CN104519371B - Method for pushing, pusher and server - Google Patents
Method for pushing, pusher and server Download PDFInfo
- Publication number
- CN104519371B CN104519371B CN201410855843.XA CN201410855843A CN104519371B CN 104519371 B CN104519371 B CN 104519371B CN 201410855843 A CN201410855843 A CN 201410855843A CN 104519371 B CN104519371 B CN 104519371B
- Authority
- CN
- China
- Prior art keywords
- video image
- display area
- region
- sliding window
- positional information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/235—Processing of additional data, e.g. scrambling of additional data or processing content descriptors
-
- 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
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
Abstract
The invention discloses a kind of method for pushing, pusher and server, the wherein method for pushing includes：Acquisition module obtain server inner video image in can not overlay area positional information；Computing module according to the displaying size of pushed information and can not overlay area positional information to calculate the positional information of display area, wherein display area with can not overlay area do not overlap；Pushing module sends the positional information of pushed information and display area to client to show pushed information in display area.In the client, it can not be not present due to display area and between overlay area overlapping, therefore come from server pushed information will not with come from TV station additional information and between will not also produce it is overlapping, so that pushed information will not produce interference to user, and then improve the experience sense of user.
Description
Technical field
The present invention relates to technical field of media play, more particularly to a kind of method for pushing, pusher and server.
Background technology
TV station is often embedded in some additional datas in the original multimedia data for programme content to be presented, so that
Must be embedded with the video image corresponding to the multi-medium data of additional data can not only show programme content, and can also be
One or more additional information is shown on programme content, these additional informations can be programme information, TV guide, advertisement
Deng these additional informations can effectively improve the experience sense of user.
With the development of intelligent terminal and Internet technology, intellectuality has been done step-by-step in client, these intelligence
Client (such as intelligent television) can not only receive the multi-medium data of TV station, and can also real-time reception server push
Information.The exhibition of these pushed informations takes the certain display area of display screen, in the prior art, the displaying of pushed information
One piece of FX that region can be typically selected on display screen, for example：The lower left corner or the lower right corner of display screen.
However, the additional information that TV station sends sometimes can also select the lower left corner or the lower right corner in display screen to be shown
Show, the pushed information partly or completely all standing additional information that now server is sent, so as to cause user can not read annex
The content of information, i.e. pushed information cause interference to user.
The content of the invention
The present invention provides a kind of method for pushing, pusher and server, can effectively solve due to pushed information with it is attached
Plus in the presence of overlapping between information, so that the problem of producing interference to user.
To achieve the above object, the invention provides a kind of method for pushing, including：
Obtain in server inner video image can not overlay area positional information, it is described can not overlay area to be covered in
It is used for the region for showing additional information on programme content；
According to the displaying size of pushed information and it is described can not the positional information calculation of overlay area go out the video image
In display area positional information, the display area be covered on programme content be used for show the pushed information
Region, the display area with it is described can not overlay area do not overlap；
The positional information of the pushed information and the display area is sent to client with the display area exhibition
Show the pushed information.
Alternatively, it is described obtain server inner video image in can not overlay area positional information the step of include：
The video image according to the multi-medium data got generates some frames；
Video image described in some frames of generation is carried out Digital Image Processing to identify video image described in some frames
In stagnant zone, the coordinate all put in the stagnant zone constitute it is described can not overlay area positional information.
Alternatively, the Digital Image Processing is specially inter-frame difference processing.
Alternatively, it is described obtain server inner video image in can not overlay area positional information the step of before also wrap
Include：
The sampling of additional information according to each bar with preliminary dimension extracted in advance from some frame video images
Picture block is generated and several one-to-one Image Classifiers of additional information described in each bar, the positive sample of described image grader
To belong to all sampling picture blocks of additional information described in same, the negative sample of described image grader is from some frame videos
The sampling picture block of several programme contents extracted in image, the size of the sampling picture block of the programme content and each bar institute
The size for stating the sampling picture block of additional information is identical, and whole described image graders constitute additional information disaggregated model；
It is described obtain server end inner video image in can not overlay area positional information the step of include：
The video image according to the multi-medium data got generates a frame；
The first sliding window with the preliminary dimension is set up in video image described in a frame in generation, described first slides
Dynamic window is scanned with the first pre- fixed step size to video image described in a frame；
Whether additional information is belonged to the picture judged in the first sliding window according to the additional information disaggregated model, when sentencing
When the picture in the first sliding window that breaks belongs to additional information, obtain in the first sliding window each and put corresponding coordinate, described the
One sliding window complete the whole points got after being scanned to whole video image coordinates constitute it is described can not overlay area position
Confidence ceases.
Alternatively, the displaying size according to pushed information and it is described can not the positional information calculation of overlay area go out institute
Stating the positional information step of the display area in video image includes：
The second sliding window with the displaying size is set up in bottom right edge region in the video image, and described the
Two sliding windows are scanned with the second pre- fixed step size to the bottom right edge region, and the bottom right edge region is and the video
The distance of the lower side of image is less thanRegion and be less than with the distance of the right edge of the video imageRegion, its
Middle H is
Judge region in second sliding window with it is described can not overlay area whether there is overlapping region, when judging
Region in second sliding window with it is described can not overlay area overlapping region is not present when, then assert second sliding window
Interior region is optional display area, and obtains the coordinate all put in the optional display area using as described optional
The positional information of display area, the second sliding window completion is got after being scanned to whole video image can described at least one
Select display area；
One is chosen from whole optional display areas using as the display area, be picked it is described can
The positional information for selecting display area is the positional information of the display area.
Alternatively, the value that m value is 5, n is 5.
Alternatively, it is described the step of one is chosen from whole optional display areas using as the display area
Including：
Calculate the distance between the intermediate point of each optional display area and best illustrated point for pre-setting；
Choose the minimum optional display area of the distance between intermediate point and best illustrated point and be used as the display area.
Alternatively, the distance of the best illustrated point and the lower side of the video image is
The best illustrated point and the distance of the right edge of the video image are
To achieve the above object, the present invention also provides a kind of pusher, including：
Acquisition module, for obtain in server inner video image can not overlay area positional information, it is described to cover
Cover area is to be covered in be used for the region for showing additional information on programme content；
Computing module, for the displaying size according to pushed information and it is described can not the positional information calculation of overlay area go out
The positional information of display area in the video image, the display area is used to show institute to be covered on programme content
State the region of pushed information, the display area with it is described can not overlay area do not overlap；
Pushing module, for the positional information of the pushed information and the display area to be sent to client with institute
State display area and show the pushed information.
Alternatively, the acquisition module includes：
First image generation unit, for the video image according to some frames of multi-medium data generation got；
Digital image processing unit, carries out Digital Image Processing to recognize for video image described in some frames to generation
Can not the area of coverage described in the coordinate composition all put in the stagnant zone gone out in video image described in some frames, the stagnant zone
The positional information in domain.
Alternatively, the pusher also includes：
Grader generation module, according to each bar with preliminary dimension extracted in advance from some frame video images
The sampling picture block generation of additional information and several one-to-one Image Classifiers of additional information described in each bar, described image
The positive sample of grader is all sampling picture blocks for belonging to additional information described in same, the negative sample of described image grader
For the sampling picture block of several programme contents extracted from some frame video images, the sampling picture block of the programme content
Size it is identical with the size of the sampling picture block of additional information described in each bar, whole described image graders constitute additional informations
Disaggregated model；
The acquisition module includes：
Second image generation unit, for the video image according to multi-medium data one frame of generation got；
First slides window unit, for setting up the with the preliminary dimension in video image described in the frame in generation
One sliding window, first sliding window is scanned with the first pre- fixed step size to video image described in a frame；
First judging unit, for whether judging the picture in the first sliding window according to the additional information disaggregated model
Belong to additional information；
First coordinate acquiring unit, for judging that the picture in the first sliding window belongs to additional letter when the first judging unit
During breath, each corresponding coordinate of point in the first sliding window is obtained, first sliding window is completed after being scanned to whole video image
The coordinates of the whole points got constitute it is described can not overlay area positional information.
Alternatively, the computing module includes：
Second slides window unit, and being set up for the bottom right edge region in the video image has the displaying size
The second sliding window, second sliding window is scanned with the second pre- fixed step size to the bottom right edge region, the bottom right
Fringe region is less than for the distance with the lower side of the video imageRegion and with the right edge of the video image
Distance is less thanRegion, wherein, H is the height of video image, and W is the width of video image, and m and n are consolidating of pre-setting
Definite value；
Second judging unit, for judge the region in second sliding window with it is described can not overlay area whether there is
Overlapping region；
Second coordinate acquiring unit, for when the second judging unit judge region in second sliding window with it is described
Can not overlay area when overlapping region is not present, it is an optional display area to assert the region in second sliding window, and
The coordinate all put in the optional display area is obtained to slide using the positional information as the optional display area, described second
Dynamic window completes to get at least one described optional display area after scanning whole video image；
Unit is chosen, one is chosen from whole optional display areas as the display area, to be selected
The positional information of the optional display area gone out is the positional information of the display area.
Alternatively, the value that m value is 5, n is 5.
Alternatively, the selection unit includes：
Computation subunit, for the best illustrated point for calculating the intermediate point of each optional display area Yu pre-setting
The distance between；
Subelement is chosen, the optional display area conduct minimum for choosing the distance between intermediate point and best illustrated point
The display area.
Alternatively, the distance of the best illustrated point and the lower side of the video image is
The best illustrated point and the distance of the right edge of the video image are
To achieve the above object, present invention also offers a kind of server, including：Pusher, the pusher is adopted
With above-mentioned pusher.
The invention has the advantages that：
The invention provides a kind of method for pushing, pusher and server, the wherein method for pushing includes：Acquisition module
Obtain can not overlay area positional information；Computing module according to the displaying size of pushed information and can not overlay area position
Information to calculate the positional information of display area, wherein display area with can not overlay area do not overlap；Pushing module will be pushed away
The positional information for breath and the display area of delivering letters sends to client to show pushed information in display area.In the client, by
In display area and can not be not present between overlay area it is overlapping, therefore come from the pushed information of server will not be with coming from
The additional information of TV station and between will not also produce it is overlapping so that pushed information will not to user produce interference, and then
Improve the experience sense of user.
Brief description of the drawings
Fig. 1 is the structured flowchart for the pusher that the embodiment of the present invention one is provided；
Fig. 2 is the flow chart for the method for pushing that the embodiment of the present invention one is provided；
Fig. 3 is the structured flowchart for the pusher that the embodiment of the present invention two is provided；
Fig. 4 is the schematic diagram in the bottom right edge region of video image in the embodiment of the present invention two；
Fig. 5 is the flow chart for the method for pushing that the embodiment of the present invention two is provided；
Fig. 6 is the structured flowchart for the pusher that the embodiment of the present invention three is provided；
Fig. 7 is the flow chart for the method for pushing that the embodiment of the present invention three is provided.
Embodiment
To make those skilled in the art more fully understand technical scheme, the present invention is carried below in conjunction with the accompanying drawings
Method for pushing, pusher and the server of confession are described in detail.
Embodiment one
Fig. 1 is the structured flowchart for the pusher that the embodiment of the present invention one is provided, as shown in figure 1, the pusher includes：
Acquisition module 1, computing module 2 and pushing module 3, wherein, acquisition module 1 is used to obtain in server inner video image and can not cover
The positional information of cover area, can not overlay area be to be covered on programme content to be used to show the region of additional information；Calculate
Module 2 be used for according to the displaying size of pushed information with can not overlay area the displaying that goes out in video image of positional information calculation
The positional information in region, display area is to be covered on programme content to be used to showing the region of pushed information, display area with
Can not overlay area do not overlap；Pushing module 3 be used for by the positional information of pushed information and display area send to client with
Pushed information is shown in display area.
It should be noted that when TV station sends multi-medium data to client, server can also receive many matchmakers
Volume data, and server can show corresponding video image according to the multi-medium data that receives, now server operation
Business can be according to the video image in server come to the client push information related to the content of the video image.
It should be noted that using the upper left corner of video image as the origin of coordinates in the present invention, with the upper side edge of video image
For transverse axis, and the direction for pointing to the upper right corner by the upper left corner is the positive direction of transverse axis, using the left side of video image as the longitudinal axis, and by
The direction that the lower left corner is pointed in the upper left corner is the positive direction of transverse axis, sets up a coordinate system.
" size " in the present invention refers to the shape and size of corresponding region." positional information " in the present invention refers to pair
Answer the set of the coordinate of the whole points included in region in the coordinate system.
In the present embodiment, it can not be not present due to display area and between overlay area overlapping, therefore come from service
The pushed information of device will not with come from TV station additional information and between will not also produce it is overlapping so that pushed information
Interference will not be produced to user.
The course of work of the pusher shown in Fig. 1 is described in detail below in conjunction with accompanying drawing.Fig. 2 is real for the present invention
The flow chart of the method for pushing of the offer of example one is applied, as shown in Fig. 2 the method for pushing is based on the pusher shown in Fig. 1, the push
Method includes：
Step 101：Obtain server inner video image in can not overlay area positional information.
Step 101 is performed by acquisition module 1.This can not overlay area positional information by can not overlay area included
Whole points coordinates set.
Step 102：According to the displaying size of pushed information and can not the positional information calculation of overlay area go out video image
In display area positional information.
Step 102 is performed by computing module 2.Whole points that the positional information of display area is included by display area
The set of coordinate.Wherein, the display area that calculates and overlapping, the i.e. position of display area can not be not present between overlay area
Information and identical coordinate can not be not present in the positional information of overlay area.
Step 103：The positional information of pushed information and display area is sent to client to push away in display area displaying
Deliver letters breath.
Step 103 is performed by pushing module 3.In client shows video image, additional information and pushed information
It is respectively positioned on program pictures, and additional information is not overlapped with pushed information, so as to not interfere with viewing and admiring for user.
The present invention implements one and provides a kind of method for pushing and pusher, and the wherein method for pushing includes：Acquisition module
Obtain can not overlay area positional information；Computing module according to the displaying size of pushed information and can not overlay area position
Information to calculate the positional information of display area, wherein display area with can not overlay area do not overlap；Pushing module will be pushed away
The positional information for breath and the display area of delivering letters sends to client to show pushed information in display area.In the client, by
In display area and can not be not present between overlay area it is overlapping, therefore come from the pushed information of server will not be with coming from
The additional information of TV station and between will not also produce it is overlapping so that pushed information will not to user produce interference.
Embodiment two
Fig. 3 is the structured flowchart for the pusher that the embodiment of the present invention two is provided, as shown in figure 3, the pusher includes：
Acquisition module 1, computing module 2 and pushing module 3, wherein acquisition module 1 include：First image generation unit 11 and digital picture
Processing unit 12.
Wherein, the first image generation unit 11 is used to obtain generates some frame video figures according to the multi-medium data got
Picture；Digital image processing unit 12 is used to carry out digitized map to some frame video images of 11 pairs of generations of the first image generation unit
As the coordinate composition that processing is all put with identifying in the stagnant zone in some frame video images, stagnant zone can not the area of coverage
The positional information in domain.
In actual applications, because program pictures are dynamic changes, and being additional to the additional information on program pictures is
Fixed static, therefore the stagnant zone correspondence additional information being based in multi-frame video image, video image, so stagnant zone
It is interior all point coordinate constitute can not overlay area positional information.
In the present embodiment, digital image processing unit 12 identifies the quiescent centre in video image by Digital Image Processing
Domain.Alternatively, the Digital Image Processing is specially inter-frame difference processing.Handle to identify in video image by inter-frame difference
Stagnant zone substantially process it is as follows：
Frame-to-frame differences is calculated first:
Wherein, Di(x, y) represents that coordinate is the point of (x, y) between i+1 frame and the i-th frame in the i-th frame video image
Frame-to-frame differences,The red luma component of coordinate in the i-th frame video image for the point of (x, y) is represented,Represent
Coordinate is the Green brightness component of the point of (x, y) in i-th frame video image,Represent that coordinate is in the i-th frame video image
The blue luminences component of the point of (x, y), the n of i values 1,2,3 ....
Then, coordinate carries out accumulative summation for the frame-to-frame differences of the point of (x, y).
Wherein, S (x, y) denotation coordination for the accumulative of the frame-to-frame differences of the point of (x, y) and.
Finally, binaryzation is carried out to S (x, y), whether stagnant zone is belonged to using the point for judging coordinate as (x, y).
Wherein, Q (x, y) is the result to being exported after S (x, y) progress binaryzations, and q is the threshold value pre-set.When Q (x,
When y) being output as 0, then judge that coordinate belongs to stagnant zone for the point of (x, y)；When Q (x, y) is output as 1, then judge that coordinate is
The point of (x, y) is not belonging to stagnant zone.
By using frame differential method to the point of each in video image to judge whether to belong to stagnant zone, so as to
Into video image belong to stagnant zone coordinate a little, and then can obtain in video image can not overlay area position
Confidence ceases.
In this embodiment, because the algorithm realization of frame differential method is simple, programming complexity is low, operation stability
Well, the features such as identification accuracy is high, therefore can be fast and accurately identified based on multi-frame video image and using frame differential method
Stagnant zone in video image.It should be noted that identifying the stagnant zone in video image using frame differential method
Technological means is the means that are preferable to carry out of the invention, but this can't be produced in limitation, the present invention to technical scheme
The stagnant zone in video image can be also got by other Digital Image Processing, be will not enumerate herein.
Alternatively, digital image processing unit 12 includes：Second, which slides window unit 21, the second judging unit 22, second, sits
Mark acquiring unit 23 and choose unit 24.
Wherein, second window unit 21 is slided for the bottom right edge region foundation in video image with displaying size
Second sliding window, the second sliding window is scanned with the second pre- fixed step size to bottom right edge region；Second judging unit 22 is used for
Judge region in the second sliding window with can not overlay area whether there is overlapping region；If the second coordinate acquiring unit 23 is used for
Second judging unit 22 judge region in the second sliding window with can not overlay area overlapping region is not present when, then second slide
Region in dynamic window is an optional display area, obtains the coordinate all put in optional display area to be used as optional show area
The positional information in domain, can be from bottom right edge area after the second sliding window completes to scan whole video image, it is necessary to explanation
At least one optional display area is got in domain；Choose unit 24 be used to choosing from whole optional display areas one with
As display area, the positional information for the optional display area being picked is the positional information of display area.
Fig. 4 is the schematic diagram in the bottom right edge region of video image in the embodiment of the present invention two, as shown in figure 4, in reality
It is not influence user to view and admire programme content in, the bottom right edge region that pushed information is predefined in into video image is entered
Row displaying, the bottom right edge region specifically refers to be less than with the distance of the lower side of video imageRegion and with video figure
The distance of the right edge of picture is less thanRegion, wherein, H be video image height, W be video image width, m and n are
The fixed value pre-set.Preferably, the value that m value is 5, n is 5.
Still optionally further, choosing unit 24 includes：Computation subunit 25 and selection subelement 26, wherein, calculate son single
Member 25 is used for the distance between the intermediate point for calculating each optional display area and the best illustrated point S pre-set；Choose son
Unit 26 is used to choose the minimum optional display area of the distance between intermediate point and best illustrated point S as display area.From
All selection one is as follows using the detailed process as display area in optional display area：
First, the distance between the intermediate point that calculates each optional display area and the best illustrated point S that pre-sets：
Wherein, d is the distance between the intermediate point of optional display area and the best illustrated point S pre-set, and x and y divide
The abscissa and ordinate of the intermediate point of optional display area are not represented, and x0 and y0 represent best illustrated point S abscissa respectively
And ordinate.Preferably, the distance of the lower side of best illustrated point S and video image isBest illustrated point S and video
The distance of the right edge of image isThat is coordinates of the best illustrated point S in the set up coordinate system that originally carries out an invention beWhen the point midway of display area is in best illustrated point S, user can not influenceed to view and admire in program
On the premise of appearance, the visual attention location of user is obtained to greatest extent.
Then, the distance between the intermediate point that compares each optional display area and the best illustrated point S that pre-sets, and
The minimum optional display area of the distance between correspondence intermediate point and best illustrated point S is selected as display area.
In actual applications, because the intermediate point of display area is nearer from best illustrated point S, then the vision of user is got
Attention rate is higher, thus in the present embodiment by select the distance between correspondence intermediate point and best illustrated point S it is minimum can
Display area is selected as display area, can concern of the maximum guarantee user limited to pushed information.
It should be noted that by calculating the intermediate point of each optional display area and pre-setting most in the present embodiment
The distance between good displaying point S, then chooses the intermediate point optional display area work minimum with the distance between best illustrated point S
For display area technological means be the present invention be preferable to carry out means, this can't to technical scheme produce limit
System.In the present invention, can choose an arbitrary optional display area using as display area, can realize display area with not
Can overlay area no overlap.
The course of work of the pusher shown in Fig. 3 is described in detail below in conjunction with accompanying drawing.Fig. 5 is real for the present invention
The flow chart of the method for pushing of the offer of example two is applied, as shown in figure 5, the method for pushing is based on the pusher shown in Fig. 3, the push
Method includes：
Step 201：Some frame video images are generated according to the multi-medium data got.
Step 201 is performed by the first image generation unit 11.
Step 202：Some frame video images of generation are carried out Digital Image Processing to identify some frame video images
In stagnant zone, the coordinate all put in stagnant zone constitute can not overlay area positional information.
Step 202 is performed by digital image processing unit 12.It should be noted that to judge static in video image
When region or moving region, then need just be learnt according to the progress Digital Image Processing of multi-frame video image.
Preferably, Digital Image Processing is inter-frame difference processing, handles to recognize in video image by inter-frame difference
The detailed process of stagnant zone can be found in foregoing teachings, and here is omitted.The coordinate all put in stagnant zone is constituted can not
The positional information of overlay area.
Step 203：The second sliding window with displaying size is set up in bottom right edge region in video image, and second slides
Dynamic window is scanned with the second pre- fixed step size to bottom right edge region.
Step 203 is performed by the second slip window unit 21.A certain region is set up sliding window and slided in video image
The technological means of dynamic window scanning is the common technology means in this area, is not described in detail herein.
Step 204：Judge region in the second sliding window with can not overlay area whether there is overlapping region.
Step 204 is performed by the second judging unit 22.Second judging unit 22 is by contrasting in the second sliding window and not
Can overlay area whether there is the coordinate of identical point so that can determine whether region in the second sliding window with can not overlay area be
It is no to there is overlapping region.When the second judging unit 22 judge region in the second sliding window with can not overlay area exist it is overlapping
During region, then sliding window step moves to the next position, and rejudge the region in the second sliding window with can not overlay area whether
There is overlapping region；When the second judging unit 22 judge region in the second sliding window with can not overlay area be not present it is overlapping
During region, then the region in the second sliding window is an optional display area, performs step 205.
Step 205：The coordinate all put in optional display area is obtained using the positional information as optional display area.
Step 205 is performed by the second coordinate acquiring unit 23.It should be noted that step 204 and step 205 are
What two sliding windows were synchronously carried out during being scanned with the second pre- fixed step size to bottom right edge region.When the second sliding window is complete
In pairs after whole video image scanning, at least one optional display area can be got from bottom right edge region and this is optional
The positional information of display area.
Step 206：One is chosen from whole optional display areas using as display area, the optional exhibition being picked
The positional information for showing region is the positional information of display area.
The step 206 is performed by selection unit 24.Choose unit 24 can choose an arbitrary optional display area with
As display area, can realize display area with can not overlay area no overlap.
Preferably, choosing unit 24 includes：Computation subunit 25 and selection subelement 26, step 206 includes：
Step 2061：The intermediate point of each optional display area is calculated between the best illustrated point that pre-sets away from
From.
Step 2062：Choose the minimum optional display area of the distance between intermediate point and best illustrated point and be used as show area
Domain.
Step 2061 is performed by computation subunit 25, and step 2062 is performed by selection subelement 26, step 2061 and
The detailed process of step 2062 can be found in foregoing teachings, and here is omitted.
Step 207：The positional information of pushed information and display area is sent to client to push away in display area displaying
Deliver letters breath.
Step 207 is performed by pushing module 3.The detailed process of the step can be found in above-described embodiment one to step
103 description, here is omitted.
Embodiment three
Fig. 6 is the structured flowchart for the pusher that the embodiment of the present invention three is provided, as shown in fig. 6, the pusher includes：
Grader generation module 4, acquisition module 1, computing module 2 and pushing module 3, wherein acquisition module 1 include：Second image is generated
Unit 13, first slides window unit 14, the first judging unit 15 and the first coordinate acquiring unit 16.
Wherein, grader generation module 4 is used for basis has preliminary dimension from what some frame video images were extracted in advance
Each bar additional information the generation of sampling picture block and several one-to-one Image Classifiers of each bar additional information, image point
The positive sample of class device is all sampling picture blocks for belonging to same additional information, and the negative sample of Image Classifier is from some frames
The sampling picture block of several programme contents extracted in video image, the size and each bar of the sampling picture block of programme content is attached
Plus the size of the sampling picture block of information is identical, all images grader constitutes additional information disaggregated model.
Second image generation unit 13 is used to generate a frame video image according to the multi-medium data got；First slides
Window unit 14 is used in a frame video image of generation set up the first sliding window with preliminary dimension, and the first sliding window is with the
One pre- fixed step size is scanned to a frame video image；First judging unit 15 is used to according to additional information disaggregated model judge
Whether the picture in the first sliding window belongs to additional information；First coordinate acquiring unit 16 is used for when the first judging unit is judged
When picture in first sliding window belongs to additional information, each corresponding coordinate of point, the first sliding window in the first sliding window are obtained
The coordinates of whole points for completing to get after scanning whole video image constitute can not overlay area positional information.
The difference of the present embodiment and above-described embodiment two is, is to use the method for grader classification to know in the present embodiment
Do not go out in video image can not overlay area positional information, and in the present embodiment only need obtain a frame video image after i.e.
It can be identified.
In the present embodiment, use the method that grader is classified with identify in video image can not overlay area position
The substantially process of information is as follows：
First, according to manually to extract the sampling of each bar additional information with preliminary dimension out of multi-frame video image
Picture block.
Then, from the picture in multi-frame video image manually to extract the sampling picture of size and each bar additional information
The sampling picture block of several programme contents of block size identical.
Then, grader generation model is to belong to all sampling picture blocks of same additional information as positive sample, with
The sampling picture block of programme content is trained as negative sample to positive and negative samples, and bar additional information is corresponded to from obtaining can recognize that
An Image Classifier.
Image Classifier with generation is that (this TV guide can be in multi-frame video figure for recognizing a certain bar TV guide
As in occur) in case of be described.The corresponding positive sample of the Image Classifier is extracted from multi-frame video image
The sampling picture block of the TV guide, negative sample is the sampling picture block of the programme content provided from multi-frame video image, and
The size all same of each sampling picture block, may recognize that in video image using the Image Classifier in positive and negative samples
TV guide.
Based on above-mentioned principle, grader generation module can generation quantity and additional information number of species identical several
Image Classifier, Image Classifier one additional information of correspondence, all images grader constitutes an additional information classification
Model, the additional information disaggregated model may recognize that whole additional letters in all kinds of additional informations in video image, video image
The corresponding region of breath is can not overlay area.
Finally, according to additional information disaggregated model to all kinds of additional informations in video image to be identified, while sharp
With the first coordinate acquiring unit to be recorded to each corresponding coordinate of point in these additional informations, it can not be covered so as to learn
The positional information of cover area.
The course of work of the pusher shown in Fig. 6 is described in detail below in conjunction with accompanying drawing.Fig. 7 is real for the present invention
The flow chart of the method for pushing of the offer of example three is applied, as shown in fig. 7, the method for pushing is based on the pusher shown in Fig. 6, the push
Method includes：
Step 301：According to each bar additional information with preliminary dimension extracted in advance from some frame video images
Picture block of sampling generation and several one-to-one Image Classifiers of each bar additional information.
Step 301 is performed by grader generation module 4.Grader generation module 4 is generated with each bar additional information one by one
The process of corresponding each Image Classifier is referring to foregoing teachings, and here is omitted, and all images grader constitutes additional information
Disaggregated model.
Step 302：One frame video image is generated according to the multi-medium data got.
Step 302 is performed by the second image generation unit 13.
Step 303：The first sliding window with preliminary dimension, the first sliding window are set up in a frame video image of generation
One frame video image is scanned with the first pre- fixed step size.
Step 303 is performed by the first slip window unit 14.First slides window unit 14 in the video image got
The first sliding window that there is identical size with the sampling picture block in training set sample is set up, first sliding window is predetermined with first
Step-length is scanned to whole video image.
Step 304：Whether additional information is belonged to the picture judged in the first sliding window according to additional information disaggregated model.
Step 304 is performed by the first judging unit 15.During the first sliding window is scanned, the first judging unit 15
Whether additional information is belonged to the picture judged in the first sliding window using each Image Classifier in additional information disaggregated model.
Specifically, whether corresponding additional information is belonged to the picture judged in the first sliding window using each Image Classifier successively, when
When some Image Classifier judges that the picture in the first sliding window belongs to corresponding additional information, then the sliding window of surface first
Interior region belongs to can not overlay area, execution step 305.When all images grader all judges the picture in the first sliding window
When face is not belonging to corresponding additional information, then the first judging unit 15 judges that the picture in the first sliding window is not belonging to additional letter
During breath, the first sliding window step moves to the next position, and re-uses additional information disaggregated model and judge picture in the first sliding window
Whether face belongs to additional information.
It should be noted that using Image Classifier using judge the process of the corresponding additional information in the first sliding window as
The state of the art, here is omitted.
Step 305：Obtain each corresponding coordinate of point in the first sliding window.
Step 305 is performed by the first coordinate acquiring unit 16.When the first judging unit 15 is judged in the first sliding window
Picture when belonging to additional information, the first coordinate acquiring unit 16 obtains in the first sliding window each and puts corresponding coordinate.Need
Illustrate, step 304 and step 305 are the processes being scanned in the first sliding window with the first pre- fixed step size to video image
Middle synchronous progress.After the first sliding window completes to scan whole video image, what the first coordinate acquiring unit 16 was got
All point coordinates constitute can not overlay area positional information.
Step 306：According to the displaying size of pushed information and can not the positional information calculation of overlay area go out video image
In display area positional information.
Step 306 is performed by acquisition module 1.
Alternatively, the detailed process of step 306 can be realized by step 203~step 206 in above-described embodiment two, tool
Hold the description that can be found in above-described embodiment two to step 203~step 206 in vivo.
Step 307：The positional information of pushed information and display area is sent to client to push away in display area displaying
Deliver letters breath.
Step 307 is performed by pushing module 3.The detailed process of the step can be found in above-described embodiment one to step
103 description, here is omitted.
Example IV
The embodiment of the present invention four provides a kind of server, and the server includes pusher, and the pusher can be used
The pusher provided in above-described embodiment one, embodiment two or embodiment three, the method for pushing of the pusher can be found in pair
Answer the description in embodiment.
It is understood that the principle that embodiment of above is intended to be merely illustrative of the present and the exemplary implementation that uses
Mode, but the invention is not limited in this.For those skilled in the art, the essence of the present invention is not being departed from
In the case of refreshing and essence, various changes and modifications can be made therein, and these variations and modifications are also considered as protection scope of the present invention.
Claims (14)
1. a kind of method for pushing, it is characterised in that including：
The sampling picture of additional information according to each bar with preliminary dimension extracted in advance from some frame video images
Block is generated and several one-to-one Image Classifiers of additional information described in each bar, and the positive sample of described image grader is category
In all sampling picture blocks of additional information described in same, the negative sample of described image grader is from some frame video images
The sampling picture block of several programme contents of middle extraction, the size of the sampling picture block of the programme content with it is attached described in each bar
Plus the size of the sampling picture block of information is identical, whole described image graders constitute additional information disaggregated model；
Obtain in server inner video image can not overlay area positional information, it is described can not overlay area to be covered in program
It is used for the region for showing additional information on content, including：
The video image according to the multi-medium data got generates a frame；
The first sliding window with the preliminary dimension, first sliding window are set up in video image described in a frame in generation
Video image described in one frame is scanned with the first pre- fixed step size；
Whether additional information is belonged to the picture judged in the first sliding window according to the additional information disaggregated model, when judging
When picture in first sliding window belongs to additional information, each corresponding coordinate of point in the first sliding window is obtained, described first slides
Dynamic window complete the whole points got after being scanned to whole video image coordinates constitute it is described can not the position of overlay area believe
Breath；
According to the displaying size of pushed information and it is described can not the positional information calculation of overlay area go out in the video image
The positional information of display area, the display area is to be covered in be used for the area for showing the pushed information on programme content
Domain, the display area with it is described can not overlay area do not overlap；
The positional information of the pushed information and the display area is sent to client to show institute in the display area
State pushed information.
2. method for pushing according to claim 1, it is characterised in that can not be covered in the acquisition server inner video image
The step of positional information of cover area, includes：
The video image according to the multi-medium data got generates some frames；
Video image described in some frames of generation is carried out Digital Image Processing to identify in video image described in some frames
The coordinate all put in stagnant zone, the stagnant zone constitute it is described can not overlay area positional information.
3. method for pushing according to claim 2, it is characterised in that the Digital Image Processing is specially frame-to-frame differences office
Reason.
4. method for pushing according to claim 1, it is characterised in that the displaying size according to pushed information and described
Can not the positional information step of display area that goes out in the video image of positional information calculation of overlay area include：
The second sliding window with the displaying size is set up in bottom right edge region in the video image, and described second slides
Dynamic window is scanned with the second pre- fixed step size to the bottom right edge region, and the bottom right edge region is and the video image
The distance of lower side be less thanRegion and be less than with the distance of the right edge of the video imageRegion, wherein H is
The height of video image, W is the width of video image, and m and n are the fixed value pre-set；
Judge region in second sliding window with it is described can not overlay area whether there is overlapping region, it is described when judging
Region in second sliding window with it is described can not overlay area overlapping region is not present when, then assert in second sliding window
Region is an optional display area, and obtains the coordinate all put in the optional display area to be used as the optional displaying
The positional information in region, second sliding window completes to get at least one described optional exhibition after scanning whole video image
Show region；
One is chosen from whole optional display areas using as the display area, the optional exhibition being picked
The positional information for showing region is the positional information of the display area.
5. method for pushing according to claim 4, it is characterised in that
M value is 5；
N value is 5.
6. method for pushing according to claim 4, it is characterised in that described to be selected from whole optional display areas
The step of one is taken using as the display area includes：
Calculate the distance between the intermediate point of each optional display area and best illustrated point for pre-setting；
Choose the minimum optional display area of the distance between intermediate point and best illustrated point and be used as the display area.
7. method for pushing according to claim 6, it is characterised in that
The distance of the best illustrated point and the lower side of the video image is
The best illustrated point and the distance of the right edge of the video image are
8. a kind of pusher, it is characterised in that including：
Acquisition module, for obtain in server inner video image can not overlay area positional information, it is described can not the area of coverage
Domain is to be covered in be used for the region for showing additional information on programme content；The acquisition module includes：
Second image generation unit, for the video image according to multi-medium data one frame of generation got；
First slides window unit, for setting up the first cunning with the preliminary dimension in video image described in the frame in generation
Dynamic window, first sliding window is scanned with the first pre- fixed step size to video image described in a frame；
First judging unit, for judging whether the picture in the first sliding window belongs to according to the additional information disaggregated model
Additional information；
First coordinate acquiring unit, for judging that the picture in the first sliding window belongs to additional information when the first judging unit
When, each corresponding coordinate of point in the first sliding window is obtained, first sliding window completes to obtain after scanning whole video image
The coordinates for the whole points got constitute it is described can not overlay area positional information；
Computing module, for the displaying size according to pushed information and it is described can not overlay area positional information calculation go out described in
The positional information of display area in video image, the display area is used to show described push away to be covered on programme content
Deliver letters the region of breath, the display area with it is described can not overlay area do not overlap；
Pushing module, for the positional information of the pushed information and the display area to be sent to client with the exhibition
Show that region shows the pushed information；The pusher also includes：
Grader generation module, is added according to each bar with preliminary dimension extracted in advance from some frame video images
The sampling picture block generation of information and several one-to-one Image Classifiers of additional information described in each bar, described image classification
The positive sample of device is all sampling picture blocks for belonging to additional information described in same, the negative sample of described image grader be from
The sampling picture block of several programme contents extracted in some frame video images, the chi of the sampling picture block of the programme content
Very little identical with the size of the sampling picture block of additional information described in each bar, whole described image graders constitute additional information classification
Model.
9. pusher according to claim 8, it is characterised in that the acquisition module includes：
First image generation unit, for the video image according to some frames of multi-medium data generation got；
Digital image processing unit, if carrying out Digital Image Processing to identify for video image described in some frames to generation
The coordinate all put in stagnant zone in video image described in dry frame, the stagnant zone constitute it is described can not overlay area
Positional information.
10. pusher according to claim 8, it is characterised in that the computing module includes：
Second slides window unit, and the with the displaying size is set up for the bottom right edge region in the video image
Two sliding windows, second sliding window is scanned with the second pre- fixed step size to the bottom right edge region, the bottom right edge
Region is less than for the distance with the lower side of the video imageRegion and distance with the right edge of the video image
It is less thanRegion, wherein, H be video image height, W be video image width, m and n are the fixed value pre-set；
Second judging unit, for judge the region in second sliding window with it is described can not overlay area with the presence or absence of overlapping
Region；
Second coordinate acquiring unit, for when the second judging unit judge region in second sliding window with it is described can not
When overlapping region is not present in overlay area, it is an optional display area to assert the region in second sliding window, and is obtained
The coordinate all put in the optional display area is using the positional information as the optional display area, second sliding window
Complete to get at least one described optional display area after scanning whole video image；
Unit is chosen, one is chosen from whole optional display areas as the display area, to be picked
The positional information of the optional display area is the positional information of the display area.
11. pusher according to claim 10, it is characterised in that
M value is 5；
N value is 5.
12. pusher according to claim 10, it is characterised in that the selection unit includes：
Computation subunit, between the best illustrated point that calculates the intermediate point of each optional display area and pre-set
Distance；
Subelement is chosen, for choosing the optional display area of the distance between intermediate point and best illustrated point minimum as described
Display area.
13. pusher according to claim 12, it is characterised in that
The distance of the best illustrated point and the lower side of the video image is
The best illustrated point and the distance of the right edge of the video image are
14. a kind of server, it is characterised in that including any pusher in the claims 8-13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410855843.XA CN104519371B (en) | 2014-12-31 | 2014-12-31 | Method for pushing, pusher and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410855843.XA CN104519371B (en) | 2014-12-31 | 2014-12-31 | Method for pushing, pusher and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104519371A CN104519371A (en) | 2015-04-15 |
| CN104519371B true CN104519371B (en) | 2017-10-20 |
Family
ID=52793995
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410855843.XA Active CN104519371B (en) | 2014-12-31 | 2014-12-31 | Method for pushing, pusher and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104519371B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105430471B (en) * | 2015-11-26 | 2018-07-24 | 无锡天脉聚源传媒科技有限公司 | The display methods and device of barrage in a kind of video |
| CN106993227B (en) | 2016-01-20 | 2020-01-21 | 腾讯科技（北京）有限公司 | Method and device for information display |
| CN106534939B (en) * | 2016-10-09 | 2019-07-05 | 北京金山安全软件有限公司 | Multi-video preview method and device and electronic equipment |
| CN109993579B (en) * | 2019-03-28 | 2021-04-02 | 北京酷云互动科技有限公司 | Advertisement delivery area determination method, system, server and computer readable medium |
| CN112752121B (en) * | 2020-05-26 | 2023-06-09 | 腾讯科技（深圳）有限公司 | Video cover generation method and device |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103797812B (en) * | 2012-07-20 | 2018-10-12 | 松下知识产权经营株式会社 | Band comments on moving image generating means and with comment moving image generation method |
| KR102040611B1 (en) * | 2013-05-31 | 2019-11-05 | 엘지전자 주식회사 | Mobile terminal and controlling method thereof |
| US9467750B2 (en) * | 2013-05-31 | 2016-10-11 | Adobe Systems Incorporated | Placing unobtrusive overlays in video content |
| CN103634667B (en) * | 2013-12-12 | 2018-02-09 | 珠海迈越信息技术有限公司 | A kind of method and apparatus of the acquisition network push information based on set top box |
-
2014
- 2014-12-31 CN CN201410855843.XA patent/CN104519371B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104519371A (en) | 2015-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104519371B (en) | Method for pushing, pusher and server | |
| CN107347166B (en) | Video image processing method and device and terminal equipment | |
| US10713528B2 (en) | System for determining alignment of a user-marked document and method thereof | |
| EP3745339A1 (en) | Method for implanting advertisements in video, and computer device | |
| WO2017028502A1 (en) | Method and system for superimposing image in video stream | |
| CN102750555B (en) | Expression identification device applied to instant messaging tool | |
| CN107146198A (en) | A kind of intelligent method of cutting out of photo and device | |
| CN110827312B (en) | Learning method based on cooperative visual attention neural network | |
| JPWO2020179065A1 (en) | Image processing equipment, image processing methods and programs | |
| CN110827193A (en) | Panoramic video saliency detection method based on multi-channel features | |
| CN111368682B (en) | Method and system for detecting and identifying station caption based on master RCNN | |
| CN110097616B (en) | Combined drawing method and device, terminal equipment and readable storage medium | |
| CN104102904A (en) | Static gesture identification method | |
| CN105282543A (en) | Total blindness three-dimensional image quality objective evaluation method based on three-dimensional visual perception | |
| TW201905757A (en) | Electronic device and gesture recognition method applied therein | |
| CN105159923A (en) | Video image based article extraction, query and purchasing method | |
| CN113705510A (en) | Target identification tracking method, device, equipment and storage medium | |
| US20220138840A1 (en) | Systems and methods for visualizing wall coverings in an image of a scene | |
| CN107203961B (en) | Expression migration method and electronic equipment | |
| CN106780474B (en) | Kinect-based real-time depth map and color map registration and optimization method | |
| CN110782390A (en) | Image correction processing method and device and electronic equipment | |
| CN102609958A (en) | Method and device for extracting video objects | |
| US9798932B2 (en) | Video extraction method and device | |
| CN111582148B (en) | Beijing opera character recognition method, device, storage medium and apparatus | |
| CN113486941B (en) | Live image training sample generation method, model training method and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |