CN102446272A - Method and device for segmenting and recognizing station caption as well as television comprising device - Google Patents
Method and device for segmenting and recognizing station caption as well as television comprising device Download PDFInfo
- Publication number
- CN102446272A CN102446272A CN2011102606309A CN201110260630A CN102446272A CN 102446272 A CN102446272 A CN 102446272A CN 2011102606309 A CN2011102606309 A CN 2011102606309A CN 201110260630 A CN201110260630 A CN 201110260630A CN 102446272 A CN102446272 A CN 102446272A
- Authority
- CN
- China
- Prior art keywords
- station symbol
- pixel
- value
- station
- target image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention is suitable for the field of image processing and provides a method and device for segmenting and recognizing a station caption as well as a television comprising the device. The method comprises the steps of: segmenting the station caption according to change values of characteristic values of continuous multi-frame video image pixel points in a preset sensitive area and obtaining a target image of the station caption; matching the target image of the station caption with preset station captions in a station caption characteristic library; and outputting television station information matched with the target image of the station caption. According to the invention, the station caption is segmented according to the change values of the characteristic values of the pixel points so that the station caption can be segmented under the circumstance of little video image change and the real-time detection and recognition are realized; moreover, the method is widely applied in the fields of real-time monitoring of television station signal, selection and recording of specific television programs, survey and statistics of program rating, advertisement inter cut, illegal inter cut monitoring of radio stations and the like.
Description
Technical field
The invention belongs to image processing field, relate in particular to method, device and televisor that a kind of station symbol is cut apart and discerned.
Background technology
Nowadays, TV has spread all over huge numbers of families, sees that TV becomes the daily pastime of people, amusement institute necessary way.And TV station is also constantly increasing, and all there is different separately station symbols in different TV stations.The station symbol of TV station has comprised important semantic informations such as the platform name, program orientation of a TV station, is the sign of difference mutually between a plurality of TV stations, also is one of important information source of realizing video analysis, understanding and retrieval.Surge along with TV stations at different levels platform order quantity; If need monitor, select that specific television program is included to TV station's signal in real time, when network real-time monitoring and retrieval, program audience rating survey statistics, advertisement insertion and radio station are illegally intercutted and monitored; Because existing systems can not be discerned station symbol automatically; Operate and have only through the mode of manual work, make workload become very huge, work efficiency also reduces thereupon greatly; And because huge, the artificial operation mistakes are hard to avoid of workload.
Summary of the invention
The purpose of the embodiment of the invention is the method, device and the televisor that provide a kind of station symbol to cut apart and discern to be intended to solve the problem that can not discern station symbol now automatically.
The embodiment of the invention is achieved in that the method that a kind of station symbol is cut apart and discerned, and said method comprises:
Changing value according to the eigenwert of continuous multiple frames video image sensitizing range pixel is cut apart station symbol, obtains the station symbol target image, and said sensitizing range is setting in advance;
Said station symbol target image and station symbol in the preset station symbol feature database are mated;
TV station's information that output and said station symbol target image are complementary.
Another purpose of the embodiment of the invention is the device that provides a kind of station symbol to cut apart and discern, and said device comprises:
The station symbol cutting unit is used for cutting apart station symbol according to the changing value of the eigenwert of continuous multiple frames video image sensitizing range pixel, obtains the station symbol target image, and said sensitizing range is setting in advance;
The station symbol matching unit is used for the station symbol of said station symbol target image with the station symbol feature database of presetting mated;
Information output unit is used to export the TV station's information that is complementary with said station symbol target image.
Another purpose of the embodiment of the invention is to provide and comprises that said station symbol is cut apart and the televisor of the device discerned.
In embodiments of the present invention; At first set up the station symbol feature database that comprises all TV station's station symbols; Cut apart station symbol according to the changing value of the eigenwert of the sensitizing range pixel of continuous multiple frames video image then; Obtain the station symbol target image,, know TV station's information of current broadcast at last through coupling station symbol target image and station symbol storehouse.Cut apart owing to the changing value according to the pixel eigenwert carries out station symbol, change under the very little situation at video pictures and also can cut apart station symbol, accomplish to detect in real time identification.And this method to the monitoring of TV station signal in real time, select that specific television program is included, program audience rating survey statistics, advertisement insertion, radio station illegally intercut field extensive application such as monitoring.
Description of drawings
Fig. 1 is that station symbol that first embodiment of the invention provides is cut apart and the process flow diagram of recognition methods;
Fig. 2 is that station symbol that second embodiment of the invention provides is cut apart and the process flow diagram of recognition methods;
Fig. 3 be station symbol that third embodiment of the invention provides cut apart and recognition methods in come the process flow diagram of rounding table calibration method according to the changing value of the eigenwert of continuous multiple frames video image sensitizing range pixel;
Fig. 4 is that station symbol that fourth embodiment of the invention provides is cut apart and the structural representation of the device discerned;
Fig. 5 is that a kind of preferred station symbol of Fig. 4 the 4th embodiment is cut apart and the structural representation of recognition device;
Fig. 6 is that the another kind of preferred station symbol of Fig. 4 the 4th embodiment is cut apart and the structural representation of recognition device.
Embodiment
In order to make the object of the invention, technical scheme and advantage clearer,, the present invention is further elaborated below in conjunction with accompanying drawing and embodiment.Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.
The method that the embodiment of the invention provides a kind of station symbol to cut apart and discern; At first set up the station symbol feature database that comprises all TV station's station symbols; Cut apart station symbol according to the rgb value of the sensitizing range pixel of continuous multiple frames video image or the changing value of gray-scale value then; Obtain the station symbol target image,, know TV station's information of current broadcast at last through coupling station symbol target image and station symbol storehouse.
The invention provides method and device that a kind of station symbol is cut apart and discerned:
Said method comprises:
Changing value according to the eigenwert of continuous multiple frames video image sensitizing range pixel is cut apart station symbol, obtains the station symbol target image, and said sensitizing range is setting in advance;
Said station symbol target image and station symbol in the preset station symbol feature database are mated;
TV station's information that output and said station symbol target image are complementary.
Said device comprises:
The station symbol cutting unit is used for cutting apart station symbol according to the changing value of the eigenwert of continuous multiple frames video image sensitizing range pixel, obtains the station symbol target image, and said sensitizing range is setting in advance;
The station symbol matching unit is used for the station symbol of said station symbol target image with the station symbol feature database of presetting mated;
Information output unit is used to export the TV station's information that is complementary with said station symbol target image.
For technical scheme of the present invention is described, describe through specific embodiment below.
Embodiment one:
The station symbol that being illustrated in figure 1 as first embodiment of the invention provides is cut apart and the process flow diagram of the method discerned; For the ease of explanation; Only show the part relevant, be to be understood that step shown in the figure, just the general step of present embodiment with the embodiment of the invention; Ordinal relation is carried out the routine adjustment, do not influence the enforcement of examples of implementation.
In step S101, set up the station symbol feature database according to the television station channel resource in advance.
In embodiments of the present invention; In advance according to the television station channel resource,, utilize the intercepting of Flame Image Process instrument to comprise the image of station symbol to each television station channel video image of intercepting all; And this image that comprises station symbol carried out the background process of going except that station symbol, only on image, keep station symbol.Set up the station symbol feature database then, the information of all TV station's station symbols that storage obtains.
In embodiments of the present invention, the order of step S101 does not limit.
In step S102, in advance according to the sensitizing range of television video image setting video image.
In embodiments of the present invention, in advance according to the sensitizing range of television video image setting video image.Because the station symbol of TV station is usually located at the upper left corner or the upper right corner of video image, and is wherein, in the majority to be positioned at the upper left corner.Therefore, the upper left corner of video image is set to first sensitizing range in advance, and the upper right corner of video image is set to second sensitizing range.The size of sensitizing range is provided with according to the size of maximum station symbol, can both be in the sensitizing range with the station symbol of guaranteeing all TV stations.Come monitor station target position through detecting the sensitizing range, whether at first detect station symbol in first sensitizing range, if then do not detecting second sensitizing range.
As preferred version of the present invention; For the station symbol that guarantees all TV stations can both be in the sensitizing range; The size that first sensitizing range (being the upper left corner) is set is: begin from coordinate (0,0) that (Width/4-1 Height/4-1) is the determined rectangular area of diagonal line to coordinate; The size that second sensitizing range (being the upper right corner) is set in advance is: begin from coordinate (Width/4 * 3-1,0) that (width-1 Height/4-1) is the determined rectangular area of diagonal line to coordinate.Wherein, Width is the wide of view picture television video image, and Height is the height of view picture television video image, and therefore, the size of two sensitizing ranges is 1/16 of original image.
In step S103, matching times is set attempts counter.
In step S104, cut apart station symbol according to the changing value of the eigenwert of continuous multiple frames video image sensitizing range pixel, obtain the station symbol target image.
In embodiments of the present invention, the eigenwert of said pixel comprises the value or the gray values of pixel points of each component of RGB of pixel.At first; Obtain the maximal value and the minimum value of the eigenwert of continuous multiple frames video image sensitizing range pixel respectively; Cut apart station symbol according to the size of the difference of the maximal value of the eigenwert of sensitizing range pixel and minimum value then; The eigenwert of station symbol pixel is set to 255 after cutting apart, and the eigenwert of background pixel point is set to 0, obtains the station symbol target image.
Because this overwhelming majority in vision signal of station symbol remains static times, so the eigenwert of each pixel of station symbol position has relative stability, and its changing value is little.Therefore, the stool and urine of the changing value of the eigenwert through judging video image sensitizing range pixel can know that this pixel is station symbol pixel or background pixel point, obtains the station symbol target image.Here, preset threshold value is used to detect the changing value of the eigenwert of pixel.When the changing value of the eigenwert of video image sensitizing range pixel during smaller or equal to this threshold value, then this pixel is the station symbol pixel; When the changing value of the eigenwert of video image sensitizing range pixel during greater than this threshold value, then this pixel is the background pixel point.Its concrete realization is as shown in Figure 3, and back extended meeting is elaborated, and does not detail at this.
In step S105; Station symbol target image and station symbol in the preset station symbol feature database are mated, judge whether mate successfully, as if mating successfully then execution in step S106; Otherwise mate number of attempt whether greater than the judgement of optional network specific digit (as 10); Be step S107, if not greater than specific numeral then return execution in step S104, otherwise withdraw from this moment of attempting.
In embodiments of the present invention, because the face shaping of each station symbol all is unique uniqueness, and the image of station symbol can not rotate, and therefore can adopt based on the method for face shaping and mate.
Be specially; The station symbol target image is carried out rim detection; Obtain the station symbol edge detection graph,, adopt moving window search and block division method that station symbol target image and station symbol in the station symbol feature database of presetting are mated contrast then according to the station symbol template; If mate successfully then execution in step S106, otherwise execution in step S107.
Preferably, in order to prevent the skew that causes station symbol zone small number of pixels of stochastic factor such as noise, an ad hoc fixed matching threshold (being assumed to 0.8); If the matching rate after the contrast coupling is less than this matching threshold; Think that then coupling is unsuccessful, proceed to cut apart, if matching rate is greater than this matching threshold; Then think and mate successfully execution in step S106.
In step S106, TV station's information that output and station symbol target image are complementary.
In embodiments of the present invention, the TV station's information that is complementary according to output of the station symbol information in the station symbol storehouse and station symbol target image.Wherein, said TV station information includes but not limited to: information such as platform name, program orientation.
In step S107, whether judge the coupling number of attempt greater than specific numeral, be execution in step S108 then, otherwise execution in step S104.
In step S108, count program exception or mate failure and withdraw from.
In embodiments of the present invention; At first set up the station symbol feature database that comprises all TV station's station symbols; Cut apart station symbol according to the changing value of the eigenwert of the sensitizing range pixel of continuous multiple frames video image then; Obtain the station symbol target image,, know TV station's information of current broadcast at last through coupling station symbol target image and station symbol storehouse.Cut apart owing to the changing value according to the pixel eigenwert carries out station symbol, change under the very little situation at video pictures and also can cut apart station symbol, accomplish to detect in real time identification.This method to the monitoring of TV station signal in real time, select that specific television program is included, program audience rating survey statistics, advertisement insertion, radio station illegally intercut field extensive application such as monitoring.
Embodiment two:
The station symbol that being illustrated in figure 2 as second embodiment of the invention provides is cut apart and the process flow diagram of the method discerned; For the ease of explanation; Only show the part relevant, be to be understood that step shown in the figure, just the general step of present embodiment with the embodiment of the invention; Ordinal relation is carried out the routine adjustment, do not influence the enforcement of examples of implementation.
In step S201, set up the station symbol feature database according to the television station channel resource in advance;
In step S202, the sensitizing range of video image is set according to television video in advance.
In step S203, matching times is set attempts counter.
In step S204, cut apart station symbol according to the changing value of the eigenwert of continuous multiple frames video image sensitizing range pixel, obtain the station symbol target image;
In step S205, whether the mean flow rate of judging the station symbol target image greater than then return carries out the step S204 that obtain station symbol target image greater than luminance threshold, otherwise carries out the step S206 of station symbol coupling.
In embodiments of the present invention, because after station symbol cut apart, having only station symbol in the station symbol target image was high-brightness region, and the brightness value of background area is 0.Therefore, set luminance threshold K,, think that station symbol do not cut apart fully, then return execution in step S204 when the mean flow rate of station symbol target image during greater than this luminance threshold.When the mean flow rate of station symbol target image is not more than this luminance threshold, think that station symbol cuts apart fully, then execution in step S206.
In step S206, station symbol target image and station symbol in the preset station symbol feature database are mated, if mate successfully then execution in step S207, otherwise execution in step S208.
In step S207, TV station's information that output and station symbol target image are complementary.
In step S208, whether judge the coupling number of attempt greater than specific numeral (as 10), be execution in step S209 then, otherwise return step S204.
In step S209, count program exception or mate failure and withdraw from.
In embodiments of the present invention, can directly decide the station symbol image of extraction whether to cut apart fully, save the time of identification station symbol through the mean flow rate of judging the station symbol target image.
Embodiment three:
Be illustrated in figure 3 as among embodiment one that third embodiment of the invention provides and the embodiment two process flow diagram that comes the rounding table calibration method according to the changing value of the eigenwert of continuous multiple frames video image sensitizing range pixel; For the ease of explanation, only show the part relevant with the embodiment of the invention.
In step S301, obtain the maximal value and the minimum value matrix of continuous multiple frames video image sensitizing range pixel eigenwert.
In embodiments of the present invention, if the pixel eigenwert is the value of each component of pixel RGB, then set up six with sensitizing range matrix of the same size, be respectively applied for R, the G of each pixel of storage, the maximal value and the minimum value of B component, that is: R
Max, R
Min, G
Max, G
Min, B
Max, B
MinAfter having set up matrix, with the component value of first each pixel of frame video image initial value as each minute each pixel of moment matrix.
Subsequently, with the R of each pixel of second two field picture, G, B component value respectively with current matrix in R
Max, R
Min, G
Max, G
Min, B
Max, B
MinComponent is compared, if the R of the pixel of second two field picture, G, each component value of B are greater than the R of correspondence position in the current matrix
Max, G
Max, B
MaxComponent value is then with the pixel R in the current matrix
Max, G
Max, B
MaxComponent value is set to pixel R, G, the B component value of correspondence position in second two field picture; If the R of each pixel of second two field picture, G, each component value of B are less than the R of correspondence position in the current matrix
Min, G
Min, B
MinComponent value is then with R in the current matrix
Min, G
Min, B
MinCorresponding component value is set to R, G, the B component value of each pixel of second two field picture; Other situation do not change the maximal value and the minimum value of each pixel R in the current matrix, G, B component.
Be that example specifies with red component R below.
For example: (i, red component R value j) is greater than current red component matrix R to suppose the 2nd two field picture pixel
Max(i, value j) is then with current red component maximal value R for correspondence position
MaxIn the matrix (i, j) component value replaces to the 2nd two field picture (i, j) the R component value of pixel; (i, red color component value j) is less than current red component minimum value matrix R as if the 2nd two field picture pixel
MinIn (i, j) component value is then with R
MinIn (i, j) component value replaces to the 2nd two field picture pixel (i, R component value j).Then, in this way G, B component are done identical processing, each pixel of the 2nd two field picture is done same processing.
Subsequently, successively the image of other frames in the continuous multiple frames video image that is obtained is done identical processing again, finally obtain the maximal value and the minimum value matrix R of continuous multiple frames video image sensitizing range each component of pixel RGB
Max, R
Min, G
Max, G
Min, B
Max, B
Min
In embodiments of the present invention, if the pixel eigenwert is the pixel gray-scale value, then at first, the continuous multiple frames image is carried out gray processing.Wherein, the method for gray processing includes but not limited to: maximal value gray processing method, mean value gray processing method, method of weighted mean etc. do not limit at this.
Wherein:
Maximal value gray processing formula: f (x, y)=Max (R (x, y), G (x, y), B (x, y)),
Mean value gray processing formula: f (x, y)=(R (x, y)+G (x, y)+B (x, y))/3,
Method of weighted mean formula: f (x, y)=m
1* R (x, y)+m
2* G (x, y)+m
3* B (x, y).
Wherein, x, y are the coordinate position of pixel, R (x; Y) be that coordinate is that ((x is that coordinate is (x in the image y) to G for x, the R value of pixel y) in the image; The G value of pixel y), (x is that coordinate is (x, the B value of pixel y) in the image y) to B; (x is that coordinate is (x, gray values of pixel points y), m in the image y) to f
1, m
2, m
3Be respectively preset weight value.
Then, obtain continuous multiple frames video image sensitizing range pixel gray scale maximal value and minimum value matrix again.
Be specially, set up two maximal value and the minimum value that are used to store each pixel gray-scale value with sensitizing range matrix of the same size, that is: g
MaxAnd g
MinWith the initial value of each gray values of pixel points of first two field picture as each pixel point value of minimax gray matrix.
Subsequently; With each gray values of pixel points of second two field picture respectively with current matrix in the maximal value and the minimum value of each pixel gray-scale value compare, if second two field picture (i, j) gray values of pixel points is greater than (i in the current matrix; J) value; Then with in the current matrix in the current matrix (i, value j) is set to (i, j) gray values of pixel points of second two field picture; If second two field picture (i, j) gray values of pixel points less than position in the current matrix (i, value j), then with position in the current matrix (i, value j) is set to (i, j) gray values of pixel points of second two field picture; Other situation do not change the maximal value and the minimum value of each the pixel gray-scale value in the current matrix.
For example: (i, (i is j) greater than current gray level maximal value matrix g for gray-scale value f j) to suppose the 2nd two field picture pixel
Max(i, value j) is then with current g
Max(i, value j) replace to the 2nd two field picture (i, j) gray values of pixel points f (i, j); (i, gray-scale value j) is less than current gray level minimum value matrix g as if the 2nd two field picture pixel f
Min(i, j) point value is then with matrix g
Min(i, j) point value replaces to the 2nd two field picture pixel f (i, pixel value j), otherwise former maximin matrix value remains unchanged.Subsequently, successively the pixel of the image of other frames in the continuous multiple frames video image that is obtained is done identical processing again, finally obtain the maximal value and the minimum value matrix of continuous multiple frames video image sensitizing range pixel gray scale.
In step S302,, obtain the difference of pixel eigenwert maximal value and minimum value according to maximal value and minimum value matrix.
In embodiments of the present invention; When the pixel eigenwert is the value of each component of pixel RGB; Then, each component maximal value is deducted each component minimum value, obtain the difference of each component maximal value of pixel and minimum value according to the maximal value and the minimum value matrix of sensitizing range each component of pixel RGB.
Concrete available following formula obtains:
R
N(x，y)＝|R
max(x，y)-R
min(x，y)|，
G
N(x，y)＝|G
max(x，y)-G
min(x，y)|，
B
N(x，y)＝|B
max(x，y)-B
min(x，y)|，
Wherein, x, y is the coordinate position of pixel, N represents the N two field picture, R
N(x, y), G
N(x, y), B
N(x, y) be respectively before in the N two field picture coordinate be (x, the maximal value of y) pixel R value, G value and B value and the difference of minimum value, R
Min(x, y), G
Min(x, y), B
Min(x is that coordinate is (x, the minimum value of the R value of pixel y), G value and B value, R in the image y)
Max(x, y), G
Max(x, y), B
Max(x is that coordinate is (x, the maximal value of the R value of pixel y), G value and B value in the image y).
In embodiments of the present invention, when the pixel eigenwert was the pixel gray-scale value, then the gray scale maximal value with each pixel deducted minimum gray value, obtained the difference of pixel gray scale maximal value and minimum value.Can obtain through following formula:
g
N(x，y)＝|g
max(x，y)-g
min(x，y)|，
Wherein, x wherein, y is the coordinate position of pixel, N represents the N two field picture, g
N(x, y) coordinate is (x, pixel gray scale maximal value y) and the difference of minimum value, g in the preceding N two field picture of expression
Max(x is (x, the gray scale maximal value of pixel y), g for coordinate y)
Min(x is (x, the minimum gray value of pixel y) for coordinate y).
In step S303; When the difference of the maximal value of pixel eigenwert and minimum value during greater than preset threshold value; This pixel is the background pixel point, and when the difference of the maximal value of pixel eigenwert and minimum value was less than or equal to preset threshold value, this pixel was the station symbol pixel.
In embodiments of the present invention; When the pixel eigenwert is the value of each component of pixel RGB; Because this overwhelming majority time in vision signal of station symbol remains static; Each pixel R of station symbol position, G, B value have relative stability, and therefore, the maximal value of general each component of pixel and the difference of minimum value can not surpass specific value.At this; When the difference of the maximal value of the value of arbitrary component in said each component of pixel RGB and minimum value during greater than preset threshold value; This pixel is the background pixel point; When the difference of the maximal value of the said important value of pixel RGB and minimum value was less than or equal to preset threshold value, this pixel was the station symbol pixel.Just; As long as the difference of maximal value and minimum value that any one-component is arranged in the pixel three primary colours is greater than preset threshold value; Assert that then this pixel is the background pixel point, its brightness value is changed to 0, the station symbol target image that obtains with preceding frame then carries out the logical and operation; Have only when in the pixel three primary colours the difference of important maximal value and minimum value when all being not more than preset threshold value, assert that this pixel is the station symbol pixel, the brightness value of station symbol pixel is set to 255.Expression formula can be as follows:
Wherein, x, y is the coordinate position of pixel, N represents the N two field picture, R
N(x, y), G
N(x, y), B
N(x, y) be respectively before in the N two field picture coordinate for (M is a preset threshold value for x, pixel maximal value y) and the difference of minimum value.
In embodiments of the present invention; When the pixel eigenwert is the pixel gray-scale value; In embodiments of the present invention, if the difference of pixel gray scale maximal value and minimum value assert then that greater than preset threshold value this pixel is the background pixel point; Its brightness value is set to 0, and the station symbol target image that obtains with preceding frame then carries out the logical and operation; When the difference of pixel gray scale maximal value and minimum value is not more than preset threshold value, assert that this pixel is the station symbol pixel, the brightness value of station symbol pixel is set to 255.Expression formula can be as follows:
Wherein, x, y is the coordinate position of pixel, N represents the N two field picture, g
N(x, y) coordinate is that (M is a preset threshold value for x, pixel gray scale maximal value y) and the difference of minimum value in the preceding N two field picture of expression.
Embodiment four:
The station symbol that Fig. 4 provides for the embodiment of the invention is cut apart and the structural representation of the device discerned, for the ease of explanation, only shows the part relevant with the embodiment of the invention.This device can be to be built in the unit that software unit, hardware cell or software and hardware in the televisor combine, and perhaps is integrated in the application system of televisor as suspension member independently.Wherein:
Station symbol cutting unit 43 is used for cutting apart station symbol according to the changing value of continuous multiple frames video image sensitizing range pixel eigenwert, obtains the station symbol target image.
In embodiments of the present invention, the eigenwert of said pixel comprises: the value or the gray values of pixel points of each component of RGB of pixel.
In embodiments of the present invention, said station symbol cutting unit 43 comprises three modules, is respectively: matrix acquisition module 431, difference acquisition module 432 and station symbol are cut apart module 433.Wherein:
Station symbol is cut apart module 433; When being used for difference when the maximal value of pixel eigenwert and minimum value greater than preset threshold value; This pixel is the background pixel point, and when the difference of the maximal value of pixel eigenwert and minimum value was less than or equal to preset threshold value, this pixel was the station symbol pixel.
Station symbol matching unit 44 is used for the station symbol of station symbol target image with the station symbol feature database of presetting mated.
In embodiments of the present invention, station symbol matching unit 44 also comprises a module station symbol matching module 441, wherein:
Station symbol matching module 441 is used for the station symbol target image is carried out rim detection, obtains the station symbol edge detection graph, then according to the station symbol template, station symbol target image and station symbol in the preset station symbol feature database is mated contrast.
Preferably; Said device also comprises brightness judging unit 46, and said brightness judging unit 46 carries out brightness according to the station symbol target image after cutting apart and judges after station symbol cutting unit 43 is cut apart station symbol; Then, the result after station symbol matching unit 44 is judged according to brightness carries out the station symbol coupling.As shown in Figure 5:
Preferably, said device comprises that also counter is provided with unit 47, and said counter is provided with unit 47, and is as shown in Figure 6:
Counter is provided with unit 47, is used to be provided with station symbol matching unit 44 and attempts the counting of matching times, when number of attempt greater than specific numeral (as 10) time, counts program exception or the coupling failure is withdrawed from.In embodiments of the present invention; At first set up the station symbol feature database that comprises all TV station's station symbols; Cut apart station symbol according to the changing value of the eigenwert of the sensitizing range pixel of continuous multiple frames video image then; Obtain the station symbol target image,, know TV station's information of current broadcast at last through coupling station symbol target image and station symbol storehouse.Cut apart owing to the changing value according to the pixel eigenwert carries out station symbol, change under the very little situation at video pictures and also can cut apart station symbol, accomplish to detect in real time identification.And this method to the monitoring of TV station signal in real time, select that specific television program is included, program audience rating survey statistics, advertisement insertion, radio station illegally intercut field extensive application such as monitoring.
One of ordinary skill in the art will appreciate that; Realize that all or part of step in the foregoing description method is to instruct relevant hardware to accomplish through program; Described program can be in being stored in a computer read/write memory medium; Described storage medium is like ROM/RAM, disk, CD etc.
The above is merely preferred embodiment of the present invention, not in order to restriction the present invention, all any modifications of within spirit of the present invention and principle, being done, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.
Claims (11)
1. the method that station symbol is cut apart and discerned is characterized in that, said method comprises the steps:
Changing value according to the eigenwert of continuous multiple frames video image sensitizing range pixel is cut apart station symbol, obtains the station symbol target image, and said sensitizing range is setting in advance;
Said station symbol target image and station symbol in the preset station symbol feature database are mated;
TV station's information that output and said station symbol target image are complementary.
2. the method for claim 1 is characterized in that, the eigenwert of said pixel comprises: the value or the gray values of pixel points of each component of RGB of pixel.
3. according to claim 1 or claim 2 method is characterized in that, the changing value of said eigenwert according to continuous multiple frames video image sensitizing range pixel is cut apart station symbol, obtains the step of station symbol target image, is specially:
Obtain the maximal value and the minimum value matrix of continuous multiple frames video image sensitizing range pixel eigenwert;
According to the maximal value and the minimum value matrix of said eigenwert, obtain the maximal value of pixel eigenwert and the difference of minimum value;
When the difference of the maximal value of pixel eigenwert and minimum value during greater than preset threshold value, this pixel is the background pixel point, and when the difference of the maximal value of pixel eigenwert and minimum value was less than or equal to preset threshold value, this pixel was the station symbol pixel.
4. the method for claim 1; It is characterized in that; Saidly cut apart station symbol according to the rgb value of continuous multiple frames video image sensitizing range pixel or the changing value of gray-scale value; Obtain after the station symbol target image, said station symbol in said station symbol target image and the station symbol feature database is mated before, also comprise:
Whether the mean flow rate of judging the station symbol target image is greater than luminance threshold;
If the mean flow rate of station symbol target image is then mated the station symbol in said station symbol target image and the station symbol feature database greater than luminance threshold;
If the mean flow rate of station symbol target image is then returned according to the rgb value of continuous multiple frames video image sensitizing range pixel or the changing value of gray-scale value and cut apart station symbol smaller or equal to luminance threshold, obtain the station symbol target image.
5. the method for claim 1 is characterized in that, the said step that said station symbol target image and station symbol in the preset station symbol feature database are mated is specially:
Said station symbol target image is carried out rim detection, obtain the station symbol edge detection graph,, said station symbol target image and station symbol in the preset station symbol feature database are mated contrast then according to the station symbol template.
6. the device that station symbol is cut apart and discerned is characterized in that, said device comprises:
The station symbol cutting unit is used for cutting apart station symbol according to the changing value of the eigenwert of continuous multiple frames video image sensitizing range pixel, obtains the station symbol target image, and said sensitizing range is setting in advance;
The station symbol matching unit is used for the station symbol of said station symbol target image with the station symbol feature database of presetting mated;
Information output unit is used to export the TV station's information that is complementary with said station symbol target image.
7. device as claimed in claim 6 is characterized in that, the eigenwert of said pixel comprises: the value or the gray values of pixel points of each component of RGB of pixel.
8. device as claimed in claim 6 is characterized in that, said station symbol cutting unit comprises:
The matrix acquisition module is used to obtain the maximal value and the minimum value matrix of continuous multiple frames video image sensitizing range pixel eigenwert;
The difference acquisition module is used for maximal value and minimum value matrix according to said eigenwert, obtains the maximal value of pixel eigenwert and the difference of minimum value;
Station symbol is cut apart module; When being used for difference when the maximal value of pixel eigenwert and minimum value greater than preset threshold value; This pixel is the background pixel point, and when the difference of the maximal value of pixel eigenwert and minimum value was less than or equal to preset threshold value, this pixel was the station symbol pixel.
9. device as claimed in claim 6 is characterized in that, said station symbol matching unit also comprises:
The station symbol matching module is used for said station symbol target image is carried out rim detection, obtains the station symbol edge detection graph, then said station symbol target image and station symbol in the preset station symbol feature database is mated contrast.
10. device as claimed in claim 6 is characterized in that, said device also comprises:
The brightness judge module; Whether the mean flow rate that is used to judge the station symbol target image is greater than luminance threshold; If the mean flow rate of station symbol target image is greater than luminance threshold, then the station symbol matching unit matees station symbol target image and station symbol in the preset station symbol feature database, if the mean flow rate of station symbol target image is less than or equal to luminance threshold; Then the station symbol cutting unit is cut apart station symbol according to the changing value of continuous multiple frames video image sensitizing range pixel eigenwert, obtains the station symbol target image.
11. a televisor is characterized in that, described televisor comprises the device that the described station symbol of the arbitrary claim of claim 6-10 is cut apart and discerned.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102606309A CN102446272A (en) | 2011-09-05 | 2011-09-05 | Method and device for segmenting and recognizing station caption as well as television comprising device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011102606309A CN102446272A (en) | 2011-09-05 | 2011-09-05 | Method and device for segmenting and recognizing station caption as well as television comprising device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102446272A true CN102446272A (en) | 2012-05-09 |
Family
ID=46008761
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011102606309A Pending CN102446272A (en) | 2011-09-05 | 2011-09-05 | Method and device for segmenting and recognizing station caption as well as television comprising device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102446272A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103024473A (en) * | 2013-01-05 | 2013-04-03 | 张荣华 | Audience rating survey system and method |
| CN103258187A (en) * | 2013-04-16 | 2013-08-21 | 华中科技大学 | Television station caption identification method based on HOG characteristics |
| CN103455819A (en) * | 2013-08-19 | 2013-12-18 | 四川长虹电器股份有限公司 | Method for recognizing programs broadcasted by set top box of smart television |
| CN103530598A (en) * | 2013-03-08 | 2014-01-22 | Tcl集团股份有限公司 | Station logo identification method and system |
| CN103544467A (en) * | 2013-04-23 | 2014-01-29 | Tcl集团股份有限公司 | Method and device for detecting and recognizing station captions |
| CN104063701A (en) * | 2014-07-14 | 2014-09-24 | 电子科技大学 | Rapid television station caption recognition system based on SURF vocabulary tree and template matching and implementation method of rapid television station caption recognition system |
| CN104185069A (en) * | 2013-05-23 | 2014-12-03 | 北京下周科技有限公司 | Station icon identification method and identification system |
| CN105049878A (en) * | 2015-07-08 | 2015-11-11 | 无锡天脉聚源传媒科技有限公司 | Processing method of approximate TV logo and device |
| CN105282599A (en) * | 2015-10-21 | 2016-01-27 | 天脉聚源（北京）科技有限公司 | Method and device for determining television channels |
| CN106067006A (en) * | 2016-06-07 | 2016-11-02 | 东软集团股份有限公司 | For the method and apparatus identifying news program |
| CN106446889A (en) * | 2015-08-10 | 2017-02-22 | Tcl集团股份有限公司 | Local identification method and local identification device for station logo |
| CN103634653B (en) * | 2013-11-06 | 2017-03-01 | 小米科技有限责任公司 | TV station symbol recognition method, device and television set |
| WO2017088463A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Image processing method and device |
| WO2017088479A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Method of identifying digital on-screen graphic and device |
| CN106851397A (en) * | 2017-02-28 | 2017-06-13 | 青岛海信电器股份有限公司 | A kind of station symbol replacing options and device |
| CN108012160A (en) * | 2016-10-31 | 2018-05-08 | 央视国际网络无锡有限公司 | A kind of logo insertion method based on panoramic video |
| CN108093246A (en) * | 2017-11-21 | 2018-05-29 | 青岛海信电器股份有限公司 | A kind of recognition methods of top box of digital machine video playback area and device |
| CN109977859A (en) * | 2019-03-25 | 2019-07-05 | 腾讯科技（深圳）有限公司 | A kind of map logo method for distinguishing and relevant apparatus |
| CN111160215A (en) * | 2019-12-25 | 2020-05-15 | Tcl华星光电技术有限公司 | Image identifier extraction device, and brightness adjustment method and device of image identifier |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216841A (en) * | 2008-01-14 | 2008-07-09 | 南京搜拍信息技术有限公司 | Interactive type image search system and method |
| CN101295405A (en) * | 2008-06-13 | 2008-10-29 | 西北工业大学 | Portrait and vehicle recognition alarming and tracing method |
| US20090016603A1 (en) * | 2005-12-30 | 2009-01-15 | Telecom Italia S.P.A. | Contour Finding in Segmentation of Video Sequences |
-
2011
- 2011-09-05 CN CN2011102606309A patent/CN102446272A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090016603A1 (en) * | 2005-12-30 | 2009-01-15 | Telecom Italia S.P.A. | Contour Finding in Segmentation of Video Sequences |
| CN101216841A (en) * | 2008-01-14 | 2008-07-09 | 南京搜拍信息技术有限公司 | Interactive type image search system and method |
| CN101295405A (en) * | 2008-06-13 | 2008-10-29 | 西北工业大学 | Portrait and vehicle recognition alarming and tracing method |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103024473A (en) * | 2013-01-05 | 2013-04-03 | 张荣华 | Audience rating survey system and method |
| CN103530598B (en) * | 2013-03-08 | 2017-02-08 | Tcl集团股份有限公司 | Station logo identification method and system |
| CN103530598A (en) * | 2013-03-08 | 2014-01-22 | Tcl集团股份有限公司 | Station logo identification method and system |
| CN103258187A (en) * | 2013-04-16 | 2013-08-21 | 华中科技大学 | Television station caption identification method based on HOG characteristics |
| CN103544467A (en) * | 2013-04-23 | 2014-01-29 | Tcl集团股份有限公司 | Method and device for detecting and recognizing station captions |
| CN103544467B (en) * | 2013-04-23 | 2017-09-05 | Tcl集团股份有限公司 | Method for distinguishing and its device are known in a kind of station symbol detection |
| CN104185069A (en) * | 2013-05-23 | 2014-12-03 | 北京下周科技有限公司 | Station icon identification method and identification system |
| CN104185069B (en) * | 2013-05-23 | 2018-03-16 | 北京下周科技有限公司 | A kind of TV station symbol recognition method and its identifying system |
| CN103455819A (en) * | 2013-08-19 | 2013-12-18 | 四川长虹电器股份有限公司 | Method for recognizing programs broadcasted by set top box of smart television |
| CN103455819B (en) * | 2013-08-19 | 2017-06-23 | 四川长虹电器股份有限公司 | It is a kind of to realize the method that intelligent TV set plays program identification to Set Top Box |
| CN103634653B (en) * | 2013-11-06 | 2017-03-01 | 小米科技有限责任公司 | TV station symbol recognition method, device and television set |
| CN104063701B (en) * | 2014-07-14 | 2017-07-28 | 电子科技大学 | Fast electric television stations TV station symbol recognition system and its implementation based on SURF words trees and template matches |
| CN104063701A (en) * | 2014-07-14 | 2014-09-24 | 电子科技大学 | Rapid television station caption recognition system based on SURF vocabulary tree and template matching and implementation method of rapid television station caption recognition system |
| CN105049878B (en) * | 2015-07-08 | 2018-07-20 | 无锡天脉聚源传媒科技有限公司 | A kind for the treatment of method and apparatus of approximation station symbol |
| CN105049878A (en) * | 2015-07-08 | 2015-11-11 | 无锡天脉聚源传媒科技有限公司 | Processing method of approximate TV logo and device |
| CN106446889A (en) * | 2015-08-10 | 2017-02-22 | Tcl集团股份有限公司 | Local identification method and local identification device for station logo |
| CN106446889B (en) * | 2015-08-10 | 2019-09-17 | Tcl集团股份有限公司 | A kind of local recognition methods of logo and device |
| CN105282599A (en) * | 2015-10-21 | 2016-01-27 | 天脉聚源（北京）科技有限公司 | Method and device for determining television channels |
| CN105282599B (en) * | 2015-10-21 | 2018-05-08 | 天脉聚源（北京）科技有限公司 | A kind of method and device of definite television channel |
| WO2017088479A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Method of identifying digital on-screen graphic and device |
| WO2017088463A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Image processing method and device |
| CN106067006A (en) * | 2016-06-07 | 2016-11-02 | 东软集团股份有限公司 | For the method and apparatus identifying news program |
| CN108012160A (en) * | 2016-10-31 | 2018-05-08 | 央视国际网络无锡有限公司 | A kind of logo insertion method based on panoramic video |
| CN108012160B (en) * | 2016-10-31 | 2019-07-23 | 央视国际网络无锡有限公司 | A kind of logo insertion method based on panoramic video |
| CN106851397A (en) * | 2017-02-28 | 2017-06-13 | 青岛海信电器股份有限公司 | A kind of station symbol replacing options and device |
| CN108093246A (en) * | 2017-11-21 | 2018-05-29 | 青岛海信电器股份有限公司 | A kind of recognition methods of top box of digital machine video playback area and device |
| CN108093246B (en) * | 2017-11-21 | 2020-04-28 | 青岛海信电器股份有限公司 | Method and device for identifying video playing area of digital set top box |
| CN109977859A (en) * | 2019-03-25 | 2019-07-05 | 腾讯科技（深圳）有限公司 | A kind of map logo method for distinguishing and relevant apparatus |
| CN111160215A (en) * | 2019-12-25 | 2020-05-15 | Tcl华星光电技术有限公司 | Image identifier extraction device, and brightness adjustment method and device of image identifier |
| CN111160215B (en) * | 2019-12-25 | 2024-01-12 | Tcl华星光电技术有限公司 | Brightness regulator and method for image mark |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102446272A (en) | Method and device for segmenting and recognizing station caption as well as television comprising device | |
| CN108769776B (en) | Title subtitle detection method and device and electronic equipment | |
| CN106778996B (en) | It is embedded with the generation system and method for the two dimensional code of visual pattern and reads system | |
| WO2021012837A1 (en) | Method and apparatus for determining recommendation information implantation position, device and storage medium | |
| US9076071B2 (en) | Logo recognition | |
| US8355079B2 (en) | Temporally consistent caption detection on videos using a 3D spatiotemporal method | |
| CN109740444B (en) | People flow information display method and related product | |
| CN101360193A (en) | Video subtitle processing apparatus and method | |
| US20070242088A1 (en) | Method for intelligently displaying sports game video for multimedia mobile terminal | |
| CN101295398A (en) | Image processing apparatus, image processing method, and program | |
| CN109543770A (en) | Dot character recognition methods and device | |
| US20100158362A1 (en) | Image processing | |
| CN103369205A (en) | Mobile terminal and video image processing method | |
| CN105657547A (en) | Detection method and device for similar video and pirated video | |
| CN102333174A (en) | Video image processing method and device for the same | |
| CN100593948C (en) | Method and device for jointing video | |
| CN102301697A (en) | Video identifier creation device | |
| CN110769225A (en) | Projection area obtaining method based on curtain and projection device | |
| CN106507191B (en) | It is synchronous to relay content monitoring method and device | |
| CN115937267B (en) | Target track tracking method based on multi-frame video | |
| US10740922B2 (en) | Image data processing method, electronic device and storage medium | |
| CN108763491B (en) | Picture processing method and device and terminal equipment | |
| CN107222746B (en) | Video Key location positioning method and device | |
| CN107256137B (en) | Picture processing method and device | |
| CN110780832A (en) | Picture adjusting method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120509 |