described in it, claimed and from media content database between it
Any inconsistency is compared by the occurrence that the one or more of step 1 and 2 remains, this may include appointing in this paper
At least part of the step 3 where just disclosed.
Operation 908 illustrates any inconsistency remedied through test discovery, this, which is remedied, can be used to mark candidate
The boundary of item.For example, identifying as shown in Fig. 1 to Fig. 3 and/or about the result help for described in it, operating 906 and being suspected
Television commercial boundary and be evaluated no missing point (that is, all the points in two samples all match) and have
Beginning boundary and end boundary in the time threshold for the commercial advertisement initially suspected is (that is, if initially possible matching is
15 seconds long, then the result of loose content search cannot be 30 seconds long etc.).
Figure 10 illustrates the alternate embodiment of the example operational flow 500 of Fig. 5.Fig. 6 illustrates example embodiment, wherein grasping
Make 502 and operate 506 to may include at least one additional operations.Additional operations may include operation 1002.
Operation 1002 illustrate intake at least two audiovisual streams, including at least by at least two audiovisual streams at least
The associated at least some of data storage of one or more frames of one audiovisual streams, should into a program content data memory
At least some of data include and testing institute to from least one instruction wherein received to the audiovisual streams of the one or more frame
At least one occurrence of the determining candidate segment in a known advertisement data storage, and if the time is arrived in test
Selected episode returns at least one occurrence in the known advertisement data storage, then at least will be associated with the candidate segment
At least some of data are removed from the program content data memory.For example, being retouched as shown in Fig. 1 to Fig. 3 and/or about it
State, intake when, to ground transmit and/or otherwise receive more than one channel and with the data flow phase from channel
Associated data are stored in media content database, parallel or substantially in parallel to ground transimission and storage and each channel
Associated data.Specific commercial advertisement can be received on any or all channel in more than one channel, so as to
Many examples of data associated with the specific commercial advertisement being stored in media content database can be generated.It is draped over one's shoulders herein
The operation of the data associated with the advertisement in determining and media content database of dew can be used for identifying and/or handling and tool
The associated data of body advertisement, played on which channel but regardless of the specific advertisement and/or the specific advertisement at what time
It plays.
Figure 11 illustrates the alternate embodiment of the example operational flow 500 of Fig. 5.Figure 11 illustrates example embodiment, wherein
Operation 504 may include at least one additional operations.Additional operations may include operation 1102, operation 1104, operation 11011
And/or operation 1108.
Operation 1102, which illustrates, to be based at least partially at least one duration of a candidate segment and confirms the candidate segment
It is matched with a Low threshold.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, especially with regard to step 1,
If advertisement (advertisement also referred to as suspected, the commercial advertisement suspected, the television commercial suspected, institute claimed
The television advertising of suspection) at least one in other commercial advertisements shared frame duration or quantity, then the segment can be made
Claimed advertisement is selected as " Low threshold matching ".This desired target duration can be at least five seconds (because
Commercial advertisement less than five seconds is extremely rare or is not present) and/or less than 60 seconds or 120 seconds (because be longer than 60 seconds or
Commercial advertisement in 120 seconds is also extremely rare or is not present).This is not conclusive test；But the duration of frame and corresponding
Quantity be the first instruction, wherein follow-up test be intended to confirm this first instruction (that is, being confirmed with high threshold).
Further, operation 1104 illustrates the one or more fingerprints for being based at least partially on the candidate segment and one
The matching of known advertisement confirms that the candidate segment has a high threshold matching.For example, as shown in Fig. 1 to Fig. 3 and/or closing
Described in it, to data (may is that fingerprint) associated with the one or more frames of advertisement claimed and media content
The associated data of one or more frames (again, may is that fingerprint) of possible matching segment in database are compared.
These fingerprints and/or comparison can be generated by the operation disclosed in the patent of ' 781 and/or related application.Data set is compared
A result can be more strong instruction or even confirm claimed commercial advertisement and the piece from database
Section matching (" high threshold matching ").
Further, operation 1106, which illustrates determination, can be advertisement, commercial advertisement, public service bulletin, publicity, advertisement
At least part of time, at least part of advertisement pod or will present TV programme publicity at least one of piece
Section.For example, as shown in Fig. 1 to Fig. 3 and/or about described in it, associated clip disclosed in application can be by immediately
Referred to as advertisement, commercial advertisement, public service bulletin, publicity, at least part of advertising time, at least part of advertisement pod,
Or it is one or more in the TV programme publicity that will be presented.There may be other for purposes of this application will be by segment
It is expressed as the term of associated clip and all this terms within the scope of application.
Further, operation 1108, which illustrates, determines a candidate segment, which includes that at least one is taken the photograph from this
At least one or more successive video frames of the audiovisual streams taken.For example, as shown in Fig. 1 to Fig. 3 and/or about described by it
, the candidate segment of the successive frame absorbed from data flow may include continuous videos frame fragment.It is contemplated, however, that herein anyly
The system and method just disclosed can be related to other aspects of data flow, for example, audio frame, member associated with the frame of data flow
Data or other this embodiments.
Figure 12 illustrates illustrative computer program product 1200, which may include that at least one is non-
Transient state computer-readable medium.The instruction 1204 of computer program product 1200 is further illustrated in Figure 12.1204 exhibition of instruction
One or more instructions for absorbing at least one audiovisual streams are shown, including at least by one or more frames with the audiovisual streams
Associated at least some of data storage is into a program content data memory；For determining one of a candidate segment
Or multiple instruction, the candidate segment include at least one or more successive frame of at least one audiovisual streams absorbed from this；And
And at least one occurrence in a known advertisement data storage of a candidate segment determined by testing and
It at least will be with the time if at least one occurrence that test returns in the known advertisement data storage to the candidate segment
One or more instructions that the associated at least some of data of selected episode are removed from the program content data memory.For example,
As shown in Fig. 1 to Fig. 3 and Fig. 5 to Figure 11 and/or about described in it, computer program product may include that coding exists
One or more instructions in one or more non-transient computer-readable medias and/or by its storage.The one or more
Instruction causes the one or more processing equipment to execute multi-mode operation when being executed by one or more processing equipments, these operations
It include: to absorb at least one audiovisual streams, including at least by least some of number associated with one or more frames of the audiovisual streams
According to storage into a program content data memory；Determine that a candidate segment, the candidate segment include coming from least one
At least one or more successive frame of the audiovisual streams of intake；And candidate segment determined by testing is in a known advertisement data
At least one occurrence in memory, and if test returns in the known advertisement data storage to the candidate segment
At least one occurrence, then at least will at least some of data associated with the candidate segment from the program content data memory
Middle removal.Aforementioned operation at least partly can be similar to and/or be substantially similar to (but being not limited to) elsewhere
The corresponding operating of disclosure.
Figure 13 illustrates exemplary system 1300.System 1300 may include circuit 1302, circuit 1304 and/or circuit
1306。
Circuit 1302 illustrates the circuit for being configured for absorbing at least one audiovisual streams, including at least will be with the audiovisual
The associated at least some of data storage of one or more frames of stream is into a program content data memory.For example, such as Fig. 1
To shown in Fig. 3 and Fig. 5 to Figure 11 and/or about described in it, circuit 1302 can cause to have at least partly similar
In and/or be substantially similar to the operation of (be not limited to but) with the effect of the corresponding operating disclosed elsewhere.
Next, circuit 1304 illustrates the circuit for being configured for determining a candidate segment, the candidate segment packet
Include at least one or more successive frame of the audiovisual streams from least one intake.For example, such as Fig. 1 to Fig. 3 and Fig. 5 to Figure 11
Shown in and/or is at least partly similar to and/or is substantially similar to about described in it, circuit 1304 can cause to have
The operation of (be not limited to but) and the effect of the corresponding operating disclosed elsewhere.
Next, circuit 1306, which illustrates, is configured for testing an identified candidate segment wide known to one
It accuses at least one occurrence in data storage and if test returns to known advertisement data storage to the candidate segment
At least one occurrence in device then at least will at least some of data associated with the candidate segment from the program content data
The circuit removed in memory.For example, as shown in Fig. 1 to Fig. 3 and Fig. 5 to Figure 11 and/or about described in it, circuit
1306 can cause to have and at least partly be similar to and/or be substantially similar to (be not limited to but) and drape over one's shoulders elsewhere
The operation of the effect of the corresponding operating of dew.
System and method described in this specification, flow chart and structural block diagram can be in the meters including program code
It is realized in calculation machine processing system, which includes the program instruction that can be executed by computer processing system.It can also use
Other implementations.In addition, flow chart described herein and structural block diagram, which describe, supports multiple steps and corresponding function
The specific method and/or corresponding actions of (these steps and corresponding function support disclosed constructional device), and can also use
In the corresponding software configuration of realization and algorithm and its equivalent.
The embodiment of theme disclosed in this specification may be implemented as one or more computer program products,
That is, one or more modules of computer program instructions of the coding on tangible program carrier, to be held by data processing equipment
Row controls its operation.Computer storage media can be machine readable storage device, machine-readable storage substrate, memory and set
Standby or one or more of which combination.
Computer program (also known as program, software, software application, script or code) can use any type of programming language
Speech is write, including compiled language or interpreted languages or statement type language or program language, and it can use any form
Deployment, including other used in a computing environment as free-standing program or as module, component, subroutine or be suitable for
Unit.Computer program not necessarily corresponds to the file in file system.Program can store in other programs of the holding of file
Or (for example, being stored in one or more scripts in marking language document) in the part of data, be stored in and be exclusively used in being discussed
Program single file in or be stored in multiple coordination files (for example, storing one or more modules, subprogram or generation
The file of the multiple portions of code) in.Computer program can be deployed on a computer or be located at a website
Or be distributed in multiple websites and pass through suitable interconnection of telecommunication network multiple computers on be performed.
Process described in this specification or logic flow can be by one of the one or more computer programs of execution
Or multiple programmable processors execute to execute the function of being operated to input data and generate output.These processes or
Logic flow can also be executed by device, and device is also implemented as dedicated logic circuit, for example, FPGA (scene can
Program gate array) or ASIC (specific integrated circuit).
The essential element of computer be processor for executing instruction and it is one or more for storing instruction
With the memory devices of data.In general, computer also by include it is one or more for storing data storage equipment (for example,
Disk, magneto-optic disk or CD) it is either operatively connected into for transmitting data from wherein reception data or to it or both connecing
Receive transmission data again.However, computer does not need have this equipment.The processor for being adapted for carrying out computer program is only illustrated
For and not restrictive include both general purpose microprocessor and special microprocessor and any kind of digital computer
Any one or more processors.It is instructed in general, processor will be received from read-only memory or random access memory or both
And data.
It is interacted to provide with the user of system described herein or manager, theme described in this specification
Embodiment may be implemented to have for showing the display equipment of information (for example, CRT (cathode-ray tube) or LCD to user
(liquid crystal display) monitor) and user can by its to computer provide input keyboard (for example, mouse or tracking
Ball) and the computer of pointing device on.Other kinds of equipment can be used for providing and interact with user.For example, being supplied to
The feedback of user may be any type of sensory feedback, for example, visual feedback, audio feedback or touch feedback；And it can be with
Input from the user, including sound, voice or tactile input are received with any form.
The embodiment of theme described in this specification may be implemented in computing systems, which includes
The aft-end assembly of one or more data servers or including have one or more middleware components (such as application server),
Or include with graphic user interface or Web browser (user or administrator can by its with described in this specification
Certain implementations interaction of theme) front end assemblies (such as client computer) or one or more this aft-end assembly, in
Between any combination of part or front end assemblies.The component of system can be by any digital data communications form or medium (as communicated
Network) interconnection.Computing system may include client and server.Client and server is typically remote from each other and usually logical
Cross communication network interaction.Client and the relationship of server on respective computer by means of running and have a visitor each other
The computer program of family end relationship server generates.
Although this specification includes many specific implementation details, these are not necessarily to be construed as to any invention or can
The limitation of the range for the things that can be required, but be interpreted to may be specific to the feature for the specific embodiment specifically invented
Description.Certain features described in this specification can also be realized by combining form under the background of individual embodiment
In single embodiment.
On the contrary, the different characteristic described under the background of single embodiment can also be by individually or with any suitable son
Combined mode is realized in various embodiments.Moreover, although more than feature can be described as working simultaneously with certain combinations
And be even so initially required to, but from required combined one or more features in some cases can with combine
In vitro, and required combination can be for the variation of sub-portfolio or sub-portfolio.
Similarly, although depicting operation in attached drawing with specific order, this be not construed as requiring this operation with
Shown in specific order or executed with sequential order or the operation of all displayings can be performed, it is desirable to realize
As a result.In some cases, multitask processing and parallel processing may be advantageous.Moreover, the difference in above-described embodiment
The separation of system component is not construed as requiring this separation in all embodiments, and should be understood that described
Program element and system usually can be integrated together in single software product or be encapsulated into multiple software product.
Write present description illustrates optimal mode of the invention and provide multiple examples and describe the present invention
And those skilled in the art can be made to make and using the present invention.This specification write does not limit to the present invention
In the precise terms illustrated.Therefore, although being described in detail the present invention referring to examples set forth above, this field it is common
Technical staff can implement variation, modifications and changes without departing from the scope of the present invention to these examples.
Claims (17)
1. a kind of method for real-time television purposes of commercial detection, comprising:
Data associated with multiple frames of audiovisual streams are stored, wherein data storage of the data storage to programme content
In；
By data associated with the coherent frame sequence of the multiple frame and one or more candidate segment matchings, wherein described one
It is stored in the data storage, and is wherein matched using matching threshold one before a or multiple candidate segments
Or multiple candidate segments；
The matching threshold is adjusted to lower matching threshold, wherein the lower matching threshold and return are than the matching threshold
The suspection matching candidate segment of higher amount is associated；
Candidate segment is determined from the coherent frame sequence based on one or more of candidate segments, wherein the determination includes logical
It crosses data associated with the frame in the coherent frame sequence and data associated with the respective frame in the candidate segment
It is compared, to identify the starting point and end point of the coherent frame sequence, and is wherein come using the lower matching threshold
Compare the data；And
Data associated with the candidate segment are removed from the data storage.
2. the method as described in claim 1 further comprises:
Data associated with the candidate segment are stored in the data storage of known advertisement.
3. the method for claim 1, wherein true from the coherent frame sequence based on one or more of candidate segments
Determining candidate segment further comprises:
Analyze the coherent frame sequence；
Start frame and end frame are identified based on the analysis to the coherent frame sequence；And
Establish the start frame, one after the start frame and in the coherent frame sequence before the end frame
Or multiple frames and the end frame, as the candidate segment.
4. method as claimed in claim 3, wherein identify the start frame and the end frame further comprises:
The frame for determining the coherent frame sequence is monochrome frame or has one or more substantially different with and then previous frame
Fingerprint value frame；And
It is monochrome frame or with one or more frames with the and then substantially different fingerprint value of previous frame described in foundation, makees
For start frame or end frame.
5. the method for claim 1, wherein true from the coherent frame sequence based on one or more of candidate segments
Determining candidate segment further comprises:
Data associated with the coherent frame sequence and one or more of the data storage will be previously stored in
The coherent associated data of frame sequence are compared；And
The relatively data are based in part on to identify the one or more consecutive frames being stored in the data storage before
Sequence.
6. the method as described in claim 1 further comprises:
By in number of seconds length or frame count associated with the coherent frame sequence at least one of assess, to determine
Whether the coherent frame sequence is advertisement；And
The frame count based on the number of seconds length or the coherent frame sequence is substantially similar to standard advertisement length, establishes
The coherent frame sequence is advertisement.
7. the method as described in claim 1 further comprises:
The matching threshold is adjusted to higher matching threshold, wherein the higher matching threshold with it is lower than the matching threshold
The suspection matching candidate segment for returning to quantity is associated；
One or more be previously stored in the candidate segment based on the higher matching threshold in the data storage
A coherent frame sequence is matched；And
The starting point and end point of the candidate segment are determined based on the matching.
8. the matching threshold, which is the method for claim 1, wherein adjusted to lower matching threshold, further comprises:
Adjust at least one radius associated with path tracing algorithm.
9. the matching threshold, which is the method for claim 1, wherein adjusted to lower matching threshold, further comprises:
Duration relevant to case is adjusted, the case and path tracing algorithm are associated.
10. the method as described in claim 1 further comprises:
Each frame to test in the candidate segment is previously stored in one in the data storage based on the comparison
The inconsistency between respective frame in a or multiple coherent frame sequences.
11. method as claimed in claim 10, wherein based on the comparison come test each frame in the candidate segment with
The inconsistency being stored between the respective frame in one or more of coherent frame sequences in the data storage before
Include:
Determine that the frame in the candidate segment is previously stored in one or more of consecutive frames in the data storage
Do not have corresponding occurrence in sequence；And
By the frame without corresponding occurrence in the candidate segment from described one be stored in the data storage before
It is removed in a or multiple coherent frame sequences.
12. method as claimed in claim 10, wherein based on the comparison come test each frame in the candidate segment with
The inconsistency being stored between the respective frame in one or more of coherent frame sequences in the data storage before
Include:
The frame in one or more of coherent frame sequences in the data storage is stored in front of determining in the candidate
Do not have corresponding occurrence in segment；And
The frame being stored in one or more of coherent frame sequences in the data storage before is added to institute
State candidate segment.
13. method as claimed in claim 10, wherein based on the comparison come test each frame in the candidate segment with
The inconsistency being stored between the respective frame in one or more of coherent frame sequences in the data storage before
Include:
Any inconsistency by test discovery is remedied, it is described to remedy the boundary that can be used to identify the candidate segment.
14. being the method for claim 1, wherein based on one or more of candidate segments from the coherent frame sequence
Determine that candidate segment further comprises:
Duration confirmation candidate segment based on the candidate segment is matched with Low threshold；And
The matching of one or more fingerprints and known advertisement based on the candidate segment confirms that the candidate segment has high threshold
Value matching.
15. being the method for claim 1, wherein based on one or more of candidate segments from the coherent frame sequence
Determining the candidate segment is further: advertisement, commercial advertisement, public service bulletin, publicity, at least one of the advertising time
Point, at least part of advertisement pod or the TV programme publicity that will present at least one of.
16. a kind of non-transient computer-readable media, comprising:
One or more executable instructions, one or more of executable instructions make when executed by one or more processors
It obtains one or more of processors and executes following operation:
By data associated with multiple frames of audiovisual streams storage into the data storage of programme content；
By data associated with the coherent frame sequence of the multiple frame and one or more candidate segment matchings, wherein described one
It is stored in the data storage, and is wherein matched using matching threshold one before a or multiple candidate segments
Or multiple candidate segments；
The matching threshold is adjusted to lower matching threshold, wherein the lower matching threshold and return are than the matching threshold
The suspection matching candidate segment of higher amount is associated；
Candidate segment is determined from the coherent frame sequence based on one or more of candidate segments, wherein the determination includes logical
It crosses data associated with the frame in the coherent frame sequence and data associated with the respective frame in the candidate segment
It is compared, to identify the starting point and end point of the coherent frame sequence, and is wherein come using the lower matching threshold
Compare the data；And
Data associated with the candidate segment are removed from the data storage.
17. a kind of system for real-time television purposes of commercial detection, comprising:
One or more processors；
Non-transient computer-readable media including instruction, described instruction make when being executed by one or more of processors
One or more of processors execute following operation:
By data associated with multiple frames of audiovisual streams storage into the data storage of programme content；
By data associated with the coherent frame sequence of the multiple frame and one or more candidate segment matchings, wherein described one
It is stored in the data storage, and is wherein matched using matching threshold one before a or multiple candidate segments
Or multiple candidate segments；
The matching threshold is adjusted to lower matching threshold, wherein the lower matching threshold and return are than the matching threshold
The suspection matching candidate segment of higher amount is associated；
Candidate segment is determined from the coherent frame sequence based on one or more of candidate segments, wherein the determination includes logical
It crosses data associated with the frame in the coherent frame sequence and data associated with the respective frame in the candidate segment
It is compared, to identify the starting point and end point of the coherent frame sequence, and is wherein come using the lower matching threshold
Compare the data；And
Data associated with the candidate segment are removed from the data storage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811471681.4A CN109905726B (en) | 2013-03-15 | 2014-03-17 | System and method for real-time television advertisement detection |
Applications Claiming Priority (19)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201361791578P | 2013-03-15 | 2013-03-15 | |
| US61/791,578 | 2013-03-15 | ||
| US14/089,003 | 2013-11-25 | ||
| US14/089,003 US8898714B2 (en) | 2009-05-29 | 2013-11-25 | Methods for identifying video segments and displaying contextually targeted content on a connected television |
| US14/217,435 | 2014-03-17 | ||
| US14/217,039 | 2014-03-17 | ||
| PCT/US2014/030782 WO2014145929A1 (en) | 2013-03-15 | 2014-03-17 | Systems and methods for addressing a media database using distance associative hashing |
| US14/217,425 | 2014-03-17 | ||
| US14/217,039 US9055335B2 (en) | 2009-05-29 | 2014-03-17 | Systems and methods for addressing a media database using distance associative hashing |
| PCT/US2014/030795 WO2014145938A1 (en) | 2013-03-15 | 2014-03-17 | Systems and methods for real-time television ad detection using an automated content recognition database |
| US14/217,375 | 2014-03-17 | ||
| USPCT/US2014/30782 | 2014-03-17 | ||
| US14/217,075 US9055309B2 (en) | 2009-05-29 | 2014-03-17 | Systems and methods for identifying video segments for displaying contextually relevant content |
| PCT/US2014/030805 WO2014145947A1 (en) | 2013-03-15 | 2014-03-17 | Systems and methods for identifying video segments for displaying contextually relevant content |
| USPCT/US2014/30805 | 2014-03-17 | ||
| US14/217,435 US9094715B2 (en) | 2009-05-29 | 2014-03-17 | Systems and methods for multi-broadcast differentiation |
| US14/217,425 US9071868B2 (en) | 2009-05-29 | 2014-03-17 | Systems and methods for improving server and client performance in fingerprint ACR systems |
| US14/217,375 US9094714B2 (en) | 2009-05-29 | 2014-03-17 | Systems and methods for on-screen graphics detection |
| US14/217,075 | 2014-03-17 |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811471681.4A Division CN109905726B (en) | 2013-03-15 | 2014-03-17 | System and method for real-time television advertisement detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105052161A CN105052161A (en) | 2015-11-11 |
| CN105052161B true CN105052161B (en) | 2018-12-28 |
Family
ID=54261403
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201480015936.XA Active CN105052161B (en) | 2013-03-15 | 2014-03-17 | The system and method for real-time television purposes of commercial detection |
Country Status (7)
| Country | Link |
|---|---|
| CN (1) | CN105052161B (en) |
| BR (1) | BR112015023380B1 (en) |
| CA (1) | CA2906192C (en) |
| CL (1) | CL2015002619A1 (en) |
| HK (1) | HK1218193A1 (en) |
| MX (2) | MX366327B (en) |
| WO (1) | WO2014145938A1 (en) |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9961388B2 (en) | 2008-11-26 | 2018-05-01 | David Harrison | Exposure of public internet protocol addresses in an advertising exchange server to improve relevancy of advertisements |
| US10419541B2 (en) | 2008-11-26 | 2019-09-17 | Free Stream Media Corp. | Remotely control devices over a network without authentication or registration |
| US8180891B1 (en) | 2008-11-26 | 2012-05-15 | Free Stream Media Corp. | Discovery, access control, and communication with networked services from within a security sandbox |
| US9154942B2 (en) | 2008-11-26 | 2015-10-06 | Free Stream Media Corp. | Zero configuration communication between a browser and a networked media device |
| US10880340B2 (en) | 2008-11-26 | 2020-12-29 | Free Stream Media Corp. | Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device |
| US9519772B2 (en) | 2008-11-26 | 2016-12-13 | Free Stream Media Corp. | Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device |
| US9026668B2 (en) | 2012-05-26 | 2015-05-05 | Free Stream Media Corp. | Real-time and retargeted advertising on multiple screens of a user watching television |
| US10977693B2 (en) | 2008-11-26 | 2021-04-13 | Free Stream Media Corp. | Association of content identifier of audio-visual data with additional data through capture infrastructure |
| US10334324B2 (en) | 2008-11-26 | 2019-06-25 | Free Stream Media Corp. | Relevant advertisement generation based on a user operating a client device communicatively coupled with a networked media device |
| US10631068B2 (en) | 2008-11-26 | 2020-04-21 | Free Stream Media Corp. | Content exposure attribution based on renderings of related content across multiple devices |
| US9986279B2 (en) | 2008-11-26 | 2018-05-29 | Free Stream Media Corp. | Discovery, access control, and communication with networked services |
| US9386356B2 (en) | 2008-11-26 | 2016-07-05 | Free Stream Media Corp. | Targeting with television audience data across multiple screens |
| US10567823B2 (en) | 2008-11-26 | 2020-02-18 | Free Stream Media Corp. | Relevant advertisement generation based on a user operating a client device communicatively coupled with a networked media device |
| US9653094B2 (en) * | 2015-04-24 | 2017-05-16 | Cyber Resonance Corporation | Methods and systems for performing signal analysis to identify content types |
| CN105975924A (en) * | 2016-04-29 | 2016-09-28 | 杭州当虹科技有限公司 | Regression testing method for precisely recognizing ad content based on video frame statistics |
| CN107633848A (en) * | 2016-07-19 | 2018-01-26 | 阿基米德（上海）传媒有限公司 | It is a kind of to remove the method for advertisement by identifying audio-frequency fingerprint |
| CN106507141A (en) * | 2016-10-19 | 2017-03-15 | 天脉聚源（北京）科技有限公司 | A kind of channel broadcast content monitoring method and device |
| CN110537371B (en) * | 2017-06-30 | 2022-11-22 | 六科股份有限公司 | Frame certainty for automatic content recognition |
| CN109522531B (en) * | 2017-09-18 | 2023-04-07 | 腾讯科技（北京）有限公司 | Document generation method and device, storage medium and electronic device |
| CN107613326A (en) * | 2017-09-30 | 2018-01-19 | 深圳市鑫汇达机械设计有限公司 | One kind identifies accurate television advertising identifying system |
| CN108959187B (en) * | 2018-04-09 | 2023-09-05 | 中国平安人寿保险股份有限公司 | Variable box separation method and device, terminal equipment and storage medium |
| CN111077326A (en) * | 2019-12-20 | 2020-04-28 | 中译语通文娱科技(青岛)有限公司 | Detection technology based on AI technology pre-perception capability |
| CN112948636B (en) * | 2021-03-24 | 2022-09-27 | 黑龙江省能嘉教育科技有限公司 | Regional education cloud resource sharing system and method |
| CN114979691B (en) * | 2022-05-23 | 2023-07-28 | 上海影谱科技有限公司 | Statistical analysis method and system for advertisement of retransmission rights of sports event |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007114796A1 (en) * | 2006-04-05 | 2007-10-11 | Agency For Science, Technology And Research | Apparatus and method for analysing a video broadcast |
| WO2009150425A3 (en) * | 2008-06-10 | 2010-09-10 | Half Minute Media Ltd | Automatic detection of repeating video sequences |
| CN102523482A (en) * | 2011-12-07 | 2012-06-27 | 中山大学 | Advertisement monitoring technology based on video content and regression method |
| CN102760169A (en) * | 2012-06-13 | 2012-10-31 | 天脉聚源(北京)传媒科技有限公司 | Method for detecting advertising slots in television direct transmission streams |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6993245B1 (en) * | 1999-11-18 | 2006-01-31 | Vulcan Patents Llc | Iterative, maximally probable, batch-mode commercial detection for audiovisual content |
| KR100707189B1 (en) * | 2005-04-29 | 2007-04-13 | 삼성전자주식회사 | Apparatus and method for detecting advertisment of moving-picture, and compter-readable storage storing compter program controlling the apparatus |
| JP4321518B2 (en) * | 2005-12-27 | 2009-08-26 | 三菱電機株式会社 | Music section detection method and apparatus, and data recording method and apparatus |
-
2014
- 2014-03-17 CN CN201480015936.XA patent/CN105052161B/en active Active
- 2014-03-17 BR BR112015023380-5A patent/BR112015023380B1/en active IP Right Grant
- 2014-03-17 MX MX2015012511A patent/MX366327B/en active IP Right Grant
- 2014-03-17 CA CA2906192A patent/CA2906192C/en active Active
- 2014-03-17 WO PCT/US2014/030795 patent/WO2014145938A1/en active Application Filing
-
2015
- 2015-09-11 CL CL2015002619A patent/CL2015002619A1/en unknown
- 2015-09-11 MX MX2019008020A patent/MX2019008020A/en unknown
-
2016
- 2016-05-05 HK HK16105168.7A patent/HK1218193A1/en unknown
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007114796A1 (en) * | 2006-04-05 | 2007-10-11 | Agency For Science, Technology And Research | Apparatus and method for analysing a video broadcast |
| WO2009150425A3 (en) * | 2008-06-10 | 2010-09-10 | Half Minute Media Ltd | Automatic detection of repeating video sequences |
| CN102523482A (en) * | 2011-12-07 | 2012-06-27 | 中山大学 | Advertisement monitoring technology based on video content and regression method |
| CN102760169A (en) * | 2012-06-13 | 2012-10-31 | 天脉聚源(北京)传媒科技有限公司 | Method for detecting advertising slots in television direct transmission streams |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105052161A (en) | 2015-11-11 |
| CL2015002619A1 (en) | 2016-04-15 |
| CA2906192A1 (en) | 2014-09-18 |
| MX2015012511A (en) | 2016-01-12 |
| BR112015023380B1 (en) | 2023-03-28 |
| BR112015023380A8 (en) | 2018-08-14 |
| CA2906192C (en) | 2020-10-27 |
| BR112015023380A2 (en) | 2017-07-18 |
| MX2019008020A (en) | 2019-09-04 |
| WO2014145938A1 (en) | 2014-09-18 |
| HK1218193A1 (en) | 2017-02-03 |
| MX366327B (en) | 2019-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105052161B (en) | The system and method for real-time television purposes of commercial detection | |
| CN109905726A (en) | The system and method for real-time television purposes of commercial detection | |
| US8930980B2 (en) | Systems and methods for real-time television ad detection using an automated content recognition database | |
| US20220269708A1 (en) | Systems and methods for partitioning search indexes for improved efficiency in identifying media segments | |
| US9094715B2 (en) | Systems and methods for multi-broadcast differentiation | |
| US9325660B2 (en) | System for social media tag extraction | |
| US10425684B2 (en) | System and method to create a media content summary based on viewer annotations | |
| CN113923518B (en) | Tracking pixels and COOKIE for television event viewing | |
| CA2879522C (en) | Systems, methods and computer-readable media for determining outcomes for program promotions | |
| CN104813357A (en) | Systems and methods for live media content matching | |
| US9794638B2 (en) | Caption replacement service system and method for interactive service in video on demand | |
| US11093978B2 (en) | Creating derivative advertisements | |
| CN110024413A (en) | System and method for splicing advertisement in flow content | |
| US20230283821A1 (en) | Data translation for video-viewing activity | |
| CN109120954B (en) | Video message pushing method and device, computer equipment and storage medium | |
| US9794599B2 (en) | Method and system for auditing multimedia content | |
| US10141023B2 (en) | Method and system for multimedia summary generation | |
| CN113886273A (en) | Performance test method, device, test equipment and medium of application program | |
| US8352626B1 (en) | Program selection from within a plurality of active videos | |
| Mukherjee et al. | An architecture for real time television audience measurement |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20160901 Address after: Delaware Applicant after: Granville Construction Technology Co., Ltd. Address before: American California Applicant before: NEUMEIER ZEEV |
|
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1218193 Country of ref document: HK |
|
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Delaware Applicant after: Structural data Co Ltd Address before: Delaware Applicant before: Granville Construction Technology Co., Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |