CN101692269B - Method and device for processing video programs - Google Patents
Method and device for processing video programs Download PDFInfo
- Publication number
- CN101692269B CN101692269B CN200910236043.9A CN200910236043A CN101692269B CN 101692269 B CN101692269 B CN 101692269B CN 200910236043 A CN200910236043 A CN 200910236043A CN 101692269 B CN101692269 B CN 101692269B
- Authority
- CN
- China
- Prior art keywords
- video
- stable
- information
- image
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 230000008859 change Effects 0.000 claims description 16
- 238000001914 filtration Methods 0.000 claims description 15
- 239000000284 extract Substances 0.000 claims description 11
- 238000009825 accumulation Methods 0.000 claims description 7
- 230000000007 visual effect Effects 0.000 claims description 7
- 230000013011 mating Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 208000001491 myopia Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 244000309466 calf Species 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000007430 reference method Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Abstract
The invention discloses a method and a device for processing video programs. The method comprises the following steps: analyzing video to acquire stable video information in the video programs; extracting character information in the stable video information by using a character identifying method; and dividing the video programs according to the character information to acquire video contents corresponding to the same programs. The method and the device provided by the invention for processing the video programs identify the character information from the stable information of television programs, and classify and store the television programs according to the character information so that users can search the interested video programs from the recorded video programs conveniently.
Description
Technical field
The present invention relates generally to field of video processing, refers in particular to a kind of method and apparatus of automatic processing video programs.
Background technology
Along with the progress of science and technology and the raising of people's living standard, people need more stress-relieving activity to alleviate working pressure and life stress, be that the video frequency program of representative just plays such role in the life of modern with TV programme, and be also that the mankind obtain the most important channel of external information.
But there is many defects in prior art.For TV programme, present TV programme is a lot, and the work of people is extremely busy again, all TV programme all can not be finished watching, even if wanting interested program all to finish watching also has been difficult to, because TV programme has real-time.In this case, occurred recording arrangement on TV, television program record can get off by it, watches at one's leisure for people.But this recording needs user to select recording time starting point and recording time terminal, needs artificial participation.If the broadcast time of oneself liking program cannot be learnt, then can not complete the recording of TV programme.If all TV programme are all recorded, then user requires a great deal of time and search the interested program of user from recorded program, sometimes also may can not find interested program from the large-scale program recorded.Although main for TV programme above, all there is analogue in a lot of video frequency program.Therefore, in the urgent need to a kind of method of processing video programs.
Summary of the invention
The present invention proposes a kind of method and apparatus of processing video programs, can automatically divide the video frequency program comprising TV programme, thus facilitates user to search interested video frequency program.
The technical scheme of the embodiment of the present invention is achieved in that
A method for processing video programs, comprising:
Analyze video, obtain the stable video information in video frequency program;
Character recognition method is utilized to extract Word message in described stable video information;
According to described Word message, video frequency program is divided, obtain the video content that same program is corresponding.
Preferably, described analysis video, the stable video information obtained in video frequency program comprises:
Stable information search region is set;
Whether change according to image information in described stable information search region and determine stable video information.
Preferably, described whether changing according to image information in described stable information search region determines that stable video information is specially:
Stable video information is determined in brightness change according to image pixel in described stable information search region.
Preferably, the described change of the brightness according to image pixel in described stable information search region determines that stable video information comprises:
Initialization is carried out to the image pixel in stable information search region;
Calculate the variance of pixel brightness in described stable information search region
According to described σ
2determine stable video information, wherein Y
s=Y
s+ y (w, h), Y
2s=Y
2s+ y (w, h)
2, y (w, h) is the brightness of image at pixel (w, h) place, Y
sfor pixel image intensity and, be initialized as zero, Y
2sfor pixel image intensity quadratic sum, be initialized as zero, accumulation frame number T=T+1, accumulation frame number T is initialized as zero.
Preferably, the described change of the brightness according to image pixel in described stable information search region determines that stable video information comprises:
Initialization is carried out to the pixel in stable information search region;
Calculate the degree of stability of pixel and determine stable pixel according to degree of stability;
According to described stable pixel determination stable video information.
Preferably, according to the degree of stability of following formulae discovery pixel:
Wherein, y
tfor brightness, y
ifor the brightness of this pixel during moment t, K () is kernel function, N be greater than 1 positive integer.
Preferably, also comprise:
According to Word message, described video content is classified.
Preferably, also comprise after described video content being classified according to Word message:
Similar video is mated, the repetition video in the similar video of filtering.
Preferably, describedly to mate similar video, the repetition video in the similar video of filtering is specially:
From the first video content of similar video, randomly draw the first image, in the second video content of similar video, find the second image with described first images match;
Analyze after described first video content and described second video content alignment frame by frame with the position of the position of described first image and described second image, judge that whether described first image is identical with the corresponding section of described second image, if the same delete described first video content or described second video content.
Preferably, also comprise:
The index list of described video content is generated according to Word message and visual classification.
Preferably, also comprise:
Corresponding video content is recorded according to the key message that user is arranged.
A device for processing video programs, comprising:
Analytic unit, for analyzing video, obtains the stable video information in video frequency program;
Recognition unit, extracts Word message in described stable video information for utilizing character recognition method;
Acquiring unit, for dividing video frequency program according to described Word message, obtains the video content that same program is corresponding.
Preferably, also comprise:
Taxon, for classifying to described video content according to Word message.
Preferably, also comprise:
Filtering module, for mating similar video, the repetition video in the similar video of filtering.
Preferably, also comprise:
Index generation unit, for generating the index list of described video content according to Word message and visual classification.
Preferably, also comprise:
Recording elements, the key message for arranging according to user records corresponding video content.
The present invention identifies Word message from the stable information of TV programme, according to Word message, TV programme is carried out classification and stores, thus facilitates user to search interested video frequency program from the video frequency program recorded; Further, the present invention can also record the interested TV programme of user according to Word message, thus realizes the robotization of television program record, without the need to artificial participation, provides better Consumer's Experience.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme of the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the process flow diagram of method first embodiment of a kind of processing video programs of the present invention;
Fig. 2 is step S110 particular flow sheet in Fig. 1;
Fig. 3 is the process flow diagram of method second embodiment of a kind of processing video programs of the present invention;
Fig. 4 be stablize information search region in Fig. 2 in step S111 schematic diagram is set;
Fig. 5 is the composition structural drawing of device first embodiment of a kind of processing video programs of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Video frequency program in various embodiments of the present invention is described for TV programme, but technical scheme of the present invention goes for the video frequency program of other type equally.
With reference to Fig. 1, show the process flow diagram of method first embodiment of a kind of processing video programs of the present invention.The method of described processing video programs comprises step:
Step S110, analysis video, obtain the stable video information in video frequency program.
Described stable video information refers to the information that video frequency program picture remains unchanged in N continuous frame, and wherein N is more than or equal to 2, and finally, the stable video information obtained is a series of video images meeting described condition.
In video frequency program, picture is constantly change.But; recording and relay side usually can add some information in video; these information keep stable in the picture of change; continuously a lot of frame does not change, and comprises TV station symbol, the title of TV programme, collection sequence number, sports tournament live or relay printed words, title and team's title of taking part in game etc. in these information.Certainly, described stable information not only comprises foregoing, also comprises the information that other remains unchanged in video.Aforementioned stable video information is the most direct information that can obtain in TV programme, is the most direct information source obtaining video program content.
Step S120, character recognition method is utilized to extract Word message in described stable video information.
In aforementioned stable video information, except the icons such as TV station symbol, topmost has been exactly the Word message comprised in video.Word Input and recognition methods can be utilized from stable video image to obtain Word message.Word Input and identification are prior aries, and in order to length is considered, the embodiment of the present invention is not specifically described character recognition method, with reference to the data of related art.
The described Word message obtained is called as stable Word message in the present invention.
Step S130, according to described Word message, video frequency program to be divided, obtain the video content that same program is corresponding.
Video frequency program such as TV programme is Continuous Play, and different program is continuous print on video.In order to organize video frequency program, record as requested, need that segmentation is carried out to video frequency program and dividing, video segment corresponding for same program being divided and is organized in one section of video.
After obtaining video segment corresponding to same program, just can record specific video content or classification storage is carried out to the video content recorded.
With reference to Fig. 2, described step S110 can comprise the following steps:
Step S111, stable information search region is set.
Arranging stable information search region is complexity in order to reduce Processing Algorithm, thus improves processing speed, reduces hardware complexity.The establishing method of described stable region of search, can be set by system default, also can be arranged voluntarily by user.One method sets whole TV screen region as stablizing information search region.But rule of thumb, described stable information generally all appears at the left side of TV screen, the right and top edge, other place of picture needs information such as display television image and spoken title etc.Therefore, a kind of preferred plan of establishment is arranged as shown in Figure 4, and wherein, the width of region of search can be regulated by parameter.
Step S112, whether to change according to image information in described stable information search region and determine stable video information.
Determine that the method whether image changes is a lot, the method judged based on single pixel can be adopted, also can adopt the method based on piecemeal process.The image information adopted can be the monochrome information adopting pixel or region, and brightness also can be adopted in conjunction with the mode of colourity simultaneously.The criterion judged can adopt continuous multiple frames to ask the method for variance, and two continuous frames also can be adopted to ask the method for difference.
Introduce a kind of whether changing according to image information in described stable information search region below and determine the method for optimizing of stable video information, namely determine stable video information according to the brightness change of image pixel in described stable information search region.The method is by judging whether the variance of brightness in N continuous frame of single pixel in region of search is greater than threshold value and determines whether single pixel there occurs change, the video image pixel do not changed formed is as stable video information, described threshold value can wait factor to determine according to the actual requirements, and the present invention does not set this.
Introduce a kind of implementation procedure that the method is concrete below, in practice, alternate manner also can be adopted to carry out calculating and judging.
First, initialization is carried out to the pixel in stable information search region.
To each pixel in the stable information search region of setting, each pixel of initialization is in brightness of image in certain period and Y
sbe zero, brightness of image quadratic sum Y
2sbe zero, accumulation frame number T is zero.
Secondly, to certain point (w, h) in stable information search region, assuming that its brightness is in the current frame y (w, h), then upgrade Y
s=Y
s+ y (w, h), Y
2s=Y
2s+ y (w, h)
2, accumulation frame number T=T+1, calculating this variance square in accumulated time is
If σ
2be greater than a certain threshold value, then think that current pixel change too greatly, is not stable information pixels, and re-starts initialization, setting brightness of image and Y
sbe zero, brightness of image square and Y
2sbe zero, accumulation frame number T is zero; If T is greater than threshold value T1, and σ
2be less than a certain threshold value, then determine that current pixel is stable information pixels, thus determine stable video information, corresponding duration T is the time that stable video information continues.
The advantage of said method calculates simply, and memory space is little.The described change of the brightness according to pixel in described stable information search region determines that stable information is a kind of preferred version of the embodiment of the present invention, can also determine stable video information according to other Parameters variation of pixel in described stable information search region.
Introduce the another kind of implementation procedure of the method below.
Following process is carried out to all pixels of candidate region (stablizing information search region):
First, initialization stores information.
Store the collection brightness of N number of time of this position order, assuming that the monochrome information in each moment is respectively y
0, y
1... y
n-1.
Secondly, calculate the degree of stability of pixel and determine whether stable pixel according to degree of stability.
To present frame, assuming that the brightness of this position is y
t, calculating current degree of stability is
Wherein, K () is kernel function, and a kind of mode is for getting the one dimension Gaussian function at 0 centered by K ().If then degree of stability is greater than a certain threshold value, then determine that current point is stable pixel, otherwise, determine that current point is unstable pixel, wherein, N be greater than 1 positive integer, can 100 be taken as.
Again, upgrade the information stored, by up-to-date Pixel Information replacement Pixel Information the earliest, namely delete y from queue
i, add y to queue
t-N, retention time ordinal relation simultaneously.
After obtaining the position of stable information pixels, simply, the region of all stable information pixels can directly can be comprised as stable information area.Another kind of mode is that information position mask image is stablized in setting, and this image is identical with original image size or proportional, the value of each position of this image is true or false, if the pixel in the original image that this position is corresponding is judged to be stable information, then this positional value is true, otherwise is false.First filtering is carried out to stable information position mask image, then it is carried out to the expansive working of binary picture, and then carry out the etching operation of binary picture, finally connected domain analysis is carried out to the image obtained, thus obtain stable information area.The method can stress release treatment, and, pixel will be stablized and connect for region, be convenient to follow-up Text region.
In order to reduce operand, in said method processing procedure, all can not process every two field picture, but every several frame process one two field picture.
Described step S130 can comprise the following steps:
Step S131, according to described Word message, video frequency program to be split.
For same program, the stable Word message wherein comprised is generally identical, can determine same video content according to identical stable Word message.Specifically, can using video identical for continuous in time and stable Word message as same video frequency program.First, the beginning frame number of initializing set current video is present frame, and the stable Word message recorded in this frame is current video segment stablizes Word message, for follow-up each frame, if the stable Word message of this frame is identical with the stable Word message of video segment, then think that this frame belongs to current video segment.If it is different to stablize Word message, then reinitializes, create new video segment, and this frame is stablized the stable Word message of Word message as newly-built video segment.
In reality, often insertion of advertising in a lot of video frequency program, in order to obtain same video program content better, the interference of filtering advertisement, described step S130 can comprise treatment step S132 further.
Step S132, according to described Word message, union operation is carried out to video segment.
To stable video fragment obtained above, if the last period video segment end time and the start time interval of rear one section of video segment be less than a certain threshold value (as 5S, 30S, 60S......), and, the two has identical stable Word message, then merge the two as same video content.
The peculiar information of the stable Word message in rear video program as this video frequency program can be divided using every section.After the division obtaining video content and stable Word message, may be used for the recording to video, the subsequent processes such as the tissue of video and video analysis.
In another embodiment of the invention, after step S130, can further include step S140, according to Word message, described video content classified.
A kind ofly according to Word message to the implementation that described video content is classified be:
Step S141, determine the key word of video content.
Word message in dividing for each video segment obtained above, therefrom can extract the information that video content classification is relevant, using the key word of these information as this video content.The generic of key word comprises TV station's title, programm name, program duration with other key word of program, the tournament names such as competing teams title, NBA in such as sport category match.Determine that a kind of method of key word is, enumerate out all TV stations title, all television programme titles, if described in the Word message that extracts comprise above-mentioned TV station title or television programme title, then it can be used as the key word of TV station's title or television programme title.Its keyword categories belonged to of location determination that can occur according to word further.Such as, using the word on the screen left side and the right appearance as programm name candidate, and the word occurred on screen upper left side is as TV station title candidate.
Step S142, determine the classification of video content.
General TV programme can be divided into sport category, economic class, news category, amusement class, TV play, film, educational, other class etc., described sorting technique also can have alternate manner, such as, increase or reduce described classification, or adopting other sorting technique to be such as divided into national program and external program.
Step S143, set the exclusive key word of each classification, and whether comprise described exclusive key word according to the key word of video content video content is classified.
First, can be the exclusive key word of program settings programm name.Because the program of partial TV platform all belongs to a type, such as five, central authorities, Beijing Liu Tai are typical sports cast TV station, can using central authorities 5, Beijing 6 as the exclusive key word of sport category program, and central authorities 2 are as the exclusive key word of economic class program, central authorities 6 are as the exclusive key word of film class program, and central authorities 10 are as the exclusive key word of educational program.
Secondly, other exclusive key word can also be set, such as, for the exclusive key word of sport category program settings is NBA, Division A League Football Matches of Italy, rocket, calf.And by news hookup, or news 30' is as the exclusive key word of news controlling.Meanwhile, TV play name can be referred to as the exclusive key word of TV play, using the exclusive key word of movie name as film, using the exclusive key word of movie name as film.Above-mentioned exclusive key word can set according to the situation of actual programs, and the present invention is not as restriction.
Then, the type of this video content is determined according to the exclusive key word whether extracting certain type in each video content.
Further, step S144, setting non-exclusive key word can also be comprised.
For each type programs setting non-exclusive key message, the non-exclusive key message such as setting TV play be length between 20-50 minute, and in Word message, last word comprises the numerals such as, two, 29 or 1,2,29.It is that classification that non-exclusive key word that it comprises is maximum by the program partition not comprising the exclusive key word of any classification.Described non-exclusive key word can carry out setting and changing according to the situation of actual programs.
More than describe the preferred embodiment of the present invention in detail to the method for classifying to video content; it will be appreciated by those skilled in the art that; the described method to video content classification is not restricted to above disclosed kind, and other method is certainly also within protection scope of the present invention.
In another embodiment of the invention, step S150 can also be comprised after described step S140, generate the index list of described video content according to Word message and visual classification.
A kind of implementation generating the index list of described video content according to Word message and visual classification is:
Step S151, video content to be classified.
Concrete sorting technique please refer to description above, does not repeat them here.
Step S152, according to classification video content is arranged.
Such as can carry out classification arrangement according to sport category, economic class, news category, amusement class, TV play, film, order that is educational, other class to video content.
Step S153, for generic program, to arrange according to predefined procedure.
Such as, for each class, can first arrange according to the pinyin order of programm name, identical then can the arranging according to broadcast time order of programm name.
For TV play, the numeral being positioned at afterbody in TV play programm name can be extracted, as the sequence number of TV play, according to sequence number, TV play of the same name be arranged.
Further, step S154 can also be comprised, generate a representative image or video segment for each video content.
After program category, programm name generating indexes catalogue, a pair or several representative image can be extracted or represent video segment for display, being convenient to user's preview simultaneously and selecting.A kind of simple mode can extract the mode of one or more small fragment composition video segment according to set time order.After user returns, click the index being concerned about program, namely can learn programme content, and then the program selecting oneself to like is watched.
More than describe the method for the preferred generating indexes catalogue of the embodiment of the present invention in detail; it will be appreciated by those skilled in the art that; the method of described generating indexes catalogue is not restricted to above disclosed kind, and the method for other generating indexes catalogue is certainly also within protection scope of the present invention.
Further, in another embodiment of the invention, after described step S140, before step S150, step S160 can also be comprised, similar video is mated, the repetition video in the similar video of filtering.
Because the situation that different TV station plays same TV programme can be there is, therefore repetition video content may be there is.In order to save storage resources further, more excellent mode is mated similar video, the repetition video in the similar video of filtering.
Be described below in detail and similar video is mated, a kind of implementation of the repetition video in the similar video of filtering.First judge as follows between two similar video, if the two programm name is identical, then carry out video segment coupling, a kind of optimal way is as follows:
Relatively two video lengths, random selecting several image from short video, these images are looked for matching image in long video, in order to pick up speed, can preferably adopt based on histogrammic matching process, to the random image extracted from short-sighted frequency, extract its histogram, according to fixed intervals, abstract image in long video, and extract image histogram, image histogram in image histogram in long video and short-sighted frequency is compared.If the two distance is less than a certain threshold value, then determines that the two is similar, continue to find near this long video image, mate most until find.With the position of this short video image and long video image, the two alignd, and analyze frame by frame, judge whether Histogram distance is less than a certain threshold value, if be less than, then both determining, corresponding section is identical.If programm name is different, or video segment does not mate, then the two is different.If the same delete one of them, thus filter out the video segment of repetition, save storage space.
Further, in another embodiment of the invention, step S170 can also be comprised, record corresponding video content according to the key message of user's setting.
Step S171, determine the key message classification of video content.
Word message in dividing for each video segment obtained above, therefrom can extract the information that video content is relevant, these information is convenient to user as the key word of this video content and sets select target video frequency program.
The generic of key word can include but not limited to program category, TV station's title, programm name, program duration, program category and out of Memory.TV station's title, programm name can refer step 201 method.And out of Memory can comprise competing teams title in sport category match, the tournament names such as NBA.
Step S172, user set the key message of program recording.
Above-mentioned key message can comprise in program category, TV station's title, programm name and out of Memory one or more, also can comprise other key message.
Step S173, according to user arrange key message record corresponding video content.
First, program is analyzed, extract stabilized zone, Text region, extract key message, video is divided, extract the key message dividing program, if key message meets user's setting, then video content is preserved.Such as, user wants the very 6+1 recording central authorities 2, then on operation program, set TV station and be defined as central authorities 2, programm name is unusual 6+1.If then comprise the key word of central authorities 2 and unusual 6+1 in the current ready-portioned program fragment obtained, then preserve this video content.User wants that recording Division A League Football Matches of Italy, England Premier League or other TV programme then sets corresponding key word, and preserves the video content comprising corresponding key word.
The present invention identifies Word message from the stable information of TV programme, according to Word message, TV programme is carried out classification and stores, thus facilitates user to search interested video frequency program from the video frequency program recorded; Further, the present invention can also record the interested TV programme of user according to Word message, thus realizes the robotization of television program record, without the need to artificial participation, provides better Consumer's Experience.
With reference to Fig. 5, show the composition structural drawing of device first embodiment of a kind of processing video programs of the present invention.The device 500 of described processing video programs comprises analytic unit 510, recognition unit 520, acquiring unit 530.
Described analytic unit 510, for analyzing video, obtains the stable video information in video frequency program.
Described recognition unit 520, for utilizing character recognition method to extract Word message in described stable video information.
Acquiring unit 530, for dividing video frequency program according to described Word message, obtain the video content that same program is corresponding.
The device 500 of described processing video programs also comprises:
Taxon, for classifying to described video content according to Word message.
The device 500 of described processing video programs also comprises:
Filtering module, for mating similar video, the repetition video in the similar video of filtering.
The device 500 of described processing video programs also comprises:
Index generation unit, for generating the index list of described video content according to Word message and visual classification.
The device 500 of described processing video programs also comprises:
Recording elements, the key message for arranging according to user records corresponding video content.
Device embodiment of the present invention is corresponding with described embodiment of the method, the course of work of the device of described processing video programs and principle of work have been described in detail in embodiment of the method part, do not repeat them here, the description of the device of described processing video programs and the course of work of each comprising modules and principle of work reference method embodiment appropriate section.
One of ordinary skill in the art will appreciate that; in each embodiment of the method for the present invention; the sequence number of described each step can not be used for the sequencing limiting each step; for those of ordinary skill in the art; under the prerequisite not paying creative work, the priority of each step is changed also within protection scope of the present invention.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (9)
1. a method for processing video programs, is characterized in that, comprising:
Analyze video, obtain the stable video information in video frequency program;
Character recognition method is utilized to extract Word message in described stable video information;
According to described Word message, video frequency program is divided, obtain the video content that same program is corresponding;
Described analysis video, the stable video information obtained in video frequency program comprises:
Stable information search region is set;
Whether change according to image information in described stable information search region and determine stable video information;
Also comprise:
According to Word message, described video content is classified;
Also comprise after described video content being classified according to Word message:
Similar video is mated, the repetition video in the similar video of filtering;
Determine the key message classification of video content;
User sets the key message of program recording;
Corresponding video content is recorded according to the key message that user is arranged.
2. the method for processing video programs according to claim 1, is characterized in that, described whether changing according to image information in described stable information search region determines that stable video information is specially:
Stable video information is determined in brightness change according to image pixel in described stable information search region.
3. the method for processing video programs according to claim 2, is characterized in that, the described change of the brightness according to image pixel in described stable information search region determines that stable video information comprises:
Initialization is carried out to the image pixel in stable information search region;
Calculate the variance of pixel brightness in described stable information search region
according to described σ
2determine stable video information, wherein Y
s=Y
s+ y (w, h), Y
2s=Y
2s+ y (w, h)
2, y (w, h) is the brightness of image at pixel (w, h) place, Y
sfor pixel image intensity and, be initialized as zero, Y
2sfor pixel image intensity quadratic sum, be initialized as zero, accumulation frame number T=T+1, accumulation frame number T is initialized as zero.
4. the method for processing video programs according to claim 2, is characterized in that, the described change of the brightness according to image pixel in described stable information search region determines that stable video information comprises:
Initialization is carried out to the pixel in stable information search region;
Calculate the degree of stability of pixel and determine stable pixel according to degree of stability;
According to described stable pixel determination stable video information.
5. the method for processing video programs according to claim 4, is characterized in that, the degree of stability according to following formulae discovery pixel:
Wherein, y
tfor brightness, y
ifor the brightness of this pixel during moment t, K () is kernel function, N be greater than 1 positive integer.
6. the method for processing video programs according to claim 1, is characterized in that, describedly mates similar video, and the repetition video in the similar video of filtering is specially:
From the first video content of similar video, randomly draw the first image, in the second video content of similar video, find the second image with described first images match;
Analyze after described first video content and described second video content alignment frame by frame with the position of the position of described first image and described second image, judge that whether described first image is identical with the corresponding section of described second image, if the same delete described first video content or described second video content.
7. the method for processing video programs according to claim 1, is characterized in that, also comprises:
The index list of described video content is generated according to Word message and visual classification.
8. a device for processing video programs, is characterized in that, comprising:
Analytic unit, for analyzing video, obtains the stable video information in video frequency program;
Recognition unit, extracts Word message in described stable video information for utilizing character recognition method;
Acquiring unit, for dividing video frequency program according to described Word message, obtains the video content that same program is corresponding;
Also comprise:
Taxon, for classifying to described video content according to Word message;
Also comprise:
Filtering module, for mating similar video, the repetition video in the similar video of filtering;
Recording elements, for determining the key message classification of video content;
The key message of program recording is set for user;
And the key message for arranging according to user records corresponding video content.
9. the device of processing video programs according to claim 8, is characterized in that, also comprises:
Index generation unit, for generating the index list of described video content according to Word message and visual classification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910236043.9A CN101692269B (en) | 2009-10-16 | 2009-10-16 | Method and device for processing video programs |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910236043.9A CN101692269B (en) | 2009-10-16 | 2009-10-16 | Method and device for processing video programs |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101692269A CN101692269A (en) | 2010-04-07 |
| CN101692269B true CN101692269B (en) | 2015-06-03 |
Family
ID=42080953
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910236043.9A Active CN101692269B (en) | 2009-10-16 | 2009-10-16 | Method and device for processing video programs |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101692269B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102053563A (en) * | 2010-12-27 | 2011-05-11 | 刘宇 | Flight training data acquisition and quality evaluation system of analog machine |
| CN103179464B (en) * | 2011-12-23 | 2018-01-19 | 乐金电子(中国)研究开发中心有限公司 | The method and apparatus for obtaining programme information in external input device of television |
| CN106096031B (en) * | 2016-06-27 | 2020-06-16 | 武汉斗鱼网络科技有限公司 | Video sequencing method and device with labels |
| CN107770528B (en) * | 2016-08-19 | 2023-08-25 | 中兴通讯股份有限公司 | Video playing method and device |
| CN106612457B (en) * | 2016-11-09 | 2019-09-03 | 广州视源电子科技股份有限公司 | Video sequence alignment schemes and system |
| JP7143620B2 (en) * | 2018-04-20 | 2022-09-29 | 富士フイルムビジネスイノベーション株式会社 | Information processing device and program |
| CN109118507B (en) * | 2018-08-27 | 2019-09-13 | 嵊州市万睿科技有限公司 | Shell bears air pressure real-time alarm system |
| CN109583443B (en) * | 2018-11-15 | 2022-10-18 | 四川长虹电器股份有限公司 | Video content judgment method based on character recognition |
| CN111314775B (en) | 2018-12-12 | 2021-09-07 | 华为终端有限公司 | Video splitting method and electronic equipment |
| CN112541390B (en) * | 2020-10-30 | 2023-04-25 | 四川天翼网络股份有限公司 | Frame extraction dynamic scheduling method and system for examination video violation analysis |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1404688A (en) * | 2000-12-18 | 2003-03-19 | 皇家菲利浦电子有限公司 | Apparatus and method of program classification using observed cues in the transcript information |
| CN1437402A (en) * | 2003-03-03 | 2003-08-20 | 北京中科大洋科技发展股份有限公司 | Ditital TV program broadcasting, receiving and presenting method |
| CN1516469A (en) * | 1998-05-19 | 2004-07-28 | ������Ƶ��Ʒ��˾ | Programme guide system with request TV program glance over function |
| CN101021857A (en) * | 2006-10-20 | 2007-08-22 | 鲍东山 | Video searching system based on content analysis |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008236688A (en) * | 2007-03-23 | 2008-10-02 | Hitachi Ltd | Television broadcasting receiver |
-
2009
- 2009-10-16 CN CN200910236043.9A patent/CN101692269B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1516469A (en) * | 1998-05-19 | 2004-07-28 | ������Ƶ��Ʒ��˾ | Programme guide system with request TV program glance over function |
| CN1404688A (en) * | 2000-12-18 | 2003-03-19 | 皇家菲利浦电子有限公司 | Apparatus and method of program classification using observed cues in the transcript information |
| CN1437402A (en) * | 2003-03-03 | 2003-08-20 | 北京中科大洋科技发展股份有限公司 | Ditital TV program broadcasting, receiving and presenting method |
| CN101021857A (en) * | 2006-10-20 | 2007-08-22 | 鲍东山 | Video searching system based on content analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101692269A (en) | 2010-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101692269B (en) | Method and device for processing video programs | |
| CN108769823B (en) | Direct broadcasting room display methods, device, equipment | |
| CN110020437B (en) | Emotion analysis and visualization method combining video and barrage | |
| US8457368B2 (en) | System and method of object recognition and database population for video indexing | |
| CN108322788B (en) | Advertisement display method and device in live video | |
| CN109145784B (en) | Method and apparatus for processing video | |
| US9271035B2 (en) | Detecting key roles and their relationships from video | |
| CN110971929B (en) | Cloud game video processing method, electronic equipment and storage medium | |
| CN108898145A (en) | A kind of image well-marked target detection method of combination deep learning | |
| CN103052953A (en) | Information processing device, method of processing information, and program | |
| CN110740387A (en) | bullet screen editing method, intelligent terminal and storage medium | |
| CN103069414A (en) | Information processing device, information processing method, and program | |
| CN105872717A (en) | Video processing method and system, video player and cloud server | |
| CN103051913A (en) | Automatic 3D (three-dimensional) film source identification method | |
| CN107205172A (en) | A kind of method and device that search is initiated based on video content | |
| CN104765886A (en) | Information acquisition method and device based on images | |
| CN105657514A (en) | Method and apparatus for playing video key information on mobile device browser | |
| CN113761253A (en) | Video tag determination method, device, equipment and storage medium | |
| Shinahara et al. | Serif or sans: Visual font analytics on book covers and online advertisements | |
| CN113542833A (en) | Video playing method, device and equipment based on face recognition and storage medium | |
| CN108769831B (en) | Video preview generation method and device | |
| CN110309753A (en) | A kind of race process method of discrimination, device and computer equipment | |
| CN107846615A (en) | Visualize advertisement delivery system, put-on method and advertisement broadcast method | |
| CN112287800A (en) | Advertisement video identification method and system under no-sample condition | |
| CN111581435A (en) | Video cover image generation method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20160516 Address after: 519031 Guangdong city of Zhuhai province Hengqin Baohua Road No. 6, room 105 -478 Patentee after: GUANGDONG ZHONGXING ELECTRONICS CO., LTD. Address before: 100083, Haidian District, Xueyuan Road, Beijing No. 35, Nanjing Ning building, 15 Floor Patentee before: Beijing Vimicro Corporation |