CN102346768A - Method and device for finding video advertisement - Google Patents
Method and device for finding video advertisement Download PDFInfo
- Publication number
- CN102346768A CN102346768A CN2011102793610A CN201110279361A CN102346768A CN 102346768 A CN102346768 A CN 102346768A CN 2011102793610 A CN2011102793610 A CN 2011102793610A CN 201110279361 A CN201110279361 A CN 201110279361A CN 102346768 A CN102346768 A CN 102346768A
- Authority
- CN
- China
- Prior art keywords
- video
- fragment
- advertisement
- television
- sample
- 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.)
- Granted
Links
Images
Abstract
The invention provides a method and device for finding a video advertisement. The method for finding the video advertisement comprises the following steps of: A. determining an advertisement time period in a television video by using each pre-established advertisement video sample in an advertisement sample base; and B. comparing the television video in the advertisement time period with each advertisement video sample, removing repeated video clips in television video clips unmatched with each advertisement video sample and determining a target advertisement video from the unmatched television video clips of which the repeated video clip are removed. According to the mode, the monitoring and managing efficiency of the advertisement can be improved.
Description
[technical field]
The present invention relates to data processing technique, particularly a kind of method and device of finding advertisement video.
[background technology]
In television video, advertisement being supervised is a very important job.In the prior art; The method of from television video, finding advertisement video is fairly simple; Usually adopt television video to carry out the multiplicity comparison, find out the higher video segment of frequency of occurrence in the television video, confirm as doubtful advertisement whole day; And then, confirm as the advertisement sample through manual examination and verification.Under this method, have two problems: at first, do the very difficult whole day of finding like this and only play advertisement video once, the advertisement video of promptly finding is comprehensive inadequately; Secondly; Other repeated that whole day is play also can be confirmed as doubtful advertisement; The for example head of the song repeatedly play of whole day, TV play and run-out or the like, these videos that do not belong to advertisement also need increase the workload of audit through manual examination and verification greatly.
[summary of the invention]
Technical matters to be solved by this invention provides a kind of method and device of finding advertisement video, and when to solve prior art advertisement being supervised, the advertisement video of discovery is not comprehensive, the defective of increase manual examination and verification workload.
The technical scheme that the present invention adopts for the technical solution problem provides the said method of a kind of discovery advertisement and comprises: A. utilizes each the advertisement video sample in the advertisement video sample storehouse of setting up in advance to confirm the ad break in the television video; B. the television video in the said ad break and each advertisement video sample are compared; And with the unmatched television video fragment of each advertisement video sample in remove the repeated fragment, confirm the targeted advertisements video the unmatched television video fragment after removing the repeated fragment.
The preferred embodiment one of according to the present invention, said method further comprises behind step B: the targeted advertisements video of confirming is put into said advertisement video sample storehouse.
The preferred embodiment one of according to the present invention, said steps A comprises: in television video, gather the television video sample; Said television video sample and each advertisement video sample are compared one by one, the video segment of comparing with each advertisement video sample in the said television video sample is confirmed as each match video fragment; Each match video fragment is arranged according to the time sequencing in said television video sample; Therefrom confirm continuous videos fragment collection and independent vide fragment; Wherein said continuous videos fragment collection is the match video set of segments that the time interval between adjacent two match video fragments is no more than predetermined value, and said independent vide fragment is the match video fragment except that said continuous videos fragment collection in all match video fragments; Said continuous videos fragment collection is set the second redundant time interval between constantly of setting duration after first constantly redundant and finish time of duration before the zero hour in said television video sample, and said independent vide fragment in said television video sample, set duration before the zero hour triple redundance constantly and the 4th redundant time interval between constantly of setting duration after the finish time confirm as said ad break.
The preferred embodiment one of according to the present invention, said steps A comprises: in television video, gather a plurality of television video samples; Said a plurality of television video samples and each advertisement video sample are compared one by one, confirmed that the video segment of in the identical time period of said a plurality of television video samples, all comparing with same advertisement video sample is the match video fragment; Said match video fragment is set the 6th redundant time interval between constantly of setting duration after the 5th constantly redundant and finish time of duration before the zero hour and confirm as said ad break in said television video sample.
The preferred embodiment one of according to the present invention also comprises after said step B: the advertisement that will comprise as said television video with the television video fragment and the said targeted advertisements video of each advertisement video sample matches.
The preferred embodiment one of according to the present invention; Confirm the unmatched television video fragment after removing the repeated fragment that the targeted advertisements video comprises: utilize manual examination and verification result, confirm the targeted advertisements video to the unmatched television video fragment after the said removal repeated fragment; Perhaps, the sorter that obtains through mass advertising video data training to the judgement of classifying of the unmatched television video fragment after the said removal repeated fragment, is confirmed the targeted advertisements video.
The present invention also provides a kind of device of finding advertisement, and comprising: ad break is confirmed the unit, is used for utilizing each advertisement video sample in the advertisement video sample storehouse of setting up in advance to confirm the ad break in the television video; Go to heavy unit, be used for the television video in the said ad break and each advertisement video sample are compared, and with the unmatched television video fragment of each advertisement video sample in removal repeated fragment; Targeted advertisements is confirmed the unit, is used for confirming the targeted advertisements video from the unmatched television video fragment of removing after the repeated fragment.
The preferred embodiment one of according to the present invention, said device further comprises the insertion unit, is used for the targeted advertisements video of confirming is put into said advertisement video sample storehouse.
The preferred embodiment one of according to the present invention, said ad break confirms that the unit comprises: first gathers subelement, is used for gathering the television video sample at television video; The first comparer unit is used for said television video sample and each advertisement video sample are compared one by one, and the video segment of comparing with each advertisement video sample in the said television video sample is confirmed as each match video fragment; Set of matches is confirmed subelement; With each match video fragment according to time sequencing arrange; Therefrom confirm continuous videos fragment collection and independent vide fragment; Wherein said continuous videos fragment collection is the match video set of segments that the time interval between adjacent two match video fragments is no more than predetermined value, and said independent vide fragment is the match video fragment except that said continuous videos fragment collection in all match video fragments; First ad break is confirmed subelement; Be used for said continuous videos fragment collection is set the second redundant time interval between constantly of duration said television video sample was set first constantly redundant and finish time of duration before the zero hour after, and said independent vide fragment in said television video sample, set duration before the zero hour triple redundance constantly and the time interval of setting after the finish time between the 4th redundant moment of duration confirm as said ad break.
The preferred embodiment one of according to the present invention, said ad break confirms that the unit comprises: second gathers subelement, is used for gathering a plurality of television video samples at television video; The second comparer unit is used for said a plurality of television video samples and each advertisement video sample are compared one by one, confirms that the video segment of in the identical time period of said a plurality of television video samples, all comparing with same advertisement video sample is the match video fragment; Second ad break is confirmed subelement, is used for the time interval that said match video fragment was set said television video sample was set the 5th constantly redundant and finish time of duration before the zero hour after between the 6th redundant moment of duration is confirmed as said ad break.
The preferred embodiment one of according to the present invention, said device further comprises the advertisement extraction unit, is used for the advertisement that television video fragment and said targeted advertisements video with each advertisement video sample matches are comprised as said television video.
The manual examination and verification result that the preferred embodiment one of according to the present invention, said targeted advertisements are confirmed the unmatched television video fragment after unit by using is to said removal repeated fragment confirms the targeted advertisements video; Perhaps, said targeted advertisements is confirmed the sorter that the unit obtains through mass advertising video data training, to the judgement of classifying of the unmatched television video fragment after the said removal repeated fragment, confirms the targeted advertisements video.
Can find out by above technical scheme; Confirm the ad break in the television video through utilizing the advertisement video sample; Through television video in the ad break and advertisement video sample are compared, can improve the advertisement video quantity of from television video, finding well again, the advertisement video of only broadcasting once in the television video is extracted; Simultaneously can filter the non-advertisement video of repeat play in the television video, improve the efficient of advertisement supervision.
[description of drawings]
Fig. 1 is the schematic flow sheet of the embodiment of the method for discovery advertisement video among the present invention;
Fig. 2 is the schematic flow sheet of the embodiment one of the method for definite ad break;
Fig. 3 is the schematic flow sheet of the method for video comparison;
Fig. 4 is the synoptic diagram of ad break;
Fig. 5 is the schematic flow sheet of embodiment two of confirming the method for ad break
Fig. 6 is the structural representation block diagram of the embodiment one of the device of discovery advertisement video among the present invention;
Fig. 7 is the structural representation block diagram of the embodiment two of the device of discovery advertisement video among the present invention.
[embodiment]
In order to make the object of the invention, technical scheme and advantage clearer, describe the present invention below in conjunction with accompanying drawing and specific embodiment.
Please refer to Fig. 1, Fig. 1 is for finding the schematic flow sheet of the method embodiment of advertisement video among the present invention.As shown in Figure 1, said method comprises:
Step S101: utilize each the advertisement video sample in the advertisement video sample storehouse of setting up in advance to confirm the ad break in the television video.
Step S102: the television video in the ad break and each advertisement video sample are compared; And with the unmatched television video fragment of each advertisement video sample in remove the repeated fragment, confirm the targeted advertisements video the unmatched television video fragment after removing the repeated fragment.
Specify in the face of above-mentioned steps down.
The advertisement video sample is meant the video of single advertisement, and advertisement video sample storehouse comprises the database of a plurality of advertisement videos exactly.Advertisement video sample in the advertisement video sample storehouse at first can obtain through collecting existing advertisement video, and secondly, the targeted advertisements video of from method of the present invention, finding finally also can join in the advertisement video sample storehouse as the advertisement video sample.
Utilize each the advertisement video sample in the advertisement video sample storehouse to confirm the ad break of television video in broadcast, can adopt following dual mode to carry out.
Embodiment one:
Please refer to Fig. 2, Fig. 2 is the schematic flow sheet of the embodiment one of the method for definite ad break.As shown in Figure 2, confirm that the method for ad break comprises:
Step S1011: in television video, gather the television video sample.In the present embodiment, can gather one day television video as the television video sample.
Step S1012: television video sample and each advertisement video sample are compared one by one, from the television video sample, confirmed each video segment with the advertisement video sample matches.
The method that the video comparison is adopted can be with reference to figure 3, and Fig. 3 is the schematic flow sheet of the method for video comparison.The television video sample is the source video among the figure, and the advertisement video sample is the target video among the figure.After source video and target video decoded, just can both be compared according to the method shown in Fig. 3.
As can beappreciated from fig. 3, when source video and target video are compared, be actually first frame of getting target video earlier; With each frame comparison of source video,, then get each frame comparison of the next frame and the source video of target video if on not comparing; In finding the source video with target video in the frame that can compare, at this moment, note the position of this frame in the video of source; The recycle execution is got the next frame of getting in next frame and the target video and is compared from the video of source; Up to the comparison failure, note the successful position of frame in the video of source of last comparison, at this moment; Video segment between the position of the frame of two records of source video is the match video fragment.In seeking source video and target video, during first same number of frames, also can get first frame of source video earlier,,, get the next frame of source video and each frame of target video again and compare, by that analogy if not in the comparison with each frame comparison of target video.If there is not a frame can compare successfully in source video and the target video, the comparison failure of source video and this target video then is described, the match video fragment of this target video not in the video of source.
Each advertisement video sample is carried out above-mentioned comparison process as target video respectively, just can find each match video fragment in the television video sample.
Please continue with reference to figure 2.
Step S1013: each match video fragment is arranged according to the time sequencing in the television video sample, therefrom confirm continuous videos fragment collection and independent vide fragment.
As a rule; In television video advertisement playing be have necessarily successional; These advertisements all are in a period of time, to concentrate to play, and interlude is shorter between each advertisement, utilizes this characteristics; Can find out continuous videos fragment collection, this continuous videos fragment collection is meant that the reproduction time between adjacent two match video fragments is no more than the match video set of segments of predetermined value at interval.For example, five match video fragments are arranged, the reproduction time in these five match video fragments between the adjacent in twos match video fragment is spaced apart 5 seconds, then can be with these five match video fragments as a continuous videos fragment collection.Match video fragment in all match video fragments except that continuous videos fragment collection just is referred to as the independent vide fragment.
Step S1014: time interval and independent vide fragment the time interval in television video sample of continuous videos fragment collection in the television video sample confirmed as ad break.
The time interval of continuous videos fragment collection in the television video sample is the time interval between the first redundant moment to the second redundant moment; Wherein the first redundant continuous videos fragment that constantly is arranged in concentrates first match video fragment before the zero hour of television video sample; And being carved into the continuous videos fragment during this first redundancy concentrates the time span of first match video fragment between the zero hour in the television video sample to be no more than the first threshold of setting; Second redundancy is arranged in the continuous videos fragment constantly and concentrates last match video fragment after the finish time of television video sample, and the continuous videos fragment concentrates last match video fragment to be no more than second threshold value of setting the finish time in the television video sample to the time span between this second redundant moment.In like manner; The time interval of independent vide fragment in the television video sample is that triple redundance is constantly to the time interval between the 4th redundant moment; Wherein triple redundance was arranged in the independent vide fragment constantly before the zero hour of television video sample; And be carved into the 3rd threshold value that the time span of independent vide fragment between the zero hour in the television video sample is no more than setting during this triple redundance; The 4th redundant moment was arranged in the independent vide fragment after the finish time of television video sample, and the independent vide fragment is no more than the 4th threshold value of setting the finish time in the television video sample to the time span between the 4th redundant moment.
Please refer to Fig. 4, Fig. 4 is the synoptic diagram of ad break.On scheming, can find out; In the television video sample; Comprised continuous videos fragment collection and independent vide fragment, when confirming ad break, ad break is a period of time interval and a period of time interval that comprises the independent vide fragment that comprises continuous videos fragment collection.One section adjustable time is arranged in the ad break, embody by the first threshold among the figure, second threshold value, the 3rd threshold value and the 4th threshold value.It is identical that first threshold, second threshold value, the 3rd threshold value and the 4th threshold value can be set to, and also can be set to difference.For example the continuous videos fragment collection time of living in is the 10th minute to the 20th minute, and ad break can be the 7th minute to the 23rd minute, and wherein first threshold is identical with second threshold value, is three minutes.
Utilize each the advertisement video sample in the advertisement video sample storehouse to confirm the ad break of television video in broadcast, can also adopt following mode to implement.Please refer to Fig. 5, Fig. 5 is the schematic flow sheet of embodiment two of confirming the method for ad break.
Embodiment two:
Step S101A: in television video, gather a plurality of television video samples.In the present embodiment, can gather one day television video as one of them television video sample.
Step S101B: a plurality of television video samples and each advertisement video sample are compared one by one, confirmed that the video segment of in the identical time period of a plurality of television video samples, all comparing with same advertisement video sample is the match video fragment.
Step S101C: the time interval of match video fragment in the television video sample confirmed as ad break.
The time interval of match video fragment in the television video sample is the time interval between the 5th redundant moment to the 6th redundant moment; The said the 5th redundant moment was arranged in the match video fragment before the zero hour of television video sample; And be carved into the 5th threshold value that the time span of match video fragment between the zero hour in the television video sample is no more than setting during the 5th redundancy; The 6th redundant moment was arranged in the match video fragment after the finish time of television video sample, and the match video fragment is no more than the 6th threshold value of setting the finish time in the television video sample to the time span between the 6th redundant moment.The 5th threshold value here and the 6th threshold value can be identical or different with first threshold, second threshold value, the 3rd threshold value or the 4th threshold value among the embodiment one, no longer do specifying here.
In the embodiment two of confirming ad break, confirm the match video fragment through a plurality of samples of television video, thereby find ad break, meet the rule that advertisement occurs in the television video.Be appreciated that the advertisement that in television video, occurs usually has period repeatability, promptly in television video; What occur probably playing in the section at one time in continuous many days all is identical advertisement; Therefore, adopt the method for embodiment two, also can extract ad break well.
Please continue with reference to figure 1.After having confirmed ad break, just can utilize ad break to confirm the targeted advertisements video in the television video.
In step S102; At first the television video in the ad break and each advertisement video sample are compared; At this moment, the television video in the ad break just comprised with the video segment of advertisement video sample matches and with the unmatched video segment of advertisement video sample, with the video segment of advertisement video sample matches be exactly the advertisement video in the advertisement video sample storehouse; Then need further handle with the unmatched video segment of advertisement video sample, therefrom remove the video segment of repetition.
The video segment of removing repetition is exactly that a plurality of identical video segments are carried out normalization; Keep one of them; Its implementation method is: will compare in twos with the unmatched video segment of advertisement video sample; If in the comparison then keep one of them, this video segment that keeps and the next not video segment of comparison are compared.
Also possibly determine the targeted advertisements video the video segment that do not match that after going heavily to handle, obtains, that is to say that some advertisement video possibly not included in the advertisement video sample storehouse as yet.Concrete adoptable method comprises manual examination and verification, and the auditor never selects advertisement video in the match video fragment, determines the targeted advertisements video according to auditor's selection.Owing to, have only this part video segment that do not match need carry out manual examination and verification through above-mentioned treatment scheme, greatly reduce the suspicious video that needs audit, therefore greatly reduced manpower.In addition, also can adopt machine learning algorithm, whether the sorter that obtains through mass advertising video data training to the video segment that do not match that obtains after going heavily to handle judgements of classifying, is the targeted advertisements video with definite this video segment that do not match.
The targeted advertisements video that the above-mentioned treatment scheme of process obtains; Can further add advertisement video sample storehouse to, that is to say, method of the present invention; The purpose of both having utilized the advertisement video sample to realize finding only to occur advertisement video once in the television video and get rid of the non-advertisement video that whole day repeatedly plays; The quantity that simultaneously can also utilize final output result to expand the advertisement video sample repeats this process, finally can improve greatly in the managerial efficient of television video.
In addition, adopt method of the present invention, can also be further with in the step 102 with the television video fragment of each advertisement video sample matches and targeted advertisements video as the advertisement in the television video.That is to say that method of the present invention can also be used for extracting advertisement from television video.
Please refer to Fig. 6, Fig. 6 is the structural representation block diagram of the embodiment one of the device of discovery advertisement video among the present invention.Said device comprises: ad break is confirmed unit 201, is gone heavy unit 202 and targeted advertisements to confirm unit 203.
Ad break is confirmed unit 201, is used for utilizing each advertisement video sample in the advertisement video sample storehouse of setting up in advance to confirm the ad break in the television video.
The advertisement video sample is meant the video of single advertisement, and advertisement video sample storehouse comprises the database of a plurality of advertisement videos exactly.Advertisement video sample in the advertisement video sample storehouse at first can obtain through collecting existing advertisement video, and secondly, the targeted advertisements video that utilizes the present invention to obtain finally also can join in the advertisement video sample storehouse as the advertisement video sample.
Ad break confirms that unit 201 comprises that first gathers subelement 2011, the first comparer unit 2012, set of matches and confirm that the subelement 2013 and first ad break confirm subelement 2014.
Wherein gather subelement 2011, be used for gathering the television video sample at television video.In the present embodiment, can gather one day television video as the television video sample.
The first comparer unit 2012 is used for television video sample and each advertisement video sample are compared one by one, from the television video sample, confirms each video segment with the advertisement video sample matches.
The method that the video comparison is adopted can be with reference to figure 3, and Fig. 3 is the schematic flow sheet of the method for video comparison.The television video sample is the source video among the figure, and the advertisement video sample is the target video among the figure.After source video and target video decoded, just can both be compared according to the method shown in Fig. 3.
As can beappreciated from fig. 3, when source video and target video are compared, be actually first frame of getting target video earlier, with each frame comparison of source video; If in the comparison, then do not get the next frame of target video and compare with each frame of source video, in finding the source video with target video in the frame that can compare; At this moment, note the position of this frame in the video of source, from the video of source, get next frame again; Compare with the next frame of getting in the target video,, note the successful position of frame in the video of source of last comparison up to the comparison failure; At this moment, the video segment between the position of the frame of two records of source video is the match video fragment.In seeking source video and target video, during first same number of frames, also can get first frame of source video earlier,,, get the next frame of source video and each frame of target video again and compare, by that analogy if not in the comparison with each frame comparison of target video.If there is not a frame can compare successfully in source video and the target video, the comparison failure of source video and this target video then is described, the match video fragment of this target video not in the video of source.
Each advertisement video sample is carried out above-mentioned comparison process as target video respectively, just can find each match video fragment in the television video sample.
Please continue with reference to figure 6.
Set of matches is confirmed subelement 2013, is used for each match video fragment is arranged according to the time sequencing at the television video sample, therefrom confirms continuous videos fragment collection and independent vide fragment.
As a rule; In television video advertisement playing be have necessarily successional; These advertisements all are in a period of time, to concentrate to play, and interlude is shorter between each advertisement, utilizes this characteristics; Can find out continuous videos fragment collection, this continuous videos fragment collection is meant that the reproduction time between adjacent two match video fragments is no more than the match video set of segments of predetermined value at interval.For example, five match video fragments are arranged, the reproduction time in these five match video fragments between the adjacent in twos match video fragment is spaced apart 5 seconds, then can be with these five match video fragments as a continuous videos fragment collection.Match video fragment in all match video fragments except that continuous videos fragment collection just is referred to as the independent vide fragment.
First ad break is confirmed subelement 2014, is used for continuous videos fragment collection is confirmed as ad break at the time interval and the time interval of independent vide fragment in the television video sample of television video sample.
The time interval of continuous videos fragment collection in the television video sample is the time interval between the first redundant moment to the second redundant moment; Wherein the first redundant continuous videos fragment that constantly is arranged in concentrates first match video fragment before the zero hour of television video sample; And being carved into the continuous videos fragment during this first redundancy concentrates the time span of first match video fragment between the zero hour in the television video sample to be no more than the first threshold of setting; Second redundancy is arranged in the continuous videos fragment constantly and concentrates last match video fragment after the finish time of television video sample, and the continuous videos fragment concentrates last match video fragment to be no more than second threshold value of setting the finish time in the television video sample to the time span between this second redundant moment.In like manner; The time interval of independent vide fragment in the television video sample is that triple redundance is constantly to the time interval between the 4th redundant moment; Wherein triple redundance was arranged in the independent vide fragment constantly before the zero hour of television video sample; And be carved into the 3rd threshold value that the time span of independent vide fragment between the zero hour in the television video sample is no more than setting during this triple redundance; The 4th redundant moment was arranged in the independent vide fragment after the finish time of television video sample, and the independent vide fragment is no more than the 4th threshold value of setting the finish time in the television video sample to the time span between the 4th redundant moment.
Please refer to Fig. 4, Fig. 4 is the synoptic diagram of ad break.On scheming, can find out; In the television video sample; Comprised continuous videos fragment collection and independent vide fragment, when confirming ad break, ad break is a period of time interval and a period of time interval that comprises the independent vide fragment that comprises continuous videos fragment collection.One section adjustable time is arranged in the ad break, embody by the first threshold among the figure, second threshold value, the 3rd threshold value and the 4th threshold value.It is identical that first threshold, second threshold value, the 3rd threshold value and the 4th threshold value can be set to, and also can be set to difference.For example the continuous videos fragment collection time of living in is the 10th minute to the 20th minute, and then ad break can be the 7th minute to the 23rd minute, and wherein first threshold is identical with second threshold value, is three minutes.
Go to heavy unit 202, be used for the television video of playing in the ad break and each advertisement video sample are compared, and with the unmatched television video fragment of each advertisement video sample in remove the repeated fragment.
The video segment of removing repetition is exactly that a plurality of identical video segments are carried out normalization; Keep one of them; Its implementation method is: will compare in twos with the unmatched video segment of advertisement video sample; If in the comparison then keep one of them, this video segment that keeps and the next not video segment of comparison are compared.
Targeted advertisements is confirmed unit 203, is used for confirming the targeted advertisements video from the unmatched television video fragment of removing after the repeated fragment.
Targeted advertisements is confirmed can confirm the targeted advertisements video the video segment that do not match that unit 203 obtains after going heavily to handle; Comprise dual mode: a kind of is the manual examination and verification result who utilizes the unmatched television video fragment after the removal repeated fragment, confirms the targeted advertisements video.Owing to, have only this part video segment that do not match need carry out manual examination and verification through above-mentioned treatment scheme, greatly reduce the suspicious video that needs audit, therefore greatly reduced manpower.The second way is that targeted advertisements is confirmed the sorter that unit 203 obtains through mass advertising video data training, to the judgement of classifying of the unmatched television video fragment after the said removal repeated fragment, confirms the targeted advertisements video.
Please refer to Fig. 7, Fig. 7 is the structural representation block diagram of the embodiment two of the device of discovery advertisement video among the present invention.As shown in Figure 7, said device comprises: ad break confirms unit 301, go heavy unit 302, targeted advertisements to confirm unit 303, insert unit 304 and advertisement extraction unit 305.
Wherein ad break is confirmed unit 301, is used for utilizing each advertisement video sample in the advertisement video sample storehouse of setting up in advance to confirm the ad break in the television video.Ad break is confirmed unit 301, comprises that second gathers subelement 3011, the second comparer unit 3012 and second ad break and confirm subelement 3013.
Wherein second gather subelement 3011, be used for gathering a plurality of television video samples at television video.In the present embodiment, can gather one day television video as one of them television video sample.
The second comparer unit 3012 is used for a plurality of television video samples and each advertisement video sample are compared one by one, confirms that the video segment of in the identical time period of a plurality of television video samples, all comparing with same advertisement video sample is the match video fragment.
Second ad break is confirmed subelement 3013, is used for the match video fragment is confirmed as ad break at the time interval of television video sample.
The time interval of match video fragment in the television video sample is the time interval between the 5th redundant moment to the 6th redundant moment; The said the 5th redundant moment was arranged in the match video fragment before the zero hour of television video sample; And be carved into the 5th threshold value that the time span of match video fragment between the zero hour in the television video sample is no more than setting during the 5th redundancy; The 6th redundant moment was arranged in the match video fragment after the finish time of television video sample, and the match video fragment is no more than the 6th threshold value of setting the finish time in the television video sample to the time span between the 6th redundant moment.
Go to heavy unit 302, be used for the television video in the ad break and each advertisement video sample are compared, and with the unmatched television video fragment of each advertisement video sample in remove the repeated fragment.
Targeted advertisements is confirmed unit 303, is used for confirming the targeted advertisements video from the unmatched television video fragment of removing after the repeated fragment.
The above is merely preferred embodiment of the present invention, and is in order to restriction the present invention, not all within spirit of the present invention and principle, any modification of being made, is equal to replacement, improvement etc., all should be included within the scope that the present invention protects.
Claims (12)
1. a method of finding advertisement is characterized in that, said method comprises:
A. utilize each the advertisement video sample in the advertisement video sample storehouse of setting up in advance to confirm the ad break in the television video;
B. the television video in the said ad break and each advertisement video sample are compared; And with the unmatched television video fragment of each advertisement video sample in remove the repeated fragment, confirm the targeted advertisements video the unmatched television video fragment after removing the repeated fragment.
2. method according to claim 1 is characterized in that, said method further comprises behind step B: the targeted advertisements video of confirming is put into said advertisement video sample storehouse.
3. method according to claim 1 is characterized in that, said steps A comprises:
In television video, gather the television video sample;
Said television video sample and each advertisement video sample are compared one by one, the video segment of comparing with each advertisement video sample in the said television video sample is confirmed as each match video fragment;
Each match video fragment is arranged according to the time sequencing in said television video sample; Therefrom confirm continuous videos fragment collection and independent vide fragment; Wherein said continuous videos fragment collection is the match video set of segments that the time interval between adjacent two match video fragments is no more than predetermined value, and said independent vide fragment is the match video fragment except that said continuous videos fragment collection in all match video fragments;
Said continuous videos fragment collection is set the second redundant time interval between constantly of setting duration after first constantly redundant and finish time of duration before the zero hour in said television video sample, and said independent vide fragment in said television video sample, set duration before the zero hour triple redundance constantly and the 4th redundant time interval between constantly of setting duration after the finish time confirm as said ad break.
4. method according to claim 1 is characterized in that, said steps A comprises:
In television video, gather a plurality of television video samples;
Said a plurality of television video samples and each advertisement video sample are compared one by one, confirmed that the video segment of in the identical time period of said a plurality of television video samples, all comparing with same advertisement video sample is the match video fragment;
Said match video fragment is set the 6th redundant time interval between constantly of setting duration after the 5th constantly redundant and finish time of duration before the zero hour and confirm as said ad break in said television video sample.
5. method according to claim 1 is characterized in that, after said step B, also comprises: the advertisement that will comprise as said television video with the television video fragment and the said targeted advertisements video of each advertisement video sample matches.
6. according to the described method of arbitrary claim in the claim 1 to 5, it is characterized in that, confirm the unmatched television video fragment after removing the repeated fragment that the targeted advertisements video comprises:
The manual examination and verification result of the unmatched television video fragment of utilization after to said removal repeated fragment confirms the targeted advertisements video; Perhaps,
The sorter that obtains through mass advertising video data training to the judgement of classifying of the unmatched television video fragment after the said removal repeated fragment, is confirmed the targeted advertisements video.
7. a device of finding advertisement is characterized in that, said device comprises:
Ad break is confirmed the unit, is used for utilizing each advertisement video sample in the advertisement video sample storehouse of setting up in advance to confirm the ad break in the television video;
Go to heavy unit, be used for the television video in the said ad break and each advertisement video sample are compared, and with the unmatched television video fragment of each advertisement video sample in removal repeated fragment;
Targeted advertisements is confirmed the unit, is used for confirming the targeted advertisements video from the unmatched television video fragment of removing after the repeated fragment.
8. device according to claim 7 is characterized in that said device further comprises the insertion unit, is used for the targeted advertisements video of confirming is put into said advertisement video sample storehouse.
9. device according to claim 7 is characterized in that, said ad break confirms that the unit comprises:
First gathers subelement, is used for gathering the television video sample at television video;
The first comparer unit is used for said television video sample and each advertisement video sample are compared one by one, and the video segment of comparing with each advertisement video sample in the said television video sample is confirmed as each match video fragment;
Set of matches is confirmed subelement; With each match video fragment according to time sequencing arrange; Therefrom confirm continuous videos fragment collection and independent vide fragment; Wherein said continuous videos fragment collection is the match video set of segments that the time interval between adjacent two match video fragments is no more than predetermined value, and said independent vide fragment is the match video fragment except that said continuous videos fragment collection in all match video fragments;
First ad break is confirmed subelement; Be used for said continuous videos fragment collection is set the second redundant time interval between constantly of duration said television video sample was set first constantly redundant and finish time of duration before the zero hour after, and said independent vide fragment in said television video sample, set duration before the zero hour triple redundance constantly and the time interval of setting after the finish time between the 4th redundant moment of duration confirm as said ad break.
10. device according to claim 7 is characterized in that, said ad break confirms that the unit comprises:
Second gathers subelement, is used for gathering a plurality of television video samples at television video;
The second comparer unit is used for said a plurality of television video samples and each advertisement video sample are compared one by one, confirms that the video segment of in the identical time period of said a plurality of television video samples, all comparing with same advertisement video sample is the match video fragment;
Second ad break is confirmed subelement, is used for the time interval that said match video fragment was set said television video sample was set the 5th constantly redundant and finish time of duration before the zero hour after between the 6th redundant moment of duration is confirmed as said ad break.
11. device according to claim 7 is characterized in that, said device further comprises the advertisement extraction unit, is used for the advertisement that television video fragment and said targeted advertisements video with each advertisement video sample matches are comprised as said television video.
12., it is characterized in that the manual examination and verification result that said targeted advertisements is confirmed the unmatched television video fragment after unit by using is to said removal repeated fragment confirms the targeted advertisements video according to the described device of arbitrary claim in the claim 7 to 11; Perhaps, said targeted advertisements is confirmed the sorter that the unit obtains through mass advertising video data training, to the judgement of classifying of the unmatched television video fragment after the said removal repeated fragment, confirms the targeted advertisements video.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210482882.0A CN102984585B (en) | 2011-09-20 | 2011-09-20 | A kind of method and device determining advertisement video |
| CN 201110279361 CN102346768B (en) | 2011-09-20 | 2011-09-20 | Method and device for finding video advertisement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110279361 CN102346768B (en) | 2011-09-20 | 2011-09-20 | Method and device for finding video advertisement |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210482882.0A Division CN102984585B (en) | 2011-09-20 | 2011-09-20 | A kind of method and device determining advertisement video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102346768A true CN102346768A (en) | 2012-02-08 |
| CN102346768B CN102346768B (en) | 2013-01-16 |
Family
ID=45545445
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210482882.0A Expired - Fee Related CN102984585B (en) | 2011-09-20 | 2011-09-20 | A kind of method and device determining advertisement video |
| CN 201110279361 Expired - Fee Related CN102346768B (en) | 2011-09-20 | 2011-09-20 | Method and device for finding video advertisement |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210482882.0A Expired - Fee Related CN102984585B (en) | 2011-09-20 | 2011-09-20 | A kind of method and device determining advertisement video |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN102984585B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177099A (en) * | 2013-03-20 | 2013-06-26 | 深圳先进技术研究院 | Video comparison method and video comparison system |
| CN103793423A (en) * | 2012-10-26 | 2014-05-14 | 北京海逸华清科技发展有限公司 | Method and system for discovering advertisements on basis of video |
| CN104254002A (en) * | 2013-06-25 | 2014-12-31 | 上海尚恩华科网络科技股份有限公司 | Multi-place multi-channel real-time advertisement detecting system and method |
| CN112235599A (en) * | 2020-10-14 | 2021-01-15 | 广州欢网科技有限责任公司 | Video processing method and system |
| EP3534615B1 (en) * | 2013-03-15 | 2024-01-17 | Inscape Data, Inc. | Systems and methods for real-time television ad detection using an automated content recognition database |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113194332B (en) * | 2021-04-27 | 2022-04-29 | 北京市博汇科技股份有限公司 | Multi-policy-based new advertisement discovery method, electronic device and readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1543096A (en) * | 2003-04-30 | 2004-11-03 | ���µ�����ҵ��ʽ���� | Equipment and method for automatic detection of television advertisement |
| US20050149968A1 (en) * | 2003-03-07 | 2005-07-07 | Richard Konig | Ending advertisement insertion |
| CN101540907A (en) * | 2009-04-28 | 2009-09-23 | 无锡天脉聚源传媒科技有限公司 | Method for auditing safety of internet video |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7089575B2 (en) * | 2001-09-04 | 2006-08-08 | Koninklijke Philips Electronics N.V. | Method of using transcript information to identify and learn commercial portions of a program |
| US8874468B2 (en) * | 2007-04-20 | 2014-10-28 | Google Inc. | Media advertising |
| CN101996313A (en) * | 2009-08-18 | 2011-03-30 | 宏碁股份有限公司 | System and method for detecting advertisement |
-
2011
- 2011-09-20 CN CN201210482882.0A patent/CN102984585B/en not_active Expired - Fee Related
- 2011-09-20 CN CN 201110279361 patent/CN102346768B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050149968A1 (en) * | 2003-03-07 | 2005-07-07 | Richard Konig | Ending advertisement insertion |
| CN1543096A (en) * | 2003-04-30 | 2004-11-03 | ���µ�����ҵ��ʽ���� | Equipment and method for automatic detection of television advertisement |
| CN101540907A (en) * | 2009-04-28 | 2009-09-23 | 无锡天脉聚源传媒科技有限公司 | Method for auditing safety of internet video |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793423A (en) * | 2012-10-26 | 2014-05-14 | 北京海逸华清科技发展有限公司 | Method and system for discovering advertisements on basis of video |
| EP3534615B1 (en) * | 2013-03-15 | 2024-01-17 | Inscape Data, Inc. | Systems and methods for real-time television ad detection using an automated content recognition database |
| CN103177099A (en) * | 2013-03-20 | 2013-06-26 | 深圳先进技术研究院 | Video comparison method and video comparison system |
| CN103177099B (en) * | 2013-03-20 | 2017-05-24 | 深圳先进技术研究院 | Video comparison method and video comparison system |
| CN104254002A (en) * | 2013-06-25 | 2014-12-31 | 上海尚恩华科网络科技股份有限公司 | Multi-place multi-channel real-time advertisement detecting system and method |
| CN112235599A (en) * | 2020-10-14 | 2021-01-15 | 广州欢网科技有限责任公司 | Video processing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102984585B (en) | 2015-09-23 |
| CN102984585A (en) | 2013-03-20 |
| CN102346768B (en) | 2013-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102346768B (en) | Method and device for finding video advertisement | |
| CN104394433B (en) | The method and apparatus for detecting the broadcasting time of multimedia file in television channel | |
| KR101456926B1 (en) | System and method for detecting advertisement based on fingerprint | |
| CN103299648B (en) | Method for receiving enhanced service and display apparatus thereof | |
| CN108521582B (en) | Copyright video full-network viewing record system based on block chain technology | |
| CN101821734B (en) | Detection and classification of matches between time-based media | |
| CN102314916B (en) | Video processing method and system | |
| WO2015074503A1 (en) | Statistical method and apparatus for webpage access data | |
| CN103488635A (en) | Method and device for acquiring product information | |
| TW201713120A (en) | Method and device for monitoring video advertisement displaying | |
| EP2127400A1 (en) | System and method for monitoring and recognizing broadcast data | |
| CN102682024B (en) | Method for recombining incomplete JPEG file fragmentation | |
| CN1577600A (en) | Network system, server, data recording and playing device, method for the same, and program | |
| CN105681908A (en) | Broadcast television system based on individual watching behaviour and personalized programme recommendation method thereof | |
| EP3776263B1 (en) | System and method for detecting repeating content, including commercials, in a video data stream using audio-based and video-based automated content recognition | |
| US20100329563A1 (en) | System and Method for Real-time New Event Detection on Video Streams | |
| CN103794209A (en) | System for monitoring and playing advertisement before movie playing based on audio fingerprint identification technology | |
| CN106604068B (en) | A kind of method and its system of more new media program | |
| CN104185056A (en) | Uniform program arranging method and apparatus, and LED program displaying system | |
| CN103596022A (en) | Method, device and system for content matching | |
| George et al. | Youtube decade: Cultural convergence in recorded music | |
| US10614121B2 (en) | Program and syndicated content detection | |
| CN110162513A (en) | Tables of data connection method and computer readable storage medium for the processing of smart grid big data | |
| JP3983532B2 (en) | Scene extraction device | |
| CN101883252A (en) | Digital television visual media information feedback system and method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 100085, B, block 1-2322, golden building, No. 38 clear road, Beijing, Haidian District Patentee after: Beijing Sunway Eagle Polytron Technologies Inc Address before: 100085, B, block 1-2322, golden building, No. 38 clear road, Beijing, Haidian District Patentee before: Beijing Pengrun Hongtu Technology Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130116 Termination date: 20190920 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |