A kind of determination method and device interacting footmark promptTechnical field
The present invention relates to interaction footmark detection technique field, in particular to it is a kind of interact footmark prompt determination method andDevice.
Background technique
Currently, when interaction footmark occurs, can receive the interactive signal of televiewer during program interaction, and it isPreferably confirm interactive signal, needs to carry out detection identification to the interaction footmark occurred in video screen in time.Sometimes, it isThe needs of interaction carry out angle after countdown appearance interaction footmark will appear prompt (such as: 5s falls to count) before occurring when againMark detection, it is possible to reduce footmark erroneous detection, therefore, carrying out detection identification to the prompt of interaction footmark has good meaning, butThe position of prompt on the tv screen may be not fixed, and television video is complicated, thus increase the detection difficulty of prompt.
Summary of the invention
The present invention provides a kind of determination method and device for interacting footmark prompt, can be in current video frame automaticallyColor detection is carried out, after determining the color of object connected region in the current video frame, by identifying that the color of object connectsMark in logical region, can determine the confidence level of the mark and the default prompt, and then judge whether the confidence level is greater thanThe default confidence level illustrates the mark identified and the default prompt if the two confidence level is greater than the default confidence levelThe similarity of symbol is higher, then can automatically determine in color of object connected region and be identified as default prompt, this also illustrates to work asDefault prompt is had already appeared in the current video frame in preceding broadcasting program, which interaction angle will occursMark, user will participate in interacting.
The present invention provides a kind of determination method for interacting footmark prompt, comprising:
According to the pre-set color template of the corresponding default prompt of interaction footmark, determines in current video frame and preset with describedThe color of object connected region that color template matches；
It is default to judge whether the confidence level of mark and the default prompt in the color of object connected region is greater thanConfidence level；
When the confidence level is greater than the default confidence level, determines in the color of object connected region and be identified as instituteState default prompt.
In one embodiment, the default prompt includes: at least one countdown number；
Whether the mark judged in the color of object connected region and the confidence level of the default prompt are greater thanDefault confidence level, comprising:
Using the convolutional neural networks pre-established, determine mark in the color of object connected region and it is described at leastThe confidence level of each countdown number in one countdown number；
It is selected from least one described countdown number and the highest target number of the confidence level of the mark；
Determine whether the corresponding confidence level of the target number is greater than the default confidence level；
It is described when the confidence level be greater than the default confidence level when, determine the mark in the color of object connected regionFor the default prompt, comprising:
When the corresponding confidence level of the target number is greater than the default confidence level, the color of object connected region is determinedThe target number is identified as in domain.
In one embodiment, determine in the color of object connected region be identified as the target number before,The method also includes:
By the default face of the first pixel value of each pixel and the target number in the color of object connected regionSecond pixel value of color template is compared；
According to comparison result, the pixel value of all pixels point in the color of object connected region is subjected to binaryzation；
According to the pixel value of the pixel in the color of object connected region after binaryzation and the target numberThe pixel value of pixel in Character mother plate is determined and is identified whether in the color of object connected region as the number of targetsWord.
In one embodiment, described according to comparison result, by all pixels point in the color of object connected regionPixel value carries out binaryzation, comprising:
Judge the default of the first pixel value of each pixel and the target number in the color of object connected regionWhether the pixel value difference of the second pixel value of color template is greater than or equal to presetted pixel threshold value；
By the picture in the color of object connected region with the second pixel value of the pre-set color template of the target numberThe first pixel value that plain difference is less than or equal to the pixel of the presetted pixel threshold value is set as first object pixel value；
By the picture in the color of object connected region with the second pixel value of the pre-set color template of the target numberThe first pixel value that plain difference is greater than the pixel of the presetted pixel value threshold value is set as the second target pixel value.
In one embodiment, the pixel of the pixel in the color of object connected region according to after binaryzationThe pixel value of value and the pixel in the Character mother plate of the target number, determines the mark in the color of object connected regionIt whether is the target number, comprising:
Pixel in the color of object connected region after determining binaryzation with corresponding position in the Character mother plateThe numbers of the equal target pixel points of pixel value whether be greater than preset number；
When the number of the target pixel points is greater than the preset number, determine in the color of object connected regionIt is identified as the target number.
The present invention also provides a kind of determining devices for interacting footmark prompt, comprising:
First determining module determines current for the pre-set color template according to the corresponding default prompt of interaction footmarkThe color of object connected region to match in video frame with the pre-set color template；
Judgment module, for judging mark and the confidence level of the default prompt in the color of object connected regionWhether default confidence level is greater than；
Second determining module, for determining that the color of object connects when the confidence level is greater than the default confidence levelThe default prompt is identified as in logical region.
In one embodiment, the default prompt includes: at least one countdown number；
The judgment module includes:
First determines that submodule determines the color of object connected region for using the convolutional neural networks pre-establishedThe confidence level of mark and each countdown number at least one described countdown number in domain；
Submodule is selected, for selecting the confidence level highest with the mark from least one described countdown numberTarget number；
Second determines submodule, for determining whether the corresponding confidence level of the target number is greater than the default confidenceDegree；
Second determining module includes:
Third determines submodule, is used for when the corresponding confidence level of the target number is greater than the default confidence level, reallyThe target number is identified as in the fixed color of object connected region.
In one embodiment, described device further include:
Comparison module, for determine in the color of object connected region be identified as the target number before, willThe of the pre-set color template of the first pixel value and target number of each pixel in the color of object connected regionTwo pixel values are compared；
Processing module is used for according to comparison result, by the pixel value of all pixels point in the color of object connected regionCarry out binaryzation；
Third determining module, for the pixel value according to the pixel in the color of object connected region after binaryzationWith the pixel value of the pixel in the Character mother plate of the target number, determine that the mark in the color of object connected region isNo is the target number.
In one embodiment, the processing module includes:
Judging submodule, for judge in the color of object connected region the first pixel value of each pixel with it is describedWhether the pixel value difference of the second pixel value of the pre-set color template of target number is greater than or equal to presetted pixel threshold value；
First setting submodule, for by the pre-set color mould in the color of object connected region with the target numberThe first pixel value that the pixel value difference of second pixel value of plate is less than or equal to the pixel of the presetted pixel threshold value is set asFirst object pixel value；
Second setting submodule, for by the pre-set color mould in the color of object connected region with the target numberThe first pixel value that the pixel value difference of second pixel value of plate is greater than the pixel of the presetted pixel value threshold value is set as secondTarget pixel value.
In one embodiment, the third determining module includes:
4th determines submodule, for determine in the color of object connected region after binaryzation with the Character mother plateWhether the number of the equal target pixel points of the pixel value of the pixel of middle corresponding position is greater than preset number；
5th determines submodule, described in determining when the number of the target pixel points is greater than the preset numberThe target number is identified as in color of object connected region.
The technical scheme provided by this disclosed embodiment can include the following benefits:
Color detection can be automatically carried out in the current video frame of current playing program, determine the current video frameIn color of object connected region after, by identifying the mark in the color of object connected region, can determine the mark withThe confidence level of the default prompt, and then judge whether the confidence level is greater than the default confidence level, if the two confidence level is greater thanThe default confidence level then illustrates that the mark identified and the similarity of the default prompt are higher, then can automatically determine meshIt is identified as default prompt in mark color-connected regions, this also illustrates in the current video frame in current playing programThere is default prompt, which interaction footmark will occurs, and user will participate in interacting.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specificationIt obtains it is clear that understand through the implementation of the invention.The objectives and other advantages of the invention can be by written explanationSpecifically noted structure is achieved and obtained in book, claims and attached drawing.
Below by drawings and examples, technical scheme of the present invention will be described in further detail.
Detailed description of the invention
Attached drawing is used to provide further understanding of the present invention, and constitutes part of specification, with reality of the inventionIt applies example to be used to explain the present invention together, not be construed as limiting the invention.In the accompanying drawings:
Fig. 1 is a kind of flow chart of determination method for interacting footmark prompt shown according to an exemplary embodiment.
Fig. 2 is the flow chart of the determination method of another interaction footmark prompt shown according to an exemplary embodiment.
Fig. 3 is the flow chart of the determination method of another interaction footmark prompt shown according to an exemplary embodiment.
Fig. 4 is a kind of block diagram of determining device for interacting footmark prompt shown according to an exemplary embodiment.
Fig. 5 is the block diagram of the determining device of another interaction footmark prompt shown according to an exemplary embodiment.
Fig. 6 is the block diagram of the determining device of another interaction footmark prompt shown according to an exemplary embodiment.
Fig. 7 is the block diagram of the determining device of another interaction footmark prompt shown according to an exemplary embodiment.
Fig. 8 is the block diagram of the determining device of another interaction footmark prompt shown according to an exemplary embodiment.
Specific embodiment
Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings, it should be understood that preferred reality described hereinApply example only for the purpose of illustrating and explaining the present invention and is not intended to limit the present invention.
In the related technology, during program interaction, when interaction footmark occurs, it can receive the interaction letter of televiewerNumber, and in order to preferably confirm interactive signal, it needs to carry out detection identification to the interaction footmark occurred in video screen in time.Sometimes, for the needs of interaction, when will appear prompt (such as: 5s falls to count) before interacting footmark appearance, after countdown appearanceFootmark detection is carried out again, it is possible to reduce footmark erroneous detection, therefore, carrying out detection identification to the prompt of interaction footmark has wellMeaning, but the position of prompt on the tv screen may be not fixed, and television video is complicated, thus increase the inspection of promptSurvey difficulty.
In order to solve the above-mentioned technical problem, the embodiment of the present disclosure provides a kind of determination method for interacting footmark prompt,For this method suitable for the determination program, system or the device that interact footmark prompt, executing subject can be smart television, handThe equipment that machine, computer etc. can play video, as shown in Figure 1,
Step S101 is determined in current video frame according to the pre-set color template of the corresponding default prompt of interaction footmarkThe color of object connected region to match with the pre-set color template；
Interaction footmark can be any one program of any one TV station for interacting footmark with user interaction(being such as able to use family to participate in the prize drawing footmark of prize drawing, be able to use family and participate in shaking movable movable footmark), and the interactionThe corresponding default prompt of footmark is usually countdown number, such as 5,4,3,2,1, is also possible to the use such as text, figure, letterWill occur in prompt user interaction footmark, the prompt of interactive process can be entered at once.
In addition, when color in automatic identification current video frame, the face with the pre-set color template that system identification goes outThe identical region of color base sheet (the lesser area of pixel difference of the pixel value of pixel value and pre-set color template i.e. in current video frameDomain) it may be multiple zone of dispersion, at this point it is possible to the multiple zone of dispersion being closer are connected to, to obtain and default faceThe color of object connected region that color template matches after being connected to multiple zone of dispersion, is still likely to be obtained multiple mesh certainlyColor-connected regions are marked, and since the size for presetting the region of prompt is certain, therefore, it is possible to which it is biggish to remove areaConnected region and the lesser connected region of area, thus by being connected to color of object similar in the size in the region of default promptCandidate region of the region as prompt.
Step S102 judges identifying in the color of object connected region and whether the confidence level of the default promptGreater than default confidence level；
Color detection is being carried out, it, should by identification after determining the color of object connected region in the current video frameMark in color of object connected region, can determine the confidence level of the mark and the default prompt, and then judge the confidenceWhether degree is greater than the default confidence level.
Step S103 is determined in the color of object connected region when the confidence level is greater than the default confidence levelBe identified as the default prompt.
If the two confidence level is greater than the default confidence level, illustrate the phase of the mark identified with the default promptIt is higher like spending, then it can automatically determine in the color of object connected region and be identified as the default prompt, this also illustratesDefault prompt is had already appeared in the current video frame in current playing program, which will interactFootmark, user will participate in interacting, wherein confidence level is higher, indicates that the mark and the similarity of the default prompt are got overHeight, the mark in the color of object connected region are exactly that the definitive result of the default prompt is more accurate.
In one embodiment, the default prompt includes: at least one countdown number；
The default prompt can be at least one countdown number.
As shown in Fig. 2, above-mentioned steps S102 can be performed as:
Step A1, using the convolutional neural networks pre-established, determine mark in the color of object connected region withThe confidence level of each countdown number at least one described countdown number；
Convolutional neural networks are that the preparatory video flowing training using largely comprising at least one countdown number forms,In training process, countdown number how many kind, the output layer of convolutional neural networks can have it is several layers of, in this manner it is possible to makeWith the convolutional neural networks, the mark in different target color-connected regions is identified, to determine that the mark is respectivelyThe confidence level of each countdown number.
Step A2 is selected and the highest number of targets of the confidence level of the mark from least one described countdown numberWord；
Step A3, determines whether the corresponding confidence level of the target number is greater than the default confidence level；
Since the confidence level that the mark is each countdown number respectively is different, and the more high corresponding meter of confidence levelWhen number, more may be the legitimate reading of the mark, such as: at least one countdown number is 5,4,3,2,1 respectively, and shouldIt is respectively 94%, 2%, 2%, 1%, 1% that mark, which is 5,4,3,2,1 confidence level respectively, then the mark, which is most likely to be, countsWhen number 5, therefore, it is possible to from least one countdown number, select with the highest target number of the confidence level of the mark,And the mark is accidentally identified as the target number in order to prevent, and substantially ensure the recognition correct rate of the mark, it can further reallyWhether fixed the corresponding confidence level of the target number (i.e. the confidence level of the target number and the mark) is greater than the default confidence level (such as0.9)。
Above-mentioned steps S103 can be performed as:
Step A4 determines the target face when the corresponding confidence level of the target number is greater than the default confidence levelThe target number is identified as in color connected region.
When the corresponding confidence level of the target number is greater than the default confidence level, it can accurately determine that the color of object is connected toMark in region is exactly the target number, further relates to have already appeared in the current video frame in the current playing programThe default prompt will occur interacting footmark, user will be able to carry out interaction.
As shown in figure 3, in one embodiment, determining the corresponding confidence level of target number greater than the default confidence levelLater, and determine in the color of object connected region be identified as the target number before, the method also includes:
Step S301, by the first pixel value of each pixel in the color of object connected region and the target numberThe second pixel value of pre-set color template be compared；
Step S302 carries out the pixel value of all pixels point in the color of object connected region according to comparison resultBinaryzation；
Step S303, according to the pixel value of the pixel in the color of object connected region after binaryzation and the meshThe pixel value for marking the pixel in the Character mother plate of number, determines in the color of object connected region identify whether to be describedTarget number.
Determine in the color of object connected region be identified as the target number before, in order to further prevent accidentally shouldIt is identified as the target number, it can also be by the first pixel value of pixel each in color of object connected region and the targetThe second pixel value in the pre-set color template of number is compared, and then by all pixels point in color of object connected regionPixel value carries out binaryzation after being normalized again, it can by the pre-set color in color of object connected region with target numberFirst pixel value of pixel similar in template is set as white pixel value, and color differs the first pixel value of farther away pixelIt is set as black pixel value, and then according to the pixel value and number of targets of the pixel in the color of object connected region after binaryzationThe pixel value of pixel in the Character mother plate of word is further accurately determined in the color of object connected region and identifies whetherFor the target number, wherein the pixel value of the pixel in Character mother plate is also by binaryzation.
In one embodiment, above-mentioned steps S302 can be performed as:
Judge the default of the first pixel value of each pixel and the target number in the color of object connected regionWhether the pixel value difference of the second pixel value of color template is greater than or equal to presetted pixel threshold value；Wherein, presetted pixel threshold value canIt is even more small to be 10pixels etc., wherein presetted pixel threshold value is smaller, and the color of object connected region after binaryzation is more smartIt is quasi-, wherein the pre-set color template of target number is often monochromatic, thus the second of the pre-set color template of the target numberPixel value can be a fixed pixel value, be also possible to a small pixel value section.
By the picture in the color of object connected region with the second pixel value of the pre-set color template of the target numberThe first pixel value that plain difference is less than or equal to the pixel of the presetted pixel threshold value is set as first object pixel value；
By in the color of object connected region with respective pixel point in the pre-set color template of the target numberThe first pixel value that the pixel value difference of two pixel values is greater than the pixel of the presetted pixel value threshold value is set as the second target pictureElement value.
Wherein, first object pixel value can be the pixel value i.e. 255pixels of white, and the second target pixel value can bePixel value, that is, 0pixels of black.
By will be close (i.e. with the second pixel value of pre-set color template of the target number in color of object connected regionPixel value difference is less than or equal to presetted pixel threshold value) the first pixel value of pixel be set as first object pixel value, with thisSecond pixel value of the pre-set color template of target number differs the picture of (i.e. pixel value difference is greater than the presetted pixel threshold value) farther outFirst pixel value of vegetarian refreshments is set as the second target pixel value, and the face of the pre-set color template according to the target number may be implementedColor carries out binary conversion treatment to the color of object connected region.
In one embodiment, above-mentioned steps S303 can be performed as:
Pixel in the color of object connected region after determining binaryzation with corresponding position in the Character mother plateThe numbers of the equal target pixel points of pixel value whether be greater than preset number, wherein Character mother plate is to describe the target numberTemplate (such as: when target number is 5, Character mother plate description number be exactly 5, when target number is 3, Character mother plateThe number of description is exactly that 3), and the pixel of corresponding position refers in the target number of the Character mother plate in the Character mother plate,Pixel identical with each pixel position in the mark of the color of object connected region；
When the number of the target pixel points is greater than the preset number, determine in the color of object connected regionIt is identified as the target number.
Target after carrying out binary conversion treatment to the color of object connected region, after may further determine that the binaryzationIn color-connected regions the target pixel points equal with the pixel value of pixel of corresponding position in the Character mother plate and theseThe specific number of target pixel points, if it is greater than preset number, then it is assumed that successful match can accurately determine that the color of object connectsMark in logical region is exactly the target number.
As shown in figure 4, the present invention also provides a kind of determining devices for interacting footmark prompt, comprising:
First determining module 401, for the pre-set color template according to the corresponding default prompt of interaction footmark, determination is worked asThe color of object connected region to match in preceding video frame with pre-set color template；
Judgment module 402, for judging whether are the mark in color of object connected region and the confidence level of default promptGreater than default confidence level；
Second determining module 403, for determining in color of object connected region when confidence level is greater than default confidence levelIt is identified as default prompt.
As shown in figure 5, in one embodiment, default prompt includes: at least one countdown number；
Judgment module 402 includes:
First determines that submodule 4021 determines color of object connected region for using the convolutional neural networks pre-establishedThe confidence level of mark and each countdown number at least one countdown number in domain；
Submodule 4022 is selected, for selecting the highest mesh of confidence level with mark from least one countdown numberMark number；
Second determines submodule 4023, for determining whether the corresponding confidence level of target number is greater than default confidence level；
Second determining module 403 includes:
Third determines submodule 4031, for determining mesh when the corresponding confidence level of target number is greater than default confidence levelTarget number is identified as in mark color-connected regions.
As shown in fig. 6, in one embodiment, device further include:
Comparison module 601, for determine in color of object connected region be identified as target number before, by target faceThe second pixel value of the pre-set color template of the first pixel value and target number of each pixel is compared in color connected regionCompared with；
Processing module 602, for according to comparison result, by the pixel value of all pixels point in color of object connected region intoRow binaryzation；
Third determining module 603, for the pixel value according to the pixel in the color of object connected region after binaryzationWith the pixel value of the pixel in the Character mother plate of target number, determines and identify whether in color of object connected region as targetNumber.
As shown in fig. 7, in one embodiment, processing module 602 includes:
Judging submodule 6021, for judging the first pixel value and target of each pixel in color of object connected regionWhether the pixel value difference of the second pixel value of the pre-set color template of number is greater than or equal to presetted pixel threshold value；
First setting submodule 6022, for by color of object connected region with the pre-set color template of target numberThe first pixel value that the pixel value difference of second pixel value is less than or equal to the pixel of presetted pixel threshold value is set as first objectPixel value；
Second setting submodule 6023, for by color of object connected region with the pre-set color template of target numberThe first pixel value that the pixel value difference of second pixel value is greater than the pixel of presetted pixel value threshold value is set as the second object pixelValue.
As shown in figure 8, in one embodiment, third determining module 603 includes:
4th determines submodule 6031, for determine in the color of object connected region after binaryzation with phase in Character mother plateWhether the number for the target pixel points for answering the pixel value of the pixel of position equal is greater than preset number；
5th determines submodule 6032, for determining color of object when the number of target pixel points is greater than preset numberTarget number is identified as in connected region.
About the device in above-described embodiment, wherein modules execute the concrete mode of operation in related this methodEmbodiment in be described in detail, no detailed explanation will be given here.
Finally, the determining device of the interaction footmark prompt in the present invention is suitable for terminal device.For example, it may be mobilePhone, computer, digital broadcasting terminal, messaging device, game console, tablet device, Medical Devices, body-building equipment,Personal digital assistant etc..
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer programProduct.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present inventionApply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,The shape for the computer program product implemented in usable storage medium (including but not limited to magnetic disk storage and optical memory etc.)Formula.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program productFigure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructionsThe combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programsInstruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produceA raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for realThe device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spyDetermine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram orThe function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that countingSeries of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer orThe instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram oneThe step of function of being specified in a box or multiple boxes.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the artMind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologiesWithin, then the present invention is also intended to include these modifications and variations.