CN106375867A - Method and device for cutting advertisement in television video - Google Patents
Method and device for cutting advertisement in television video Download PDFInfo
- Publication number
- CN106375867A CN106375867A CN201610704826.5A CN201610704826A CN106375867A CN 106375867 A CN106375867 A CN 106375867A CN 201610704826 A CN201610704826 A CN 201610704826A CN 106375867 A CN106375867 A CN 106375867A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- time point
- frame
- television video
- video stream
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005520 cutting process Methods 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 10
- 230000008569 process Effects 0.000 description 12
- 238000004590 computer program Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000005611 electricity Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003860 storage Methods 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]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application
- H04N21/47815—Electronic shopping
-
- 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
- H04N21/439—Processing of audio elementary streams
-
- 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 or rendering scenes according to encoded video stream scene graphs
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- Marketing (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention discloses a method and a device for cutting an advertisement in a television video, which are used for realizing a purpose of cutting the advertisement in the television video accurately and quickly. The method comprises the following steps of obtaining a television video stream and recognizing voice features in the television video stream; comparing the voice features in the television video stream with voice features of advertisements of a preset advertisement library and outputting a comparison result once at every first duration to obtain a plurality of comparison results; determining a starting time point and an end time point of the advertisement in the television video according to the comparison results; and carrying out cutting processing on the advertisement in the television video according to the starting time point and the end time point.
Description
Technical field
The present invention relates to technical field of internet application, particularly to a kind of method cutting advertisement in television video and
Device.
Background technology
Television advertising is a kind of advertisement form via TV broadcast, is commonly used to publicize commodity, service, tissue, concept
Deng.Most television advertising is to be made by advertising company, and buys number when playing to television station.Television advertising is developed so far
My god, its length all has from the several seconds to several minutes, also has the advertisement magazine up to 10 minutes, and up to whole program time-interval
" information type advertisement ", also known as TV shopping.
At present, when processing to television video it is sometimes desirable in television video, the advertisement broadcasted is accurately identified
And cut out, wasted time and energy by the method that artificial cognition confirms it would be highly desirable to solve this problem in correlation technique.
Content of the invention
The present invention provides a kind of method and device cutting advertisement in television video, in order to realize accurate, quick in electricity
Depending on cutting the purpose of advertisement in video.
The present invention provides a kind of method cutting advertisement in television video, comprising:
Obtain television video stream, identify the sound characteristic in described television video stream；
The sound characteristic of the sound characteristic in described television video stream and advertisement in default advertisement base is compared, often
First duration exports a comparison result, obtains multiple comparison results；
According to the plurality of comparison result, determine the start time point of advertisement in described television video stream and end time
Point；
According to described start time point and point of described end time, cut place is carried out to the advertisement in described television video stream
Reason.
In an embodiment of the present invention, described comparison result includes:
Sound characteristic in described television video stream belongs to the sound characteristic of advertisement in described default advertisement base；Or
Sound characteristic in described television video stream is not belonging to the sound characteristic of advertisement in described default advertisement base.
In an embodiment of the present invention, according to the plurality of comparison result, determine advertisement in described television video stream
Start time point and end time point, comprising:
The time of adjacent and consistent comparison result in the plurality of comparison result is merged, when obtaining merging
Between；
Using the start time point of described merging time as the advertisement in described television video stream start time point, and
Using the end time point of described merging time as the advertisement in described television video stream end time point.
In an embodiment of the present invention, according to described start time point and point of described end time, to described television video
Advertisement in stream carries out cutting process, comprising:
Described start time point is expanded forward the second duration, the start time point after being expanded；
Described end time point is expanded the 3rd duration backward, the end time point after being expanded；
Identification described expand after start time point and the end time point after described expansion between advertisement start frame and
Advertisement end frame；
Based on described advertisement start frame and described advertisement end frame, cut place is carried out to the advertisement in described television video stream
Reason.
In an embodiment of the present invention, the start time point after the described expansion of identification and the end time point after described expansion
Between advertisement start frame and advertisement end frame, comprising:
Extract the phash feature of the phash feature of the first frame of advertisement and tail frame in described default advertisement base；
Start time point after described expansion starts to extract from front to back the phash feature of picture frame, by described image
The phash feature of frame with described head frame phash feature compare, determine with described head frame phash feature similarity
More than the first specified threshold picture frame as described advertisement start frame；
End time point after described expansion starts to extract from back to front the phash feature of picture frame, by described image
The phash feature of frame is compared with the phash feature of described tail frame, determines the similarity with the phash feature of described tail frame
More than the second specified threshold picture frame as described advertisement end frame.
In an embodiment of the present invention, the start time point after the described expansion of identification and the end time point after described expansion
Between advertisement start frame, comprising:
Carry out profile using the picture frame that canny operator starts from front to back to the start time point after described expansion to carry
Take；
Using otsu algorithm to the contour images binaryzation obtaining, obtain a black and white binary map comprising profile information,
Black and white binary map is divided into the fritter of specified quantity, calculates the ratio that white pixel point in each fritter accounts for all pixels point,
Obtain the vector of described image frame；
Calculate the vector of described image frame and the Euclidean distance of the vector of first frame of advertisement in described default advertisement base, really
Determine Euclidean distance and be less than the picture frame of the first distance to a declared goal threshold value as described advertisement start frame.
In an embodiment of the present invention, the start time point after the described expansion of identification and the end time point after described expansion
Between advertisement end frame, comprising:
Carry out profile using the picture frame that canny operator starts from back to front to the end time point after described expansion to carry
Take；
Using otsu algorithm to the contour images binaryzation obtaining, obtain a black and white binary map comprising profile information,
Black and white binary map is divided into the fritter of specified quantity, calculates the ratio that white pixel point in each fritter accounts for all pixels point,
Obtain the vector of described image frame；
Calculate the vector of described image frame and the Euclidean distance of the vector of tail frame of advertisement in described default advertisement base, really
Determine Euclidean distance and be less than the picture frame of the second distance to a declared goal threshold value as described advertisement end frame.
The present invention also provides a kind of device cutting advertisement in television video, comprising:
Identification module, for obtaining television video stream, identifies the sound characteristic in described television video stream；
Comparing module, for will in the sound characteristic in described television video stream and default advertisement base advertisement sound special
Levy and compare, every first duration exports a comparison result, obtains multiple comparison results；
Determining module, for according to the plurality of comparison result, determine advertisement in described television video stream initial when
Between point and end time point；
Cutting module, for according to described start time point and point of described end time, in described television video stream
Advertisement carries out cutting process.
In an embodiment of the present invention, described comparison result includes:
Sound characteristic in described television video stream belongs to the sound characteristic of advertisement in described default advertisement base；Or
Sound characteristic in described television video stream is not belonging to the sound characteristic of advertisement in described default advertisement base.
In an embodiment of the present invention, described determining module is additionally operable to:
The time of adjacent and consistent comparison result in the plurality of comparison result is merged, when obtaining merging
Between；
Using the start time point of described merging time as the advertisement in described television video stream start time point, and
Using the end time point of described merging time as the advertisement in described television video stream end time point.
In embodiments of the present invention, television video stream, the sound characteristic in identification television video stream are obtained；By television video
The sound characteristic of the sound characteristic in stream and advertisement in default advertisement base is compared, and every first duration output once compares knot
Really, multiple comparison results are obtained；Subsequently, according to multiple comparison results, determine the start time point of the advertisement in television video stream
With end time point；And then according to start time point and end time point, cutting process is carried out to the advertisement in television video stream.
Sound is one-dimensional signal, and the embodiment of the present invention identifies the advertisement in television video based on sound characteristic, and amount of calculation is few, is capable of
Accurately, the quick purpose cutting advertisement in television video.
Other features and advantages of the present invention will illustrate in the following description, and, partly become from description
Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages can be by the explanations write
In book, claims and accompanying drawing, specifically noted structure is realizing and to obtain.
Below by drawings and Examples, technical scheme is described in further detail.
Brief description
Accompanying drawing is used for providing a further understanding of the present invention, and constitutes a part for description, the reality with the present invention
Apply example and be used for explaining the present invention together, be not construed as limiting the invention.In the accompanying drawings:
The flow chart that Fig. 1 show the method cutting advertisement in one embodiment of the invention in television video；
The flow chart that Fig. 2 show the method cutting advertisement in another embodiment of the present invention in television video；And
Fig. 3 show the structural representation of the device cutting advertisement in one embodiment of the invention in television video.
Specific embodiment
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are illustrated it will be appreciated that preferred reality described herein
Apply example to be merely to illustrate and explain the present invention, be not intended to limit the present invention.
The flow chart being illustrated in figure 1 the method cutting advertisement in one embodiment of the invention in television video, the method
May comprise steps of s11-s14.
Step s11, obtains television video stream, the sound characteristic in identification television video stream.
In this step, it is possible to use existing voice recognition technology identifying the sound characteristic in television video stream, this
Place repeats no more.
Step s12, the sound characteristic of the sound characteristic in television video stream and advertisement in default advertisement base is compared
Right, every first duration exports a comparison result, obtains multiple comparison results.
In this step, the sound characteristic that comparison result is included in television video stream belongs to advertisement in default advertisement base
Sound characteristic；Or the sound characteristic in television video stream is not belonging to the sound characteristic of advertisement in default advertisement base.When first
Length can be configured according to the actual requirements, a length of 2 seconds etc. when such as arranging first, the invention is not restricted to this.
Step s13, according to multiple comparison results, determines the start time point of advertisement and the end time in television video stream
Point.
Step s14, according to start time point and end time point, carries out cutting process to the advertisement in television video stream.
In embodiments of the present invention, television video stream, the sound characteristic in identification television video stream are obtained；By television video
The sound characteristic of the sound characteristic in stream and advertisement in default advertisement base is compared, and every first duration output once compares knot
Really, multiple comparison results are obtained；Subsequently, according to multiple comparison results, determine the start time point of the advertisement in television video stream
With end time point；And then according to start time point and end time point, cutting process is carried out to the advertisement in television video stream.
Sound is one-dimensional signal, and the embodiment of the present invention identifies the advertisement in television video based on sound characteristic, and amount of calculation is few, is capable of
Accurately, the quick purpose cutting advertisement in television video.
According to multiple comparison results in above step s13, determine the start time point of advertisement and the knot in television video stream
Bundle time point, embodiments provides a kind of optional scheme, the program comprises the following steps a1-a2.
Step a1, is merged, is merged the time of adjacent and consistent comparison result in multiple comparison results
Time.
Step a2, using merge the time start time point as the advertisement in television video stream start time point, and
Using merge the time end time point as the advertisement in television video stream end time point.
In the above-described embodiments, using voice recognition technology, advertisement identification is carried out to television video stream, in advertisement base
Advertisement is compared, comparison result of output in every 2 seconds, is then closed the time of the adjacent identical comparison result of output
And.Due to can there is the factor such as quiet in advertisement during broadcasting it is thus possible to lead to have certain mistake output, due to reality
In there is not the advertisement of 2 seconds, there are certain two seconds negligible, Ke Yihes different from other in comparison result in therefore one time period
And, the approximate time point of certain advertisement in television video stream is so quickly obtained by voice recognition.
According to start time point and end time point in above step s14, the advertisement in television video stream is cut
Process, embodiments provide a kind of optional scheme, the program comprises the following steps b1-b4.
Step b1, start time point is expanded forward the second duration, the start time point after being expanded.
Step b2, end time point is expanded the 3rd duration backward, the end time point after being expanded.
Step b3, identification expand after start time point and the end time point after expanding between advertisement start frame and extensively
Come to an end frame.
Step b4, based on advertisement start frame and advertisement end frame, carries out cutting process to the advertisement in television video stream.
In the alternative embodiment of the present invention, the start time point of the time spot section that voice recognition is obtained expands forward 5
Second, end time point expands 5 seconds backward, and then carries out the accurate judgement of start frame and end frame using image recognition algorithm.Initial
Frame is searched for from front to back, and end frame is searched for from back to front.For avoiding the impact such as word of edge's appearance, image in video playback
Compare operation can the heart 3/4 part carry out in the picture.Be described in detail below using image recognition algorithm carry out advertisement initiate
Frame and the accurate judgement of advertisement end frame.
Mode one, extracts the mode of the phash feature of picture frame.
Phash feature saves the low-frequency information of image, as long as the overall structure of advertisement head (tail) frame is constant, phash is special
Levy to compare and have very high similarity.So, when identifying advertisement start frame and advertisement end frame, first, extraction is default wide
Accuse the phash feature of first frame of advertisement and the phash feature of tail frame in storehouse；Then, from expand after start time point start from
Extract the phash feature of picture frame after forward direction, the phash feature of the phash feature of picture frame and first frame is compared, determines
It is more than the picture frame of the first specified threshold with the similarity of the phash feature of first frame as advertisement start frame；Knot after expanding
Bundle time point starts to extract from back to front the phash feature of picture frame, will be special with the phash of tail frame for the phash feature of picture frame
Levy and compare, determine that the similarity with the phash feature of tail frame terminates as advertisement more than the picture frame of the second specified threshold
Frame.
Mode two, extracts the mode of the profile of picture frame.
When identifying advertisement start frame, first with canny operator, (target of canny is to find an optimum edge
Detection algorithm) contours extract is carried out to the picture frame that starts from front to back of start time point after expanding；(maximum using otsu
Ostu method) algorithm, to the contour images binaryzation obtaining, obtains a black and white binary map comprising profile information, by black and white
Binary map is divided into the fritter of specified quantity (as 64 etc.), calculates the ratio that white pixel point in each fritter accounts for all pixels point
Example, obtains the vector of picture frame；Finally, calculate the Europe of the vector of first frame of advertisement in the vectorial and default advertisement base of picture frame
Formula distance, determines that Euclidean distance is less than the picture frame of the first distance to a declared goal threshold value as advertisement start frame.
When identifying advertisement end frame, first with canny operator, the end time point after expanding is started from backward
Front picture frame carries out contours extract；Using otsu algorithm to the contour images binaryzation obtaining, obtain one and comprise profile letter
The black and white binary map of breath, black and white binary map is divided into the fritter of specified quantity, calculates white pixel point in each fritter and account for institute
There is the ratio of pixel, obtain the vector of picture frame；Calculate the tail frame of advertisement in the vectorial and default advertisement base of picture frame
The Euclidean distance of vector, determines that Euclidean distance is less than the picture frame of the second distance to a declared goal threshold value as advertisement end frame.
In actual applications, the identification of advertisement start frame and advertisement end frame can be carried out with employing mode one or mode two,
The mode that combines of employing mode one and mode two can also carry out advertisement start frame and the identification of advertisement end frame.
It is described above multiple implementations of the links of embodiment illustrated in fig. 1, below by a specific embodiment
Come to be discussed in detail the present invention in television video cut advertisement method realize process.
The flow chart being illustrated in figure 2 the method cutting advertisement in another embodiment of the present invention in television video, the party
Method may comprise steps of s21-s27.
Step s21, obtains television video stream, the sound characteristic in identification television video stream.
In this step, it is possible to use existing voice recognition technology identifying the sound characteristic in television video stream, this
Place repeats no more.
Step s22, the sound characteristic of the sound characteristic in television video stream and advertisement in default advertisement base is compared
Right, every first duration exports a comparison result, obtains multiple comparison results.
In this step, the sound characteristic that comparison result is included in television video stream belongs to advertisement in default advertisement base
Sound characteristic；Or the sound characteristic in television video stream is not belonging to the sound characteristic of advertisement in default advertisement base.When first
Length can be configured according to the actual requirements, a length of 2 seconds etc. when such as arranging first, the invention is not restricted to this.
Step s23, is merged, is merged the time of adjacent and consistent comparison result in multiple comparison results
Time.
Step s24, using merge the time start time point as the advertisement in television video stream start time point, and
Using merge the time end time point as the advertisement in television video stream end time point.
Due to can there is the factor such as quiet in advertisement during broadcasting it is thus possible to lead to have certain mistake output, by
Exist in comparison result in the advertisement not having 2 seconds in practice, therefore one time period certain two seconds different from other negligible,
Can merge, so be quickly obtained the approximate time point of certain advertisement in television video stream by voice recognition.
Step s25, start time point is expanded forward the second duration, the start time point after being expanded；At the end of inciting somebody to action
Between put backward expansion the 3rd duration, the end time point after being expanded.
Step s26, identification expand after start time point and expand after end time point between advertisement start frame and
Advertisement end frame.
In this step, the advertisement start frame between the start time point after identification is expanded and the end time point after expansion
Can in the way of using introducing above one or mode two with advertisement end frame, it would however also be possible to employ the side that mode one and mode two combine
Formula, here is omitted.
Step s27, based on advertisement start frame and advertisement end frame, carries out cutting process to the advertisement in television video stream.
In embodiments of the present invention, it is primarily based on sound characteristic and carries out preliminary treatment, because sound is one-dimensional signal, so
The approximate time point of certain advertisement in television video stream can quickly be obtained.Further, do accurate based on image recognition algorithm
Join, from the purpose realizing accurate, quick cutting advertisement television video.
It should be noted that in actual applications, above-mentioned all optional embodiments can be any in the way of using combining
Combination, forms the alternative embodiment of the present invention, this is no longer going to repeat them.
Method corresponding to cutting advertisement in above-described embodiment in television video, the present invention also provides one kind in TV
The device of advertisement is cut in video.It is illustrated in figure 3 the device cutting advertisement in one embodiment of the invention in television video
Structural representation, comprising:
Identification module 31, for obtaining television video stream, identifies the sound characteristic in described television video stream；
Comparing module 32, for by the sound of advertisement in the sound characteristic in described television video stream and default advertisement base
Feature is compared, and every first duration exports a comparison result, obtains multiple comparison results；
Determining module 33, for according to the plurality of comparison result, determining initiateing of the advertisement in described television video stream
Time point and end time point；
Cutting module 34, for according to described start time point and point of described end time, in described television video stream
Advertisement carry out cutting process.
In an embodiment of the present invention, described comparison result includes:
Sound characteristic in described television video stream belongs to the sound characteristic of advertisement in described default advertisement base；Or
Sound characteristic in described television video stream is not belonging to the sound characteristic of advertisement in described default advertisement base.
In an embodiment of the present invention, described determining module 33 is additionally operable to:
The time of adjacent and consistent comparison result in the plurality of comparison result is merged, when obtaining merging
Between；
Using the start time point of described merging time as the advertisement in described television video stream start time point, and
Using the end time point of described merging time as the advertisement in described television video stream end time point.
In an embodiment of the present invention, described cutting module 34 is additionally operable to:
Described start time point is expanded forward the second duration, the start time point after being expanded；
Described end time point is expanded the 3rd duration backward, the end time point after being expanded；
Identification described expand after start time point and the end time point after described expansion between advertisement start frame and
Advertisement end frame；
Based on described advertisement start frame and described advertisement end frame, cut place is carried out to the advertisement in described television video stream
Reason.
In an embodiment of the present invention, described cutting module 34 is additionally operable to:
Extract the phash feature of the phash feature of the first frame of advertisement and tail frame in described default advertisement base；
Start time point after described expansion starts to extract from front to back the phash feature of picture frame, by described image
The phash feature of frame with described head frame phash feature compare, determine with described head frame phash feature similarity
More than the first specified threshold picture frame as described advertisement start frame；
End time point after described expansion starts to extract from back to front the phash feature of picture frame, by described image
The phash feature of frame is compared with the phash feature of described tail frame, determines the similarity with the phash feature of described tail frame
More than the second specified threshold picture frame as described advertisement end frame.
In an embodiment of the present invention, described cutting module 34 is additionally operable to:
Carry out profile using the picture frame that canny operator starts from front to back to the start time point after described expansion to carry
Take；
Using otsu algorithm to the contour images binaryzation obtaining, obtain a black and white binary map comprising profile information,
Black and white binary map is divided into the fritter of specified quantity, calculates the ratio that white pixel point in each fritter accounts for all pixels point,
Obtain the vector of described image frame；
Calculate the vector of described image frame and the Euclidean distance of the vector of first frame of advertisement in described default advertisement base, really
Determine Euclidean distance and be less than the picture frame of the first distance to a declared goal threshold value as described advertisement start frame.
In an embodiment of the present invention, described cutting module 34 is additionally operable to:
Carry out profile using the picture frame that canny operator starts from back to front to the end time point after described expansion to carry
Take；
Using otsu algorithm to the contour images binaryzation obtaining, obtain a black and white binary map comprising profile information,
Black and white binary map is divided into the fritter of specified quantity, calculates the ratio that white pixel point in each fritter accounts for all pixels point,
Obtain the vector of described image frame；
Calculate the vector of described image frame and the Euclidean distance of the vector of tail frame of advertisement in described default advertisement base, really
Determine Euclidean distance and be less than the picture frame of the second distance to a declared goal threshold value as described advertisement end frame.
The said apparatus of the embodiment of the present invention: obtain television video stream, the sound characteristic in identification television video stream；By electricity
Sound characteristic depending on the sound characteristic in video flowing and advertisement in default advertisement base is compared, and every first duration exports once
Comparison result, obtains multiple comparison results；Subsequently, according to multiple comparison results, determine initiateing of the advertisement in television video stream
Time point and end time point；And then according to start time point and end time point, the advertisement in television video stream is cut
Cut process.Sound is one-dimensional signal, and the embodiment of the present invention identifies the advertisement in television video based on sound characteristic, and amount of calculation is few,
It is capable of accurate, the quick purpose cutting advertisement in television video.
Those skilled in the art are it should be appreciated that embodiments of the invention can be provided as method, system or computer program
Product.Therefore, the present invention can be using complete hardware embodiment, complete software embodiment or the reality combining software and hardware aspect
Apply the form of example.And, the present invention can be using in one or more computers wherein including computer usable program code
The shape of the upper computer program implemented of usable storage medium (including but not limited to disk memory and optical memory etc.)
Formula.
The present invention is the flow process with reference to method according to embodiments of the present invention, equipment (system) and computer program
Figure and/or block diagram are describing.It should be understood that can be by each stream in computer program instructions flowchart and/or block diagram
Flow process in journey and/or square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processor instructing general purpose computer, special-purpose computer, Embedded Processor or other programmable data processing device is to produce
A raw machine is so that produced for reality by the instruction of computer or the computing device of other programmable data processing device
The device of the function of specifying in present one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions may be alternatively stored in and can guide computer or other programmable data processing device with spy
Determine in the computer-readable memory that mode works so that the instruction generation inclusion being stored in this computer-readable memory refers to
Make the manufacture of device, this command device realize in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or
The function of specifying in multiple square frames.
These computer program instructions also can be loaded in computer or other programmable data processing device so that counting
On calculation machine or other programmable devices, execution series of operation steps to be to produce computer implemented process, thus in computer or
On other programmable devices, the instruction of execution is provided for realizing in one flow process of flow chart or multiple flow process and/or block diagram one
The step of the function of specifying in individual square frame or multiple square frame.
Obviously, those skilled in the art can carry out the various changes and modification essence without deviating from the present invention to the present invention
God and scope.So, if these modifications of the present invention and modification belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprise these changes and modification.
Claims (10)
1. a kind of method cutting advertisement in television video is it is characterised in that include:
Obtain television video stream, identify the sound characteristic in described television video stream；
The sound characteristic of the sound characteristic in described television video stream and advertisement in default advertisement base is compared, every first
Duration exports a comparison result, obtains multiple comparison results；
According to the plurality of comparison result, determine the start time point of advertisement in described television video stream and end time point；
According to described start time point and point of described end time, cutting process is carried out to the advertisement in described television video stream.
2. method according to claim 1 is it is characterised in that described comparison result includes:
Sound characteristic in described television video stream belongs to the sound characteristic of advertisement in described default advertisement base；Or
Sound characteristic in described television video stream is not belonging to the sound characteristic of advertisement in described default advertisement base.
3. method according to claim 1 and 2 is it is characterised in that according to the plurality of comparison result, determine described TV
The start time point of the advertisement in video flowing and end time point, comprising:
The time of adjacent and consistent comparison result in the plurality of comparison result is merged, obtains the merging time；
Using the start time point of described merging time as the advertisement in described television video stream start time point, and by institute
State the end time point as the advertisement in described television video stream for the end time point of merging time.
4. method according to claim 3 is it is characterised in that according to described start time point and point of described end time,
Cutting process is carried out to the advertisement in described television video stream, comprising:
Described start time point is expanded forward the second duration, the start time point after being expanded；
Described end time point is expanded the 3rd duration backward, the end time point after being expanded；
Advertisement start frame between start time point after the described expansion of identification and the end time point after described expansion and advertisement
End frame；
Based on described advertisement start frame and described advertisement end frame, cutting process is carried out to the advertisement in described television video stream.
5. method according to claim 4 is it is characterised in that identify the start time point after described expansion and described expansion
The advertisement start frame between end time point afterwards and advertisement end frame, comprising:
Extract the phash feature of the phash feature of the first frame of advertisement and tail frame in described default advertisement base；
Start time point after described expansion starts to extract from front to back the phash feature of picture frame, by described image frame
The phash feature of phash feature and described head frame is compared, and determines that the similarity with the phash feature of described first frame is more than
The picture frame of the first specified threshold is as described advertisement start frame；
End time point after described expansion starts to extract from back to front the phash feature of picture frame, by described image frame
The phash feature of phash feature and described tail frame is compared, and determines that the similarity with the phash feature of described tail frame is more than
The picture frame of the second specified threshold is as described advertisement end frame.
6. method according to claim 4 is it is characterised in that identify the start time point after described expansion and described expansion
The advertisement start frame between end time point afterwards, comprising:
Carry out contours extract using the picture frame that canny operator starts from front to back to the start time point after described expansion；
Using otsu algorithm to the contour images binaryzation obtaining, obtain a black and white binary map comprising profile information, will be black
White binary map is divided into the fritter of specified quantity, calculates the ratio that white pixel point in each fritter accounts for all pixels point, obtains
The vector of described image frame；
Calculate vector and the Euclidean distance of the vector of first frame of advertisement in described default advertisement base of described image frame, determine Europe
Formula distance less than the first distance to a declared goal threshold value picture frame as described advertisement start frame.
7. method according to claim 4 is it is characterised in that identify the start time point after described expansion and described expansion
The advertisement end frame between end time point afterwards, comprising:
Carry out contours extract using the picture frame that canny operator starts from back to front to the end time point after described expansion；
Using otsu algorithm to the contour images binaryzation obtaining, obtain a black and white binary map comprising profile information, will be black
White binary map is divided into the fritter of specified quantity, calculates the ratio that white pixel point in each fritter accounts for all pixels point, obtains
The vector of described image frame；
Calculate vector and the Euclidean distance of the vector of tail frame of advertisement in described default advertisement base of described image frame, determine Europe
Formula distance less than the second distance to a declared goal threshold value picture frame as described advertisement end frame.
8. a kind of device cutting advertisement in television video is it is characterised in that include:
Identification module, for obtaining television video stream, identifies the sound characteristic in described television video stream；
Comparing module, for entering the sound characteristic of the sound characteristic in described television video stream and advertisement in default advertisement base
Row compares, and every first duration exports a comparison result, obtains multiple comparison results；
Determining module, for according to the plurality of comparison result, determining the start time point of the advertisement in described television video stream
With end time point；
Cutting module, for according to described start time point and point of described end time, to the advertisement in described television video stream
Carry out cutting process.
9. device according to claim 8 is it is characterised in that described comparison result includes:
Sound characteristic in described television video stream belongs to the sound characteristic of advertisement in described default advertisement base；Or
Sound characteristic in described television video stream is not belonging to the sound characteristic of advertisement in described default advertisement base.
10. device according to claim 8 or claim 9 is it is characterised in that described determining module is additionally operable to:
The time of adjacent and consistent comparison result in the plurality of comparison result is merged, obtains the merging time；
Using the start time point of described merging time as the advertisement in described television video stream start time point, and by institute
State the end time point as the advertisement in described television video stream for the end time point of merging time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610704826.5A CN106375867A (en) | 2016-08-22 | 2016-08-22 | Method and device for cutting advertisement in television video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610704826.5A CN106375867A (en) | 2016-08-22 | 2016-08-22 | Method and device for cutting advertisement in television video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106375867A true CN106375867A (en) | 2017-02-01 |
Family
ID=57879447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610704826.5A Pending CN106375867A (en) | 2016-08-22 | 2016-08-22 | Method and device for cutting advertisement in television video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106375867A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107948734A (en) * | 2017-12-05 | 2018-04-20 | 深圳市茁壮网络股份有限公司 | The screening technique and system of a kind of advertisement video |
| CN108737786A (en) * | 2018-05-22 | 2018-11-02 | 浙江磐至科技有限公司 | A kind of intelligent monitor system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW200529665A (en) * | 2004-02-25 | 2005-09-01 | Bing-Chiang Jeng | Method and system for detection of advertisements located within TV programs |
| US8494234B1 (en) * | 2007-03-07 | 2013-07-23 | MotionDSP, Inc. | Video hashing system and method |
| CN103605666A (en) * | 2013-10-28 | 2014-02-26 | 复旦大学 | Video copying detection method for advertisement detection |
| CN103617263A (en) * | 2013-11-29 | 2014-03-05 | 安徽大学 | Automatic TV advertisement movie clip detection method based on multi-mode features |
| CN104678963A (en) * | 2015-02-03 | 2015-06-03 | 葛武 | System and method for acquiring instrumentation and equipment information based on computer vision |
| CN105184610A (en) * | 2015-09-02 | 2015-12-23 | 王磊 | Real-time mobile advertisement synchronous putting method and device based on audio fingerprints |
-
2016
- 2016-08-22 CN CN201610704826.5A patent/CN106375867A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW200529665A (en) * | 2004-02-25 | 2005-09-01 | Bing-Chiang Jeng | Method and system for detection of advertisements located within TV programs |
| US8494234B1 (en) * | 2007-03-07 | 2013-07-23 | MotionDSP, Inc. | Video hashing system and method |
| CN103605666A (en) * | 2013-10-28 | 2014-02-26 | 复旦大学 | Video copying detection method for advertisement detection |
| CN103617263A (en) * | 2013-11-29 | 2014-03-05 | 安徽大学 | Automatic TV advertisement movie clip detection method based on multi-mode features |
| CN104678963A (en) * | 2015-02-03 | 2015-06-03 | 葛武 | System and method for acquiring instrumentation and equipment information based on computer vision |
| CN105184610A (en) * | 2015-09-02 | 2015-12-23 | 王磊 | Real-time mobile advertisement synchronous putting method and device based on audio fingerprints |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107948734A (en) * | 2017-12-05 | 2018-04-20 | 深圳市茁壮网络股份有限公司 | The screening technique and system of a kind of advertisement video |
| CN107948734B (en) * | 2017-12-05 | 2020-09-11 | 深圳市茁壮网络股份有限公司 | Method and system for screening advertisement videos |
| CN108737786A (en) * | 2018-05-22 | 2018-11-02 | 浙江磐至科技有限公司 | A kind of intelligent monitor system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111460219B (en) | Video processing method and device and short video platform | |
| CN109145784B (en) | Method and apparatus for processing video | |
| KR100717402B1 (en) | Apparatus and method for determining genre of multimedia data | |
| US20100172591A1 (en) | Image-sound segment corresponding apparatus, method and program | |
| CN111861572B (en) | Advertisement putting method and device, electronic equipment and computer readable storage medium | |
| US20040062520A1 (en) | Enhanced commercial detection through fusion of video and audio signatures | |
| CN104516501A (en) | Display device and content display method | |
| KR20200054613A (en) | Video metadata tagging system and method thereof | |
| CN111988658A (en) | Video generation method and device | |
| JP2007143114A (en) | Multi-modal based video abstract generation method and apparatus thereof | |
| CN109977779B (en) | Method for identifying advertisement inserted in video creative | |
| CA2977680C (en) | Detecting of graphical objects to identify video demarcations | |
| CN113613065A (en) | Video editing method and device, electronic equipment and storage medium | |
| CN113822972B (en) | Video-based processing method, device and readable medium | |
| CN110856039A (en) | Video processing method and device and storage medium | |
| CN113343831A (en) | Method and device for classifying speakers in video, electronic equipment and storage medium | |
| CN114143575A (en) | Video editing method and device, computing equipment and storage medium | |
| CN114598933B (en) | Video content processing method, system, terminal and storage medium | |
| CN106375867A (en) | Method and device for cutting advertisement in television video | |
| CN106507175A (en) | Method of video image processing and device | |
| CN114339451A (en) | Video editing method and device, computing equipment and storage medium | |
| CN114302174A (en) | Video editing method and device, computing equipment and storage medium | |
| CN109522799A (en) | Information cuing method, device, computer equipment and storage medium | |
| US9113269B2 (en) | Audio processing device, audio processing method, audio processing program and audio processing integrated circuit | |
| CN112287771A (en) | Method, apparatus, server and medium for detecting video event |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170201 |
|
| RJ01 | Rejection of invention patent application after publication |