CN105721933A - Method for creating advertisement video information base, advertisement video identification method, apparatus for creating advertisement video information base and advertisement video identification apparatus - Google Patents
Method for creating advertisement video information base, advertisement video identification method, apparatus for creating advertisement video information base and advertisement video identification apparatus Download PDFInfo
- Publication number
- CN105721933A CN105721933A CN201610173635.0A CN201610173635A CN105721933A CN 105721933 A CN105721933 A CN 105721933A CN 201610173635 A CN201610173635 A CN 201610173635A CN 105721933 A CN105721933 A CN 105721933A
- Authority
- CN
- China
- Prior art keywords
- value
- eigenvalue
- key
- video
- fingerprint
- 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 239000000284 extract Substances 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 7
- 238000013481 data capture Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 12
- 239000012634 fragment Substances 0.000 abstract 6
- 230000008569 process Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000003542 behavioural effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 229910002056 binary alloy Inorganic materials 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010422 painting 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
- H04N21/43—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
- H04N21/44008—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 involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/845—Structuring of content, e.g. decomposing content into time segments
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/845—Structuring of content, e.g. decomposing content into time segments
- H04N21/8456—Structuring of content, e.g. decomposing content into time segments by decomposing the content in the time domain, e.g. in time segments
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention is applicable to the field of advertisement identification, and provides a method for creating an advertisement video information base, an advertisement video identification method, an apparatus for creating the advertisement video information base and an advertisement video identification apparatus. The method comprises the steps of: acquiring video file fragments, a video file name of a video file in which the video file fragments are positioned, and video head data; reading and retaining video data corresponding to key frames of the video file fragments, and reading a first key frame behind an end position of the video file fragments; according to the video head data and the first key frame, respectively encoding each video file fragment, and respectively extracting a feature value of each video file fragment; determining Key values according to the feature values, wherein each Key value corresponds to one value, and the values are determined by the feature values and the video file name; and storing key value pairs consisting of the Key values and the values. By the method, a speed of processing video data can be improved.
Description
Technical field
The embodiment of the present invention belongs to advertisement identification field, particularly relates to the creation method of a kind of advertisement video information bank, advertisement video recognition methods and device.
Background technology
User behavior data is the useful data that each commercial field carries out client's expansion.
At present, the method obtaining TV user behavioral data is generally: primary advertising video is processed, and creates video image information storehouse according to result, and the process that primary advertising video is carried out is based on the distributed frame of Hadoop.Due in Hadoop distributed file system (HDFS), video data will be become multiple burst (split) according to byte by logical division, this makes continuous print frame be divided to different split, video data in the split of its correspondence cannot individually be processed by therefore corresponding for each split Map task, thus causing that video data processing speed is excessively slow, and then it is slow to cause that TV user behavioral data obtained.
Summary of the invention
Embodiments provide the creation method of a kind of advertisement video information bank, advertisement video recognition methods and device, it is intended to solve the problem that the speed of existing method process video data is excessively slow.
The embodiment of the present invention is achieved in that the creation method of a kind of advertisement video information bank, and described method includes:
Video file burst is obtained from Hadoop distributed file system；
Obtain video file name and the video heads data at video file burst place；
Read and retain the video data that the key frame of video file burst is corresponding, and reading first key frame after the end position of described video file burst；
Decode each video file burst respectively according to described video heads data and described first key frame, and extract the eigenvalue of each video file burst respectively；
Determining at least one Key value according at least one eigenvalue described, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name；
The key-value pair that storage is made up of described Key value and value value.
The another object of the embodiment of the present invention is in that to provide a kind of advertisement video recognition methods, and described method includes:
The television channel that monitoring is being play, intercepts a frame video image of described television channel；
Obtain the eigenvalue of described video image；
Determining a Key value according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined；
The Key value of the Key value determined with the advertisement video information bank being pre-created is compared, searches the value value that the Key value identical with the described Key value determined is corresponding；
Whether the value value corresponding according to the described Key value determined and the value value described video image of identification corresponding to the identical Key value of the described and described Key value determined are advertisement video.
The another object of the embodiment of the present invention is in that to provide the establishment device of a kind of advertisement video information bank, and described device includes:
Video file burst acquiring unit, for obtaining video file burst from Hadoop distributed file system；
Video heads data capture unit, for obtaining video file name and the video heads data at video file burst place；
First key frame reads unit, for reading and retaining the video data that the key frame of video file burst is corresponding, and reads first key frame after the end position of described video file burst；
Characteristics extraction unit, for decoding each video file burst respectively according to described video heads data and described first key frame, and extracts the eigenvalue of each video file burst respectively；
Key value determines unit, and for determining Key value according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name；
Key-value pair memory element, for storing the key-value pair being made up of described Key value and value value.
The another object of the embodiment of the present invention is in that to provide a kind of advertisement video identification device, and described device includes:
Video image interception unit, for monitoring the television channel play, intercepts a frame video image of described television channel；
Eigenvalue acquiring unit, for obtaining the eigenvalue of described video image；
Key-value pair determines unit, and for determining a Key value according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined；
Value value searches unit, for the Key value of the Key value determined with the advertisement video information bank being pre-created being compared, searches the value value that the Key value identical with the described Key value determined is corresponding；
Advertisement video recognition unit, whether the value value identification described video image corresponding for Key value that the value value corresponding according to the described Key value determined and the described and described Key value determined are identical is advertisement video.
In embodiments of the present invention, owing to have read first key frame after the end position of video file burst so that each video file burst is interrelated, therefore, respectively each video file burst can be processed, thus improve the processing speed of video data.
Accompanying drawing explanation
Fig. 1 is the flow chart of the creation method of a kind of advertisement video information bank that first embodiment of the invention provides；
Fig. 2 is the video heads data transmission schematic diagram that first embodiment of the invention provides；
Fig. 3 is the flow chart of a kind of advertisement video recognition methods that second embodiment of the invention provides；
Fig. 4 is the structure chart of a kind of advertisement video identification device that third embodiment of the invention provides；
Fig. 5 is the structure chart creating device of a kind of advertisement video information bank that fourth embodiment of the invention provides.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein is only in order to explain the present invention, is not intended to limit the present invention.
In the embodiment of the present invention, video file burst is obtained from Hadoop distributed file system, video heads data according to the video file burst obtained, read and retain the video data that the key frame of video file burst is corresponding, and read first key frame after the end position of described video file burst, according to described video heads data, and described first key frame decodes each video file burst respectively, and extract the eigenvalue of each video file burst respectively, Key value is determined further according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name, the key-value pair that storage is made up of described Key value and value value.
In order to technical solutions according to the invention are described, illustrate below by specific embodiment.
Embodiment one:
Fig. 1 illustrates the creation method flow chart of a kind of advertisement video information bank that first embodiment of the invention provides, and details are as follows:
Step S11, obtains video file burst from Hadoop distributed file system.
In this step, utilize Fuse-dfs instrument, the video file burst (split) in Hadoop distributed file system (HDFS) is mapped to local file system (LocalFS such as Fig. 2).Here video file burst is the video file burst that primary advertising video file is corresponding.
Step S12, obtains video file name and the video heads data at video file burst place.
Utilize Fuse-dfs instrument, for each split initializing decoder (Map), and using the video file name at split place as input, the function that each decoder is corresponding reads in video heads data from the first of video file split (video heads data location), namely video heads data transmit portion to each decoding Map task node, as shown in Figure 2.
Step S13, reads and retains the video data that the key frame of video file burst is corresponding, and reading first key frame after the end position of described video file burst.
Wherein, key frame is equivalent to the original painting in 2 D animation, refers to that frame residing for key operations in role or object of which movement or change, and the key frame of video file is generally I frame.
In this step, the size of split block is identical with predetermined BlocK size, the logical division position of split defines according to BlocK start-stop byte location, when the video data of split is read in decoding, the non-key frame data read in split original position being abandoned, just starting to retain the video data read until reading first key frame；When reading the end position of split, continue to read, until first key frame after reading split end position.
Alternatively, described reading also retains the video data that the key frame of video file burst is corresponding, and read first key frame after the end position of described video file burst, specifically includes:
A1, read video file burst original position frame, it is judged that whether the frame of described video file burst original position is key frame.
A2, when the frame of described video file burst original position is key frame, retain the video data that described key frame is corresponding, when the frame of described video file burst original position is not key frame, ignore the video data that described key frame is corresponding.
A3, continue to read the next frame of video file burst, and judge whether the next frame of video file burst read is key frame, until reading first key frame after the end position of video file burst just terminate reading.
In above-mentioned A1～A3, owing to have read first key frame after the end position of video file burst, remain the information of subsequent frame, therefore, it is simple to follow-up individually the video data in each split is processed, improve processing speed.
It is pointed out that the frame of the original position due to video file burst (split) is generally non-key frame, in order to accelerate reading speed, following read method can be adopted:
A. the non-key frame of split original position is ignored；
B. the next frame of split is read；
C. determine whether I frame (key frame), if it is not, then forward step b to, if I frame, then enter step d；
D. this frame is mapped to a Map node (a picture group is mapped to a Map node)；
E. continue to read next frame；
F. judging whether it is first I frame after split end position, if then terminating to read, otherwise going to step d.
Step S14, decodes each video file burst respectively according to described video heads data and described first key frame, and extracts the eigenvalue of each video file burst respectively.
In this step, owing to remaining the information of subsequent frame, therefore, each video file burst can be decoded respectively, and extracting the eigenvalue of each video file burst respectively, eigenvalue here includes condition code (condition codes such as 160 bytes) and the timestamp of video image (referring here to the timestamp that a frame video image is corresponding).
Step S15, determines Key value according to described eigenvalue, and the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name.
Alternatively, described determine Key value according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name, specifically includes:
B1, calculating the fingerprint of described eigenvalue, described fingerprint is the binary code more than 1.Specifically, its SimHash code of eigenvalue calculation to all or part of split extracted, this SimHash code is as the fingerprint of eigenvalue, and it is a binary code more than 1, is generally the binary code of 64.
B2, using at least one binary code of the fingerprint of described eigenvalue as Key value, the corresponding value value of each Key value, the timestamp of the video image that described value value includes the fingerprint of described eigenvalue, described eigenvalue includes and video file name.Wherein, eigenvalue includes the timestamp of video image, and value value corresponding to Key value includes the video file filename that the fingerprint of eigenvalue, the timestamp of video image and this video image are corresponding.In this step, to obtaining a Key value after single eigenvalue calculation, for instance, if 10 eigenvalues are calculated, then obtain 10 Key.
Further, described B2 specifically includes:
B21, the fingerprint of described eigenvalue splitting into the binary code of 4 16, the fingerprint of described eigenvalue is the binary code of 64.Specifically, when eigenvalue has 4, and during binary code that the fingerprint of eigenvalue is 64, the fingerprint of eigenvalue is on average split in order the binary code of 4 16.
B22, respectively to split the binary code of obtain 4 16 as 4 Key, each Key value one value value of correspondence, the timestamp of the video image that described value value includes the fingerprint of described eigenvalue, described eigenvalue includes and video file name.
Step S16, the key-value pair that storage is made up of described Key value and value value.
In first embodiment of the invention, video file burst is obtained from Hadoop distributed file system, video heads data according to the video file burst obtained, read and retain the video data that the key frame of video file burst is corresponding, and read first key frame after the end position of described video file burst, according to described video heads data, and described first key frame decodes each video file burst respectively, and extract the eigenvalue of each video file burst respectively, Key value is determined further according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name, the key-value pair that storage is made up of described Key value and value value.Owing to have read first key frame after the end position of video file burst so that each video file burst is interrelated, therefore, it is possible to respectively each video file burst is processed, thus improve the processing speed of video data.
Should be understood that in embodiments of the present invention, the size of the sequence number of above-mentioned each process is not meant to the priority of execution sequence, and the execution sequence of each process should be determined with its function and internal logic, and the implementation process of the embodiment of the present invention should not constituted any restriction.
Embodiment two:
Fig. 3 illustrates the flow chart of a kind of advertisement video recognition methods that second embodiment of the invention provides, and in the present embodiment, make use of the advertisement video information bank that embodiment one provides to carry out advertisement video identification, and details are as follows:
Step S31, the television channel that monitoring is being play, intercept a frame video image of described television channel.
In this step, the television channel that monitor in real time user is watching, and intercept a frame video image at set intervals, and such as intercepting a frame video image etc. every 5 seconds or 10 seconds, concrete interval time can adjust as desired by backstage.
Step S32, obtains the eigenvalue of described video image.
Here eigenvalue includes condition code (condition codes such as 160 bytes) and the timestamp of video image (referring here to the timestamp that a frame video image is corresponding).
Step S33, determines a Key value according to described eigenvalue, and the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined.
Alternatively, described determine a Key value according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined, specifically includes:
C1, calculating the fingerprint of described eigenvalue, described fingerprint is the binary code more than 1.Specifically, its SimHash code of eigenvalue calculation to each split extracted, this SimHash code is as the fingerprint of eigenvalue, and it is a binary code more than 1, is generally the binary code of 64.
C2, using at least one binary code of the fingerprint of described eigenvalue as Key value, the corresponding value value of described Key value, timestamp and the video file name of the video image that described value value is included by the fingerprint of described eigenvalue, described eigenvalue form.Specifically, generally binary code corresponding for fingerprint is split as in order multiple binary code, and using the Part I binary code that splits as Key value.Wherein, eigenvalue includes the timestamp of video image, and value value corresponding to Key value is made up of the video file filename that the fingerprint of eigenvalue, the timestamp of video image and this video image are corresponding.
Step S34, it is to be determined to Key value compare with the Key value of the advertisement video information bank being pre-created, search the value value that the Key value identical with the described Key value determined is corresponding.
Wherein, each video file burst can be processed by advertisement video information bank respectively that be pre-created.
Specifically, it is to be determined to Key value compare with the Key value of the advertisement video information bank being pre-created, if finding identical Key value in the advertisement video information bank being pre-created, then extract the value value that the Key value identical with the described Key value determined is corresponding.If not finding identical Key value in the advertisement video information bank being pre-created, then after redefining a Key value, then perform step S34.Such as, assume the binary code that fingerprint is 64 of eigenvalue, this 64 binary code being is split as 4 16 ary codes in order, and using split gained first 16 ary codes as Key value, if not finding the Key value identical with first Key value in the advertisement video information bank being pre-created, then using split gained second 16 ary codes as second Key value, again this second Key value is compared with the Key value of the advertisement video information bank being pre-created, if also not finding, then continue to determine the 3rd Key value, until 4 Key values are all inquired about complete.
According to the value value that the Key value that the value value of the described Key value determined correspondence and the described and described Key value determined are identical is corresponding, step S35, identifies whether described video image is advertisement video.
Specifically, whether the described value value corresponding according to the described Key value determined and the value value described video image of identification corresponding to the identical Key value of the described and described Key value determined are advertisement video, specifically include:
D1, the parsing value value according to the described Key value determined correspondence, obtain the fingerprint of the First Eigenvalue.Owing to value value includes the fingerprint of eigenvalue, therefore resolve value value time, it is possible to obtain the fingerprint of an eigenvalue, in order to follow-up differentiation, the fingerprint of the eigenvalue now obtained is defined as the fingerprint of the First Eigenvalue.
D2, resolve the value value that the Key value identical with the described Key value determined is corresponding, obtain the fingerprint of Second Eigenvalue.Owing to value value includes the fingerprint of eigenvalue, when therefore resolving value value, it is possible to obtaining the fingerprint of an eigenvalue, in order to the fingerprint with the First Eigenvalue above is distinguished, the fingerprint of the eigenvalue now obtained is defined as the fingerprint of Second Eigenvalue.
D3, calculate the fingerprint of described the First Eigenvalue and the similarity of the fingerprint of described Second Eigenvalue.
D4, when the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is less than or equal to default threshold value, identify that described video image is advertisement video, when the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is more than default threshold value, identify that described video image is for advertisement video.When the similarity of the fingerprint of two eigenvalues meets default threshold value, it is determined that the video image of intercepting is advertisement video, otherwise, it is determined that the video image intercepted is not advertisement video.
Further, when judging the video image intercepted as advertisement video, resolve value value to obtain timestamp and the video file name of video image.
Further, when the similarity of the fingerprint of the described fingerprint at described the First Eigenvalue and described Second Eigenvalue is more than default threshold value, identify that described video image is not for advertisement video, specifically includes:
When the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is more than default threshold value, continue to read and (assume the binary code that fingerprint is 64 of eigenvalue, this 64 binary code being is split as 4 16 ary codes in order, and using split gained first 16 ary codes as Key value) next 16 ary codes are as Key value, perform step S34 and step S35 again, circulation performs, until have read 4 16 ary codes as Key value.If the similarity of the fingerprint of the eigenvalue obtained according to 4 Key values is all unsatisfactory for the threshold value preset, identify that described video image is for advertisement video.
In second embodiment of the invention, the television channel that monitoring is being play, intercept a frame video image of described television channel, obtain the eigenvalue of described video image, a Key value is determined according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined, the Key value of the Key value determined with the advertisement video information bank being pre-created is compared, search the value value that the Key value identical with the described Key value determined is corresponding, whether the value value corresponding according to the described Key value determined and the value value described video image of identification corresponding to the identical Key value of the described and described Key value determined are advertisement video.Each video file burst is processed by advertisement video information bank owing to being pre-created respectively, this improves the processing speed of video data, and then improves the recognition speed of advertisement video.
Embodiment three:
Fig. 4 illustrates the structure chart of a kind of advertisement video identification device that third embodiment of the invention provides, this advertisement video identification device can apply in various intelligent terminal, this intelligent terminal includes through the wireless access network RAN subscriber equipment communicated with one or more core net, this subscriber equipment can be mobile phone (or being called " honeycomb " phone), there is the computer etc. of mobile equipment, such as, subscriber equipment can also is that portable, pocket, hand-held, built-in computer or vehicle-mounted mobile device, they are with wireless access network switched voice and/or data.Again such as, this moves equipment and can include smart mobile phone, panel computer, personal digital assistant PDA, point-of-sale terminal POS or vehicle-mounted computer etc..For the ease of illustrating, illustrate only the part relevant to the embodiment of the present invention.
This advertisement video identification device includes: video image interception unit 41, eigenvalue acquiring unit 42, key-value pair determine that unit 43, value value search unit 44, advertisement video recognition unit 45.Wherein:
Video image interception unit 41, for monitoring the television channel play, intercepts a frame video image of described television channel.
Specifically, the television channel that monitor in real time user is watching, and intercept a frame video image at set intervals, and such as intercepting a frame video image etc. every 5 seconds or 10 seconds, concrete interval time can adjust as desired by backstage.
Eigenvalue acquiring unit 42, for obtaining the eigenvalue of described video image.
Here eigenvalue includes condition code (condition codes such as 160 bytes) and the timestamp of video image (referring here to the timestamp that a frame video image is corresponding).
Key-value pair determines unit 43, and for determining a Key value according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined.
Wherein, described key-value pair determines that unit 43 includes:
Fingerprint computing module, for calculating the fingerprint of described eigenvalue, described fingerprint is the binary code more than 1.Specifically, its SimHash code of eigenvalue calculation to each split extracted, this SimHash code is as the fingerprint of eigenvalue, and it is a binary code more than 1, is generally the binary code of 64.
Fingerprint splits module, for using at least one binary code of the fingerprint of described eigenvalue as Key value, the corresponding value value of described Key value, timestamp and the video file name of the video image that described value value is included by the fingerprint of described eigenvalue, described eigenvalue form.Specifically, generally binary code corresponding for fingerprint is split as in order multiple binary code, and using the Part I binary code that splits as Key value.
Value value searches unit 44, for the Key value of the Key value determined with the advertisement video information bank being pre-created being compared, searches the value value that the Key value identical with the described Key value determined is corresponding.
Wherein, each video file burst can be processed by advertisement video information bank respectively that be pre-created.
Specifically, it is to be determined to Key value compare with the Key value of the advertisement video information bank being pre-created, if not finding identical Key value in the advertisement video information bank being pre-created, then after redefining a Key value, then perform value value search unit 44.Such as, assume the binary code that fingerprint is 64 of eigenvalue, this 64 binary code being is split as 4 16 ary codes in order, and using split gained first 16 ary codes as Key value, if not finding the Key value identical with first Key value in the advertisement video information bank being pre-created, then using split gained second 16 ary codes as second Key value, again this second Key value is compared with the Key value of the advertisement video information bank being pre-created, if also not finding, then continue to determine the 3rd Key value, until 4 Key values are all inquired about complete.
Advertisement video recognition unit 45, whether the value value identification described video image corresponding for Key value that the value value corresponding according to the described Key value determined and the described and described Key value determined are identical is advertisement video.
Wherein, described advertisement video recognition unit 45 includes:
The fingerprint determination module of the First Eigenvalue, for resolving the value value corresponding according to the described Key value determined, obtains the fingerprint of the First Eigenvalue.
The fingerprint determination module of Second Eigenvalue, for resolving the value value that the Key value identical with the described Key value determined is corresponding, obtains the fingerprint of Second Eigenvalue.
Fingerprint similarity calculation module, is used for calculating the similarity of the fingerprint of the fingerprint of described the First Eigenvalue and described Second Eigenvalue.
Advertisement video determination module, during for the similarity of the fingerprint at the fingerprint of described the First Eigenvalue and described Second Eigenvalue less than or equal to default threshold value, identify that described video image is advertisement video, when the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is more than default threshold value, identify that described video image is for advertisement video.
Further, when judging the video image intercepted as advertisement video, resolve value value to obtain timestamp and the video file name of video image.
Further, when the similarity of the fingerprint of the described fingerprint at described the First Eigenvalue and described Second Eigenvalue is more than default threshold value, identify that described video image is not for advertisement video, specifically includes:
When the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is more than default threshold value, continue to read and (assume the binary code that fingerprint is 64 of eigenvalue, this 64 binary code being is split as 4 16 ary codes in order, and using split gained first 16 ary codes as Key value) next 16 ary codes are as Key value, perform value value again and search unit 44 and advertisement video recognition unit 45, circulation performs, until have read 4 16 ary codes as Key value.If the similarity of the fingerprint of the eigenvalue obtained according to 4 Key values is all unsatisfactory for the threshold value preset, identify that described video image is for advertisement video.
In third embodiment of the invention, owing to each video file burst is processed by advertisement video information bank respectively that be pre-created, this improves the processing speed of video data, and then improve the recognition speed of advertisement video.
Embodiment four:
Fig. 5 illustrates the structure chart creating device of a kind of advertisement video information bank that fourth embodiment of the invention provides, and for the ease of illustrating, illustrate only part related to the present embodiment.
The establishment device of this advertisement video information bank includes: video file burst acquiring unit 51, video heads data capture unit 52, first key frame reading unit 53, characteristics extraction unit 54, Key value determine unit 55, key-value pair memory element 56.Wherein:
Video file burst acquiring unit 51, for obtaining video file burst from Hadoop distributed file system.
Video heads data capture unit 52, for obtaining video file name and the video heads data at video file burst place.
First key frame reads unit 53, for reading and retaining the video data that the key frame of video file burst is corresponding, and reads first key frame after the end position of described video file burst.
Alternatively, described first key frame reading unit 53 includes:
Key frame judge module, for reading the frame of video file burst original position, it is judged that whether the frame of described video file burst original position is key frame.
The processing module of different frame, for when the frame of described video file burst original position is key frame, retain the video data that described key frame is corresponding, when the frame of described video file burst original position is not key frame, ignore the video data that described key frame is corresponding.
Reading and terminate judge module, for continuing to read the next frame of video file burst, and judging whether the next frame of video file burst read is key frame, until reading first key frame after the end position of video file burst just terminate reading.
Above-mentioned key frame judge module, the processing module of different frame, reading terminate in judge module, owing to have read first key frame after the end position of video file burst, remain the information of subsequent frame, therefore, it is easy to follow-up individually the video data in each split be processed, improves processing speed.
Characteristics extraction unit 54, for decoding each video file burst respectively according to described video heads data and described first key frame, and extracts the eigenvalue of each video file burst respectively.
Key value determines unit 55, and for determining Key value according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name.
Alternatively, described Key value determines that unit 55 includes:
Fingerprint computing module, for calculating the fingerprint of described eigenvalue, described fingerprint is the binary code more than 1.Specifically, its SimHash code of eigenvalue calculation to all or part of split extracted, this SimHash code is as the fingerprint of eigenvalue, and it is a binary code more than 1, is generally the binary code of 64.
Fingerprint and Key value modular converter, for using at least one binary code of the fingerprint of described eigenvalue as Key value, the corresponding value value of each Key value, the timestamp of the video image that described value value includes the fingerprint of described eigenvalue, described eigenvalue includes and video file name.
Further, described fingerprint and Key value modular converter include:
Fingerprint splits module, and for the fingerprint of described eigenvalue splits into the binary code of 4 16, the fingerprint of described eigenvalue is the binary code of 64.Specifically, when eigenvalue has 4, and during binary code that the fingerprint of eigenvalue is 64, the fingerprint of eigenvalue is on average split in order the binary code of 4 16.
Binary system and Key value respective modules, for respectively to split the binary code of obtain 4 16 as 4 Key, the corresponding value value of each Key value, the timestamp of the video image that described value value includes the fingerprint of described eigenvalue, described eigenvalue includes and video file name.
Key-value pair memory element 56, for storing the key-value pair being made up of described Key value and value value.
In fourth embodiment of the invention, owing to have read first key frame after the end position of video file burst so that each video file burst is interrelated, therefore, respectively each video file burst can be processed, thus improve the processing speed of video data.
Those of ordinary skill in the art are it is to be appreciated that the unit of each example that describes in conjunction with the embodiments described herein and algorithm steps, it is possible to being implemented in combination in of electronic hardware or computer software and electronic hardware.These functions perform with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme.Professional and technical personnel specifically can should be used for using different methods to realize described function to each, but this realization is it is not considered that beyond the scope of this invention.
Those skilled in the art is it can be understood that arrive, for convenience and simplicity of description, and the specific works process of the system of foregoing description, device and unit, it is possible to reference to the corresponding process in preceding method embodiment, do not repeat them here.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method, it is possible to realize by another way.Such as, device embodiment described above is merely schematic, such as, the division of described unit, being only a kind of logic function to divide, actual can have other dividing mode when realizing, for instance multiple unit or assembly can in conjunction with or be desirably integrated into another system, or some features can ignore, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be through INDIRECT COUPLING or the communication connection of some interfaces, device or unit, it is possible to be electrical, machinery or other form.
The described unit illustrated as separating component can be or may not be physically separate, and the parts shown as unit can be or may not be physical location, namely may be located at a place, or can also be distributed on multiple NE.Some or all of unit therein can be selected according to the actual needs to realize the purpose of the present embodiment scheme.
It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to be that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.
If described function is using the form realization of SFU software functional unit and as independent production marketing or use, it is possible to be stored in a computer read/write memory medium.Based on such understanding, part or the part of this technical scheme that prior art is contributed by technical scheme substantially in other words can embody with the form of software product, this computer software product is stored in a storage medium, including some instructions with so that a computer equipment (can be personal computer, server, or the network equipment etc.) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (ROM, Read-OnlyMemory), the various media that can store program code such as random access memory (RAM, RandomAccessMemory), magnetic disc or CD.
The above; being only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, any those familiar with the art is in the technical scope that the invention discloses; change can be readily occurred in or replace, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should described be as the criterion with scope of the claims.
Claims (11)
1. the creation method of an advertisement video information bank, it is characterised in that described method includes:
Video file burst is obtained from Hadoop distributed file system；
Obtain video file name and the video heads data at video file burst place；
Read and retain the video data that the key frame of video file burst is corresponding, and reading first key frame after the end position of described video file burst；
Decode each video file burst respectively according to described video heads data and described first key frame, and extract the eigenvalue of each video file burst respectively；
Determining Key value according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name；
The key-value pair that storage is made up of described Key value and value value.
2. method according to claim 1, it is characterised in that described reading also retains the video data that the key frame of video file burst is corresponding, and read first key frame after the end position of described video file burst, specifically includes:
Read the frame of video file burst original position, it is judged that whether the frame of described video file burst original position is key frame；
When the frame of described video file burst original position is key frame, retain the video data that described key frame is corresponding, when the frame of described video file burst original position is not key frame, ignore the video data that described key frame is corresponding；
Continue to read the next frame of video file burst, and judge whether the next frame of video file burst read is key frame, until reading first key frame after the end position of video file burst just terminate reading.
3. method according to claim 1, it is characterised in that described determine Key value according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name, specifically includes:
Calculating the fingerprint of described eigenvalue, described fingerprint is the binary code more than 1；
Using at least one binary code of the fingerprint of described eigenvalue as Key value, the corresponding value value of each Key value, described value value includes the fingerprint of described eigenvalue, described eigenvalue includes timestamp and the video file name of video image.
4. method according to claim 3, it is characterized in that, at least one binary code of the described fingerprint using described eigenvalue is as Key value, the corresponding value value of each Key value, the timestamp of the video image that described value value includes the fingerprint of described eigenvalue, described eigenvalue includes and video file name, specifically include:
The fingerprint of described eigenvalue splits into the binary code of 4 16, and the fingerprint of described eigenvalue is the binary code of 64；
Respectively to split the binary code of obtain 4 16 as 4 Key, each Key value one value value of correspondence, the timestamp of the video image that described value value includes the fingerprint of described eigenvalue, described eigenvalue includes and video file name.
5. an advertisement video recognition methods, it is characterised in that described method includes:
The television channel that monitoring is being play, intercepts a frame video image of described television channel；
Obtain the eigenvalue of described video image；
Determining a Key value according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined；
The Key value of the Key value determined with the advertisement video information bank being pre-created is compared, searches the value value that the Key value identical with the described Key value determined is corresponding；
Whether the value value corresponding according to the described Key value determined and the value value described video image of identification corresponding to the identical Key value of the described and described Key value determined are advertisement video.
6. method according to claim 5, it is characterised in that described determine a Key value according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined, specifically includes:
Calculating the fingerprint of described eigenvalue, described fingerprint is the binary code more than 1；
Using at least one binary code of the fingerprint of described eigenvalue as Key value, the corresponding value value of described Key value, timestamp and the video file name of the video image that described value value is included by the fingerprint of described eigenvalue, described eigenvalue form.
7. method according to claim 6, it is characterized in that, whether the described value value corresponding according to the described Key value determined and the value value described video image of identification corresponding to the identical Key value of the described and described Key value determined are advertisement video, specifically include:
Resolve the value value corresponding according to the described Key value determined, obtain the fingerprint of the First Eigenvalue；
Resolve the value value that the Key value identical with the described Key value determined is corresponding, obtain the fingerprint of Second Eigenvalue；
Calculate the fingerprint of described the First Eigenvalue and the similarity of the fingerprint of described Second Eigenvalue；
When the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is less than or equal to default threshold value, identify that described video image is advertisement video, when the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is more than default threshold value, identify that described video image is for advertisement video.
8. the establishment device of an advertisement video information bank, it is characterised in that described device includes:
Video file burst acquiring unit, for obtaining video file burst from Hadoop distributed file system；
Video heads data capture unit, for obtaining video file name and the video heads data at video file burst place；
First key frame reads unit, for reading and retaining the video data that the key frame of video file burst is corresponding, and reads first key frame after the end position of described video file burst；
Characteristics extraction unit, for decoding each video file burst respectively according to described video heads data and described first key frame, and extracts the eigenvalue of each video file burst respectively；
Key value determines unit, and for determining Key value according to described eigenvalue, the corresponding value value of each Key value, described value value is determined by described eigenvalue and video file name；
Key-value pair memory element, for storing the key-value pair being made up of described Key value and value value.
9. an advertisement video identification device, it is characterised in that described device includes:
Video image interception unit, for monitoring the television channel play, intercepts a frame video image of described television channel；
Eigenvalue acquiring unit, for obtaining the eigenvalue of described video image；
Key-value pair determines unit, and for determining a Key value according to described eigenvalue, the corresponding value value of described Key value, described value value video file name belonging to described eigenvalue and video image is determined；
Value value searches unit, for the Key value of the Key value determined with the advertisement video information bank being pre-created being compared, searches the value value that the Key value identical with the described Key value determined is corresponding；
Advertisement video recognition unit, whether the value value identification described video image corresponding for Key value that the value value corresponding according to the described Key value determined and the described and described Key value determined are identical is advertisement video.
10. device according to claim 9, it is characterised in that described key-value pair determines that unit includes:
Fingerprint computing module, for calculating the fingerprint of described eigenvalue, described fingerprint is the binary code more than 1；
Fingerprint splits module, for using at least one binary code of the fingerprint of described eigenvalue as Key value, the corresponding value value of described Key value, timestamp and the video file name of the video image that described value value is included by the fingerprint of described eigenvalue, described eigenvalue form.
11. device according to claim 10, it is characterised in that described advertisement video recognition unit includes:
The fingerprint determination module of the First Eigenvalue, for resolving the value value corresponding according to the described Key value determined, obtains the fingerprint of the First Eigenvalue；
The fingerprint determination module of Second Eigenvalue, for resolving the value value that the Key value identical with the described Key value determined is corresponding, obtains the fingerprint of Second Eigenvalue；
Fingerprint similarity calculation module, is used for calculating the similarity of the fingerprint of the fingerprint of described the First Eigenvalue and described Second Eigenvalue；
Advertisement video determination module, during for the similarity of the fingerprint at the fingerprint of described the First Eigenvalue and described Second Eigenvalue less than or equal to default threshold value, identify that described video image is advertisement video, when the similarity of the fingerprint of described the First Eigenvalue and the fingerprint of described Second Eigenvalue is more than default threshold value, identify that described video image is for advertisement video.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610173635.0A CN105721933A (en) | 2016-03-23 | 2016-03-23 | Method for creating advertisement video information base, advertisement video identification method, apparatus for creating advertisement video information base and advertisement video identification apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610173635.0A CN105721933A (en) | 2016-03-23 | 2016-03-23 | Method for creating advertisement video information base, advertisement video identification method, apparatus for creating advertisement video information base and advertisement video identification apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105721933A true CN105721933A (en) | 2016-06-29 |
Family
ID=56159005
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610173635.0A Pending CN105721933A (en) | 2016-03-23 | 2016-03-23 | Method for creating advertisement video information base, advertisement video identification method, apparatus for creating advertisement video information base and advertisement video identification apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105721933A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109547864A (en) * | 2017-09-21 | 2019-03-29 | 腾讯科技（深圳）有限公司 | Media data processing method and device |
| CN110276040A (en) * | 2019-06-28 | 2019-09-24 | 北京金山安全软件有限公司 | Picture file processing method, picture displaying method and picture displaying device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807208A (en) * | 2010-03-26 | 2010-08-18 | 上海全土豆网络科技有限公司 | Method for quickly retrieving video fingerprints |
| CN103279521A (en) * | 2013-05-28 | 2013-09-04 | 重庆大学 | Video big data distributed decoding method based on Hadoop |
| CN103310460A (en) * | 2013-06-24 | 2013-09-18 | 安科智慧城市技术（中国）有限公司 | Image characteristic extraction method and system |
| US20140161362A1 (en) * | 2012-12-10 | 2014-06-12 | International Business Machines Corporation | Techniques for Spatial Semantic Attribute Matching for Location Identification |
| CN104394415A (en) * | 2014-12-09 | 2015-03-04 | 中国电子科技集团公司第二十八研究所 | Method for distributed decoding of video big data |
| CN105095435A (en) * | 2015-07-23 | 2015-11-25 | 北京京东尚科信息技术有限公司 | Similarity comparison method and device for high-dimensional image features |
-
2016
- 2016-03-23 CN CN201610173635.0A patent/CN105721933A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807208A (en) * | 2010-03-26 | 2010-08-18 | 上海全土豆网络科技有限公司 | Method for quickly retrieving video fingerprints |
| US20140161362A1 (en) * | 2012-12-10 | 2014-06-12 | International Business Machines Corporation | Techniques for Spatial Semantic Attribute Matching for Location Identification |
| CN103279521A (en) * | 2013-05-28 | 2013-09-04 | 重庆大学 | Video big data distributed decoding method based on Hadoop |
| CN103310460A (en) * | 2013-06-24 | 2013-09-18 | 安科智慧城市技术（中国）有限公司 | Image characteristic extraction method and system |
| CN104394415A (en) * | 2014-12-09 | 2015-03-04 | 中国电子科技集团公司第二十八研究所 | Method for distributed decoding of video big data |
| CN105095435A (en) * | 2015-07-23 | 2015-11-25 | 北京京东尚科信息技术有限公司 | Similarity comparison method and device for high-dimensional image features |
Non-Patent Citations (2)
| Title |
|---|
| GEMANTIC: "simhash简介及java实现", 《百度》 * |
| 于萧榕等: "监控系统预警视频的分布式检索设计与实现", 《计算机测量与控制》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109547864A (en) * | 2017-09-21 | 2019-03-29 | 腾讯科技（深圳）有限公司 | Media data processing method and device |
| CN109547864B (en) * | 2017-09-21 | 2021-05-07 | 腾讯科技（深圳）有限公司 | Media data processing method and device |
| CN110276040A (en) * | 2019-06-28 | 2019-09-24 | 北京金山安全软件有限公司 | Picture file processing method, picture displaying method and picture displaying device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104735468B (en) | A kind of method and system that image is synthesized to new video based on semantic analysis | |
| CN107911249B (en) | Method, device and equipment for sending command line of network equipment | |
| CN105159913B (en) | Method and device for determining file to be cleaned | |
| CN104504109A (en) | Image search method and device | |
| CN110147469B (en) | Data processing method, device and storage medium | |
| US10614312B2 (en) | Method and apparatus for determining signature actor and identifying video based on probability of appearance of signature actor | |
| CN110198482B (en) | Video key bridge segment marking method, terminal and storage medium | |
| CN109885828A (en) | Word error correction method, device, computer equipment and medium based on language model | |
| CN102148805A (en) | Feature matching method and device | |
| CN110502664A (en) | Video tab indexes base establishing method, video tab generation method and device | |
| CN105718920A (en) | Fingerprint identification method and fingerprint identification device | |
| CN103077163B (en) | Data preprocessing method, device and system | |
| CN109766422A (en) | Information processing method, apparatus and system, storage medium, terminal | |
| CN111428087B (en) | Video interception method and device, computer equipment and storage medium | |
| CN111488813B (en) | Video emotion marking method and device, electronic equipment and storage medium | |
| CN105721933A (en) | Method for creating advertisement video information base, advertisement video identification method, apparatus for creating advertisement video information base and advertisement video identification apparatus | |
| CN105045882A (en) | Hot word processing method and device | |
| AU2020104435A4 (en) | Method and apparatus for video recommendation, and refrigerator with screen | |
| CN106937127B (en) | Display method and system for intelligent search preparation | |
| CN113254712A (en) | Video matching method, video processing device, electronic equipment and medium | |
| CN107798004B (en) | Keyword searching method and device and terminal | |
| CN112954455A (en) | Subtitle tracking method and device and electronic equipment | |
| CN116226681B (en) | Text similarity judging method and device, computer equipment and storage medium | |
| CN109213972B (en) | Method, device, equipment and computer storage medium for determining document similarity | |
| CN107147947A (en) | Key frame recognition methods and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160629 |