The detection method of a kind of video source and deviceTechnical field
The present invention relates to the technical field that video data processes, particularly relate to detection method and the one of a kind of video sourceThe detection device of video source.
Background technology
Along with the development of science and technology, the entertainment life of people is also increasingly to enrich, the TV programme of each television stationOccurring like the mushrooms after rain, general user can be gathered recording TV program by video frequency collection card link television set or be passed throughThe linking of devices television set recording TV program of DVR etc, then changes the most permissible by simple editing, formUpload to network.
Because watching video on network not limited by the time, usual people like using fixing broadcasting on networkIts TV programme liked watched by device, and the size (length-width ratio) of current video source is broadly divided into 4:3,16:9,21:9, and TVPlatform provides the size of video source and is usually 16:9 or 4:3, is transmitted by satellite-signal or cable TV signal；But owing to regardingFrequently the problem in source, the video source of above two size there will be the video source of situation about mixing, the most a size of 16:9 and can mixThe video source the video source of a size of 4:3, a size of 4:3 can mix the video source of a size of 16:9, on network oneFor individual fixing player, the above-mentioned video source mixed can be carried out stretched operation or reduction operation.
But, carry out stretched operation or the fluency of video playback can be impacted by reduction operation, play simultaneouslyImage quality also can be deteriorated, and directly affects the effect of spectators' viewing.
Summary of the invention
In view of the above problems, it is proposed that the embodiment of the present invention is to provide one to overcome the problems referred to above or at least in partSolve detection method method and the detection device of corresponding a kind of video source of a kind of video source of the problems referred to above.
In order to solve the problems referred to above, the embodiment of the invention discloses the detection method of a kind of video source, described method includes:
One or more target marking area is determined from video source image to be detected；
Calculate the object feature value of described target marking area；
Calculate the object feature value of described target marking area and the fixed reference feature value in preset corresponding reference mark regionSimilarity；
When described similarity is more than predetermined threshold value, calculate the target length-width ratio of described target marking area；
Judge the target length-width ratio of described target marking area, with the reference length-width ratio in described reference mark region whether phaseWith；
The most then judge that described video source data does not changes；Otherwise, it is determined that described video source data changes.
Alternatively, from video source image to be detected, determine that the step of one or more target marking area includes:
The image-region of selected preset range in described video source image to be detected；
Image recognition is carried out, it is thus achieved that the one or more target marking area for described image-region.
Alternatively, described image recognition is carried out for described image-region, it is thus achieved that the one or more blip districtThe sub-step in territory farther includes；
Described image-region is converted to accumulative gray level image；
Described accumulative gray level image is carried out binaryzation, it is thus achieved that bianry image；
Connected domain extraction method is used to process described bianry image, it is thus achieved that the one or more target marking area.
Alternatively, the quantity of described video source image is multiframe, and the most described image-region is converted to accumulative gray scaleThe sub-step of image farther includes；
Image-region in the video source image of described multiframe is converted to the gray level image of multiframe；
Calculate the maximum pixel difference between every frame and add up, obtaining inter-pixel and add up image；
Employing negates and the most normalized mode processes described interframe and adds up image, it is thus achieved that described accumulative gray level image.
Alternatively, described described accumulative gray level image is carried out binaryzation, it is thus achieved that the step of bianry image is:
Use accumulative gray level image described in thresholding method binaryzation, it is thus achieved that described bianry image.
Alternatively, described connected domain extraction method includes direct passing marker method and bianry image connected component labeling express method,The most described employing connected domain extraction method processes described bianry image, it is thus achieved that the sub-step of the one or more target marking areaFarther include:
Direct passing marker method is used to extract described bianry image, it is thus achieved that the one or more target marking area；
Or, use bianry image connected component labeling express method to extract described bianry image, it is thus achieved that the one or moreTarget marking area.
Alternatively, it is characterised in that described object feature value includes histograms of oriented gradients feature, described in the most described calculatingThe sub-step of the object feature value of target marking area farther includes:
Described target marking area is carried out dividing processing, it is thus achieved that multiple images are interval；
Described image interval comprises interval mark；
Calculate the histograms of oriented gradients that described image is interval；
According to described interval mark, the histograms of oriented gradients that described image is interval is connected into described direction gradient NogataFigure feature.
Alternatively, the generating mode of the fixed reference feature value in described preset corresponding reference mark region is as follows:
The image-region of selected preset range in described video source image to be detected；
Image recognition is carried out, it is thus achieved that the one or more reference mark region for described image-region；
Calculate the fixed reference feature value in described reference mark region.
Alternatively, described preset range is at the described upper left corner of video source image or the 1/4 of the upper right corner.
The embodiment of the invention also discloses the detection device of a kind of video source, described device includes:
Target marking area determines module, for determining one or more blip from video source image to be detectedRegion；
Object feature value computing module, for calculating the object feature value of described target marking area；
Similarity calculation module, for calculating the object feature value of described target marking area and preset corresponding reference markThe similarity of the fixed reference feature value in will region；
Target length-width ratio computing module, for when described similarity is more than predetermined threshold value, calculating described blip districtThe target length-width ratio in territory；
Length-width ratio judge module, for judging the target length-width ratio of described target marking area, with described reference mark districtThe reference length-width ratio in territory is the most identical；
First determination module, is used for judging that described video source data does not changes；
Second determination module, is used for judging that described video source data changes.
Alternatively, described target marking area determines that module includes:
First image-region selectes submodule, for the figure of preset range selected in described video source image to be detectedAs region；
Target marking area obtains submodule, for carrying out image recognition for described image-region, it is thus achieved that oneOr multiple target marking area.
Alternatively, acquisition submodule in described target marking area farther includes；
Accumulative greyscale image transitions unit, is converted to accumulative gray level image for described image-region；
Bianry image obtains unit, for described accumulative gray level image is carried out binaryzation, it is thus achieved that bianry image；
Target marking area obtains unit, is used for using connected domain extraction method to process described bianry image, it is thus achieved that described oneIndividual or multiple target marking area.
Alternatively, the quantity of described video source image is multiframe, and the most described accumulative greyscale image transitions unit is furtherIncluding；
Greyscale image transitions subelement, for being converted to multiframe by the image-region in the video source image of described multiframeGray level image；
Inter-pixel adds up image and obtains subelement, for calculating the maximum pixel difference between every frame and adding up, it is thus achieved thatInter-pixel adds up image；
Accumulative gray level image obtains subelement, and employing negates and the most normalized mode processes the accumulative figure of described interframePicture, it is thus achieved that described accumulative gray level image.
Alternatively, described bianry image acquisition unit includes:
Bianry image obtains subelement, is used for using accumulative gray level image described in thresholding method binaryzation, it is thus achieved that instituteState bianry image.
Alternatively, it is characterised in that described connected domain extraction method includes direct passing marker method and bianry image connected domainLabelling express method, the most described target marking area obtains unit and farther includes:
First object mark region obtains subelement, is used for using described direct passing marker method to extract described binary mapPicture, it is thus achieved that the one or more target marking area；
Or, the second target marking area obtains subelement, is used for using described bianry image connected component labeling express methodExtract described bianry image, it is thus achieved that the one or more target marking area.
Alternatively, described object feature value includes histograms of oriented gradients feature, the most described object feature value computing moduleFarther include:
Image interval obtains submodule, for described target marking area is carried out dividing processing, it is thus achieved that multiple image districtsBetween；Described image interval comprises interval mark；
Histograms of oriented gradients calculating sub module, for calculating the histograms of oriented gradients that described image is interval；
Histograms of oriented gradients feature connexon module, for according to described interval mark, the side interval by described imageDescribed histograms of oriented gradients feature is connected into histogram of gradients.
Alternatively, described similarity calculation module includes following submodule:
Second image-region selectes submodule, for the figure of preset range selected in described video source image to be detectedAs region；
Reference mark region obtains submodule, for carrying out image recognition for described image-region, it is thus achieved that oneOr multiple reference mark region；
Fixed reference feature value calculating sub module, for calculating the fixed reference feature value in described reference mark region.
Alternatively, described preset range is at the described upper left corner of video source image or the 1/4 of the upper right corner.
The embodiment of the present invention includes advantages below:
In the embodiment of the present invention, the mode of calculated direction histogram of gradients feature is used to come comparison object mark region and ginsengExamine the content in mark region, because histograms of oriented gradients is the architectural feature representing gradient (edge), local can be describedShape information, and, due to histograms of oriented gradients characteristic use is the processing mode of piecemeal subdivision so that image officeRelation between portion's pixel is well characterized；Comparison content between the two is more accurate, automatically generates the reference of standardMark region and storehouse, reference mark region；Use image recognition technology, the image-region being recognized accurately in preset rangeTarget marking area, calculates object feature value and fixed reference feature value, repeatedly comparison object mark region object feature value with ginsengExamine similarity and the predetermined threshold value of the fixed reference feature value of mark region, make comparison result have robustness.
Embodiment of the present invention comparison target marking area and the length-width ratio in reference mark region, detect video source figure exactlyWhether the size of picture changes, and detects whether the source of video source changes the most exactly, improves video source and broadcastsThe fluency put and image quality, and provide certain convenience for the examination & verification of video source.
Accompanying drawing explanation
Fig. 1 shows the flow chart of steps of the detection method embodiment one of a kind of video source of the embodiment of the present invention；
Fig. 2 shows the schematic diagram in the storehouse, a kind of reference mark region of the embodiment of the present invention；
Fig. 3 shows the target marking area schematic diagram under normal size；
Fig. 4 shows the target marking area schematic diagram under deformation size；
Fig. 5 shows the schematic diagram in reference mark region；
Fig. 6 shows the flow chart of steps of the detection method embodiment two of a kind of video source of the embodiment of the present invention；
Fig. 7 shows that embodiment of the present invention one is converted to gray level image by coloured image, and greyscale image transitions is accumulativeGray level image, accumulative gray level image is reconverted into the process schematic of bianry image；
Fig. 8 shows the structured flowchart of the detection device embodiment one of a kind of video source of the embodiment of the present invention；
Fig. 9 shows a kind of terminal unit structural representation that the embodiment of the present invention provides.
Detailed description of the invention
Understandable for enabling the above-mentioned purpose of the present invention, feature and advantage to become apparent from, real with concrete below in conjunction with the accompanying drawingsThe present invention is further detailed explanation to execute mode.
One of core idea of the embodiment of the present invention is, selectes carrying out a particular range in video source image,Then particular range is carried out image recognition, extract mark region template, the mark of television station can be included, set up logo areaThe template base in territory；The video source image originating same signal later carries out the selected of particular range, then performs image recognitionStep, extract mark region, compare that it is the most consistent with picture size with the content of the mark region template of template base, reachTo the most normal effect of detection video source size.
With reference to Fig. 1, it is shown that the flow chart of steps of the detection method embodiment one of a kind of video source of the present invention, specifically may be usedTo comprise the steps:
Step 101, determines one or more target marking area from video source image to be detected；
In implementing, a video source data generally has the video source image of multiframe, by image recognition technology, from regardingFrequently determining one or more target marking area in source images, described target marking area can include the logo area of television stationTerritory.
Step 102, calculates the object feature value of described target marking area；
In another preferred embodiment in the embodiment of the present invention, described object feature value can include direction gradient NogataFigure feature, described step 102 includes following sub-step:
Sub-step S1021, carries out dividing processing to described target marking area, it is thus achieved that multiple images are interval；
Sub-step S1022, described image interval comprises interval mark；
Sub-step S1023, calculates the histograms of oriented gradients that described image is interval；
Sub-step S1024, according to described interval mark, connects into described by the histograms of oriented gradients that described image is intervalHistograms of oriented gradients feature.
Wherein, histograms of oriented gradients (Histogram ofOriented Gradients) feature, it is called for short HOG feature,It is a kind of feature describing image local texture of being in daily use of current computer vision, area of pattern recognition, piece image is dividedInterval for multiple images, multiple unit is contained in each interval, by multiple pixels (elementary cell of image) component units.
Such as, piece image is divided into 16 intervals, and 4 unit are contained in each interval, and each unit has the pixel of 8x8.MeterCalculate the histograms of oriented gradients of each unit, then obtain the histograms of oriented gradients in each interval, according to the district that image is intervalBetween identify, histograms of oriented gradients interval for each image is connected into the histograms of oriented gradients feature of target marking area.
It should be noted that described object feature value includes but not limited to histograms of oriented gradients feature, implementing thisDuring bright embodiment, other eigenvalues can be calculated according to practical situation, as long as the image local stricture of vagina of target marking area can be describedReason feature, this is not any limitation as by the embodiment of the present invention.
Step 103, calculates the object feature value of described target marking area and the ginseng in preset corresponding reference mark regionExamine the similarity of eigenvalue；
In another preferred embodiment of the embodiment of the present invention, the most described preset corresponding reference mark districtThe generating mode of the fixed reference feature value in territory includes following sub-step:
Sub-step S1031, the image-region of selected preset range in described video source image to be detected；
Sub-step S1032, carries out image recognition for described image-region, it is thus achieved that the one or more reference mark districtTerritory；
Sub-step S1033, calculates the fixed reference feature value in described reference mark region.
For making those skilled in the art be more fully understood that the embodiment of the present invention, below by way of an example to preset correspondenceThe generating mode of fixed reference feature value in reference mark region illustrates with the calculation of similarity.Described reference mark regionCan include that the region at TV-station logotype place, reference mark region are equivalent to preset logo template, set up reference mark districtStorehouse, territory, with reference to Fig. 2, it is shown that a kind of storehouse, reference mark region.Target marking area and reference mark region are belonging to same letterNumber source, simply time location is different, and the sequencing i.e. obtained is different.Described fixed reference feature value includes that direction gradient is straightSide's figure feature, calculates the fixed reference feature value in described reference mark region, is equivalent to the histograms of oriented gradients of calculation flag templateFeature, contrast target marking area and the histograms of oriented gradients feature of each unit in each interval, reference mark region, realMatter is the histograms of oriented gradients feature of comparison object mark region and the unit of the same position of logo template, calculates target markThe similarity of the histograms of oriented gradients feature of the unit of will region and reference mark region same position.
Such as, when there are 16 intervals target marking area, and 4 unit are contained in each interval, and each unit has the picture of 8x8Element；Because target marking area and reference mark region are belonging to same signal source, simply time location is different, therefore simultaneouslyThere are 16 intervals in reference mark region, and 4 unit are contained in each interval, and each unit has the pixel of 8x8, if, set and judgeRule be interval 4 unit of same position the most identical time, be an identical interval, when blip district in 16 intervalsThere are 14 identical intervals in territory and reference mark region, then it is assumed that similarity is 87.5%.It should be noted that for decision ruleSetting, the embodiment of the present invention does not make any restriction.
Step 104, when described similarity is more than predetermined threshold value, calculates the target length-width ratio of described target marking area；
In being embodied as, predetermined threshold value could be arranged to 80% or 90%, and excursion can be 0-100%, thisThis is not made any restriction by inventive embodiments.When similarity is more than predetermined threshold value, the target calculating target marking area is longWide ratio, specifically, calculates the size of the length direction of target marking area and the size of width, it is thus achieved that blip districtThe length-width ratio in territory.
For example, it is possible to arranging predetermined threshold value is 85%, by the target of the calculated described target marking area of step 103Eigenvalue is 87.5% with the similarity of the fixed reference feature value in preset corresponding reference mark region, and similarity is more than presetting thresholdValue, with reference to Fig. 3, it is shown that the target marking area under normal size, the size calculating target marking area length direction is1.2 inches, the size of width is 0.8 inch, then the length-width ratio of target marking area is 1.5；With reference to Fig. 4, it is shown that becomeTarget marking area under shape size, if calculating 1.2 inches of reference mark zone length direction, the size of width is1.0 inches, then the length-width ratio obtaining target marking area is 1.2.
Step 105, it is judged that the target length-width ratio of described target marking area, with the reference length and width in described reference mark regionRatio is the most identical；
It should be noted that the computational methods of the reference length-width ratio in reference mark region are long with the target of target marking areaThe computational methods of wide ratio are consistent, calculate both length-width ratios, and the length-width ratio then comparing both is the most identical.
With reference to Fig. 5, it is shown that reference mark region, the size calculating reference mark zone length direction is 1.2 inches,The size of width is 0.8 inch, then the reference length-width ratio obtaining reference mark region is 1.5, comparison blip simultaneouslyThe target length-width ratio in region is the most identical with the reference length-width ratio in reference mark region.
Step 106, the most then judge that described video source data does not changes；Otherwise, it is determined that described video source data is sent outChanging.
Wherein, if the reference length-width ratio in reference mark region is 1.5, and the length-width ratio of target marking area is 1.5, it is believed thatBoth length-width ratios are identical, it is determined that when video source data does not changes, and can export the normal information of video source；If with reference to markThe reference length-width ratio in will region is 1.5, and the length-width ratio of target marking area is 1.2, it is believed that both length-width ratios are different, it is determined thatWhen video source data changes, the information that video source is abnormal can be exported.
In the embodiment of the present invention, the mode of calculated direction histogram of gradients feature is used to come comparison object mark region and ginsengExamine the content in mark region, because histograms of oriented gradients is the architectural feature representing gradient (edge), local can be describedShape information, and, due to histograms of oriented gradients characteristic use is the processing mode of piecemeal subdivision so that image officeRelation between portion's pixel is well characterized；Comparison content between the two is more accurate, automatically generates the reference of standardMark region and storehouse, reference mark region；Whether the size of detection video source image changes exactly.
With reference to Fig. 6, it is shown that the flow chart of steps of the detection method embodiment two of a kind of video source of the present invention, specifically may be usedTo comprise the steps:
Step 201, the image-region of selected preset range in described video source image to be detected；
In implementing, described expected range is at the upper left corner of video source image or the 1/4 of the upper right corner.For sameThe video source image of signal source, the mark region at its place is typically mounted on the upper left corner or the upper right corner of screen.The present invention is realExecute in example, select the image-region at mark region place faintly.
Step 202, described image-region is converted to accumulative gray level image；
In a preferred embodiment in the embodiment of the present invention, the quantity of described video source image is multiframe, then stepThe sub-step of 202 farther includes；
Sub-step S2021, is converted to the gray level image of multiframe by the image-region in the video source image of described multiframe；
Sub-step S2022, calculates the maximum pixel difference between every frame and adds up, obtaining inter-pixel and add up image；
Sub-step S2023, employing negates and the most normalized mode processes described interframe and adds up image, it is thus achieved that described tiredMeter gray level image.
With reference to Fig. 7, it is shown that be converted to gray level image by coloured image, greyscale image transitions is accumulative gray level image, tiredMeter gray level image is reconverted into the process of bianry image.Specifically, the image-region in video source image is converted to gray scaleImage, really transfers coloured image to gray level image.The process that coloured image transfers to gray level image is to remove coloured imageColouring information, represent image luminance information at gray value；In prior art, coloured image typically uses rgb color patternDisplay chroma-luminance, shows the chroma-luminance of image, image with R (red), G (green), the superposition of B (blue) three kinds of colorsThe R value of middle pixel, G-value are 0-255 with the excursion of B value, and when R value, G-value are all 255 with B value numerical value, this pixel is whiteColor；When R value, G-value are all 0 with B value numerical value, this pixel is black.Coloured image picture for a three-component RGBElement, the chroma-luminance of this pixel calculates the most as follows: I=0.3B+0.59G+0.11R, and R in above-mentioned formula, G, B divideAmount is arranged to identical value, it is possible to obtain described gray level image, the method transferring gray level image to for coloured image, the present inventionEmbodiment does not make any restriction.
Wherein, the gray level image to multiframe uses following formula to add up: Isum=max (Isum, abs (Ipre-Icur))；Wherein IsumRepresent accumulative image, IpreIt is a frame gray level image, IcurFor present frame gray image, abs (Ipre-Icur) the absolute value of pixel value difference；To accumulative image IsumCarry out negating and normalization operation again, reduce light uneven rightThe impact that image causes, obtains described accumulative gray level image Inew。
Step 203, carries out binaryzation to described accumulative gray level image, it is thus achieved that bianry image；
In implementing, described accumulative gray level image is carried out binaryzation according to following setting,IbinaRepresent bianry image；Th represents that segmentation threshold, segmentation threshold th are set by empirical value, pointIt is not set to th={50,60,70 .., 240}；When in the pixel of accumulative gray level image, R value, G-value are more than set with B valueDuring segmentation threshold th, with the numerical value of B value, the R value in this pixel, G-value being both configured to 255, this pixel becomes white；WhenWhen in the pixel of accumulative gray level image, R value, G-value are less than set segmentation threshold th with B value, by the R value in this pixel, GValue is both configured to 0 with the numerical value of B value, and this pixel becomes black, by the number of the R value of described accumulative gray level image, G-value and B valueValue is both configured to 0 or 255, obtains described bianry image, described bianry image is carried out erosion operation, removes interference noise, soAfter do dilation operation again so that bianry image is more complete.
Step 204, uses connected domain extraction method to process described bianry image, it is thus achieved that the one or more blip districtTerritory.
In another kind of preferred embodiment in embodiments of the present invention, described connected domain extraction method includes direct passing markerMethod and bianry image connected component labeling express method, then the sub-step of step 204 farther includes:
Sub-step S2041, uses direct passing marker method to extract described bianry image, it is thus achieved that the one or more targetMark region；
Or, sub-step S2042, use bianry image connected component labeling express method to extract described bianry image, it is thus achieved that instituteState one or more target marking area.
In implementing, the most important method of binary image analysis is exactly that connected domain is extracted, and it is that all bianry images divideThe basis of analysis, it, by white pixel in bianry image (target marking area) is extracted labelling, allows each single connected domainForm an identified block, further just can obtain the geometry such as the profile of these blocks, boundary rectangle, barycenter, not bending momentParameter.
In embodiments of the present invention, obtain the maximum boundary rectangle of the connected region of bianry image, be defined as blipRegion, and for the accuracy of result, it may be determined that multiple target marking area, i.e. from the different video of same signal sourceSource images determines multiple target marking area.In the embodiment of the present invention, use bianry image connected component labeling express method orPerson's direct passing marker method carries out the extraction of connected domain, allows each single connected domain form an identified block, obtains blockMaximum boundary rectangle, be defined as target marking area；It should be noted that connected domain is extracted by the embodiment of the present inventionMethod is not intended to be limited in any.
Step 205, calculates the object feature value of described target marking area；
In being embodied as, calculating the histograms of oriented gradients of each unit, the direction gradient then obtaining each interval is straightFang Tu, according to the interval mark that image is interval, connects into target marking area by the histograms of oriented gradients that each image is intervalHistograms of oriented gradients feature.
Step 206, calculates the object feature value of described target marking area and the ginseng in preset corresponding reference mark regionExamine the similarity of eigenvalue；
In actual application, reference mark region is equivalent to preset logo template, and the embodiment of the present invention collects reference markRegion, sets up storehouse, reference mark region.Target marking area and reference mark region are belonging to same signal source, time simplyBetween position different, the sequencing i.e. obtained is different.Calculate the fixed reference feature value in described reference mark region, be equivalent to calculate markThe histograms of oriented gradients feature of will template, contrast target marking area and each unit in each interval, reference mark regionThe direction gradient Nogata of the unit of the same position of histograms of oriented gradients feature, i.e. comparison object mark region and logo templateFigure feature, calculates target marking area similar to the histograms of oriented gradients feature of the unit of reference mark region same positionDegree.
Step 207, when described similarity is more than predetermined threshold value, calculates the target length-width ratio of described target marking area；
In the embodiment of the present invention, target setting mark region and the number of comparisons in reference mark region, multiple by obtainTarget marking area compares with reference mark region (template), when described similarity is more than pre-in certain number of times all continuouslyIf threshold value, then it is assumed that result has robustness.It should be noted that for the number of times compared, the present invention does not make any restriction.When similarity is more than predetermined threshold value, calculates the target length-width ratio of target marking area, specifically, calculate target marking areaThe size of size and width of length direction, it is thus achieved that the length-width ratio of target marking area.
Step 208, it is judged that the target length-width ratio of described target marking area, with the reference length and width in described reference mark regionRatio is the most identical；
Step 209, the most then judge that described video source data does not changes；Otherwise, it is determined that described video source data is sent outChanging.
In actual applications, when judging that video source data does not changes, the normal information of video source can be exported；WhenJudge, when video source data changes, the information that video source is abnormal to be exported.
In the embodiment of the present invention, use image recognition technology, the mesh of the image-region being recognized accurately in preset rangeMark mark region, and set up storehouse, reference mark region, calculate object feature value and fixed reference feature value, repeatedly comparison object logo areaThe similarity of the fixed reference feature value with reference mark region of territory object feature value and predetermined threshold value, make comparative result have robustProperty, additionally, comparison target marking area and the length-width ratio in reference mark region, detect whether the source of video source is sent out exactlyChanging, the examination & verification for video source provides certain convenience.
It should be noted that for embodiment of the method, in order to be briefly described, therefore it is all expressed as a series of action groupClosing, but those skilled in the art should know, the embodiment of the present invention is not limited by described sequence of movement, because depending onAccording to the embodiment of the present invention, some step can use other orders or carry out simultaneously.Secondly, those skilled in the art also shouldKnowing, embodiment described in this description belongs to preferred embodiment, and the involved action not necessarily present invention implementsNecessary to example.
With reference to Fig. 8, it is shown that the structured flowchart of the detection device embodiment one of a kind of video source of the present invention, the most permissibleIncluding such as lower module:
Target marking area determines module 301, for determining one or more target from video source image to be detectedMark region；
Object feature value computing module 302, for calculating the object feature value of described target marking area；
In a preferred embodiment in the embodiment of the present invention, object feature value computing module 302 includes following submoduleBlock:
Image interval obtains submodule 3021, for described target marking area is carried out dividing processing, it is thus achieved that Duo GetuAs interval；Described image interval comprises interval mark；
Histograms of oriented gradients calculating sub module 3022, for calculating the histograms of oriented gradients that described image is interval；
Histograms of oriented gradients feature connexon module 3023, for according to described interval mark, described image is intervalHistograms of oriented gradients connect into described histograms of oriented gradients feature.
Similarity calculation module 303, for calculating the object feature value of described target marking area and preset corresponding ginsengExamine the similarity of the fixed reference feature value of mark region；
In another preferred embodiment of the invention, similarity calculation module 303 includes following submodule:
Second image-region selectes submodule 3031, for preset range selected in described video source image to be detectedImage-region；
Reference mark region obtains submodule 3032, for carrying out image recognition for described image-region, it is thus achieved that describedOne or more reference mark regions；
Fixed reference feature value calculating sub module 3033, for calculating the fixed reference feature value in described reference mark region.
Target length-width ratio computing module 304, for when described similarity is more than predetermined threshold value, calculating described blipThe target length-width ratio in region；
Length-width ratio judge module 305, for judging the target length-width ratio of described target marking area, with described reference markThe reference length-width ratio in region is the most identical；
First determination module 306, is used for judging that described video source data does not changes；
Second determination module 307, is used for judging that described video source data changes.
For device embodiment, due to itself and embodiment of the method basic simlarity, so describe is fairly simple, relevantPart sees the part of embodiment of the method and illustrates.
With reference to Fig. 9, it illustrates the structural representation of the terminal unit that one embodiment of the invention provides.This electronic equipmentFor implementing the detection method of the video source provided in above-described embodiment, specifically:
Electronic equipment 800 can include RF (Radio Frequency, radio frequency) circuit 810, include one or one withThe memorizer 820 of upper computer-readable recording medium, input block 830, display unit 840, sensor 850, voicefrequency circuit860, short range wireless transmission module 870, include one or more than one processes the processor 880 of core and power supply890 parts such as grade.It will be understood by those skilled in the art that the electronic devices structure shown in Fig. 9 is not intended that electronic equipmentLimit, can include that ratio illustrates more or less of parts, or combine some parts, or different parts are arranged.Wherein:
RF circuit 810 can be used for receiving and sending messages or in communication process, the reception of signal and transmission, especially, by base stationAfter downlink information receives, transfer to one or more than one processor 880 processes；It addition, be sent to relating to up dataBase station.Generally, RF circuit 810 includes but not limited to antenna, at least one amplifier, tuner, one or more agitator, receiptsSender, bonder, LNA (Low Noise Amplifier, low-noise amplifier), duplexer etc..Additionally, RF circuit 810 is alsoCan be communicated with network and other equipment by radio communication.Described radio communication can use arbitrary communication standard or agreement,Include but not limited to GSM (Global System of Mobile communication, global system for mobile communications), GPRS(General Packet Radio Service, general packet radio service), CDMA (Code Division MultipleAccess, CDMA), WCDMA (Wideband Code Division Multiple Access, WCDMA),LTE (Long Term Evolution, Long Term Evolution), Email, SMS (Short Messaging Service, short messageService) etc..Memorizer 820 can be used for storing software program and module, and such as, memorizer 820 may be used for storage and gathers languageThe software program of tone signal, realize the software program of key word identification, the software program realizing continuous speech recognition and realizationSoftware program of prompting item etc. is set.Processor 880 is stored in software program and the mould of memorizer 820 by operationBlock, thus perform the application of various function and data process, in the such as embodiment of the present invention " from video source image to be detectedDetermine one or more target marking area " function, the function of " calculating the object feature value of described target marking area "," calculate the object feature value of described target marking area and the similar of the fixed reference feature value in preset corresponding reference mark regionDegree ", " when described similarity more than predetermined threshold value time, calculate the target length-width ratio of described target marking area " " judge described meshMark mark region target length-width ratio, the most identical with the reference length-width ratio in described reference mark region ", " the most then judge instituteState video source data not change；Otherwise, it is determined that described video source data changes " function etc..Memorizer 820 canMainly include storing program area and storage data field, wherein, needed for storage program area can store operating system, at least one functionApplication program (such as sound-playing function, image player function etc.) etc.；Storage data field can store according to electronic equipment 800The data (such as voice data, phone directory etc.) etc. that created of use.Additionally, memorizer 820 can include depositing at random at a high speedAccess to memory, it is also possible to include nonvolatile memory, for example, at least one disk memory, flush memory device or other are easyThe property lost solid-state memory.Correspondingly, memorizer 820 can also include Memory Controller, to provide processor 880 and inputThe unit 830 access to memorizer 820.
Input block 830 can be used for receiving numeral or the character information of input, and produces and user setup and functionControl relevant keyboard, mouse, action bars, optics or the input of trace ball signal.Specifically, input block 830 can include touchingSensitive surfaces 831 and other input equipments 832.Touch sensitive surface 831, also referred to as touches display screen or Trackpad, can collect useFamily thereon or neighbouring touch operation (such as user uses any applicable object such as finger, stylus or adnexa at touch-sensitive tableOperation on face 831 or near Touch sensitive surface 831), and drive corresponding attachment means according to formula set in advance.Optional, Touch sensitive surface 831 can include touch detecting apparatus and two parts of touch controller.Wherein, touch detecting apparatus detection is usedThe touch orientation at family, and detect the signal that touch operation brings, transmit a signal to touch controller；Touch controller is from touchReceive touch information on detection device, and be converted into contact coordinate, then give processor 880, and processor 880 can be receivedThe order sent also is performed.Furthermore, it is possible to use the polytypes such as resistance-type, condenser type, infrared ray and surface acoustic waveRealize Touch sensitive surface 831.Except Touch sensitive surface 831, input block 830 can also include other input equipments 832.Specifically,Other input equipments 832 can include but not limited to physical keyboard, function key (such as volume control button, switch key etc.),One or more in trace ball, mouse, action bars etc..
Display unit 840 can be used for showing the information inputted by user or the information being supplied to user and electronic equipmentThe various graphical user interface of 800, these graphical user interface can be come by figure, text, icon, video and its combination in anyConstitute.Display unit 840 can include display floater 841, optionally, can use LCD (Liquid Crystal Display,Liquid crystal display), the form such as OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode) configure displayPanel 841.Further, Touch sensitive surface 831 can cover on display floater 841, when Touch sensitive surface 831 detects thereonOr after neighbouring touch operation, send processor 880 to determine the type of touch event, with preprocessor 880 according to touchThe type of event provides corresponding visual output on display floater 841.Although in fig .9, Touch sensitive surface 831 and display floater841 is to realize input and input function as two independent parts, but in some embodiments it is possible to by Touch sensitive surface831 with display floater 841 integrated and realize input and output function.
Electronic equipment 800 may also include at least one sensor 850, such as optical sensor, motion sensor and otherSensor.Specifically, optical sensor can include ambient light sensor and proximity transducer, and wherein, ambient light sensor can basisThe light and shade of ambient light regulates the brightness of display floater 841, proximity transducer can when electronic equipment 800 moves in one's ear,Close display floater 841 and/or backlight.As the one of motion sensor, Gravity accelerometer can detect all directionsThe size of upper (generally three axles) acceleration, can detect that size and the direction of gravity time static, can be used for identifying mobile phone attitudeApplication (such as horizontal/vertical screen switching, dependent game, magnetometer pose calibrating), Vibration identification correlation function (such as pedometer, strikesHit) etc.；The gyroscope that can also configure as electronic equipment 800, barometer, drimeter, thermometer, infrared ray sensor etc. otherSensor, does not repeats them here.
Voicefrequency circuit 860, speaker 861, microphone 862 can provide the audio interface between user and electronic equipment 800.The signal of telecommunication after the voice data conversion that voicefrequency circuit 860 can will receive, is transferred to speaker 861, speaker 861 changesExport for acoustical signal；On the other hand, the acoustical signal of collection is converted to the signal of telecommunication by microphone 862, voicefrequency circuit 860 connectVoice data is converted to after receipts, then after voice data output processor 880 is processed, through RF circuit 810 to be sent to another eventuallyEnd, or voice data is exported to memorizer 820 to process further.Voicefrequency circuit 860 is also possible that earphone jack,To provide the communication of peripheral hardware earphone and electronic equipment 800.
Short range wireless transmission module 870 can be WIFI (wireless fidelity, Wireless Fidelity) module or indigo plantTooth module etc..Electronic equipment 800 can help user to send and receive e-mail, browse net by short range wireless transmission module 870Page and access streaming video etc., it has provided the user wireless broadband internet and has accessed.Although Fig. 9 shows short-distance wirelessTransport module 870, but it is understood that, it is also not belonging to must be configured into of electronic equipment 800, completely can be as requiredOmit in not changing the scope of essence of invention.
Processor 880 is the control centre of electronic equipment 800, utilizes various interface and the whole electronic equipment of connectionVarious piece, is stored in the software program in memorizer 820 and/or module by running or performing, and calls and be stored inData in reservoir 820, perform the various functions of electronic equipment 800 and process data, thus electronic equipment carries out overall prisonControl.Optionally, processor 880 can include one or more process core；Preferably, processor 880 can integrated application processorAnd modem processor, wherein, application processor mainly processes operating system, user interface and application program etc., modulatedemodulateProcessor is adjusted mainly to process radio communication.It is understood that above-mentioned modem processor can not also be integrated into processorIn 880.
Electronic equipment 800 also includes the power supply 890 (such as battery) powered to all parts, it is preferred that power supply can lead toCross power-supply management system logically contiguous with processor 880, thus by power-supply management system realize management charging, electric discharge andThe functions such as power managed.Power supply 890 can also include one or more direct current or alternating current power supply, recharging system, electricitySource failure detector circuit, power supply changeover device or the random component such as inverter, power supply status indicator.
Although not shown, electronic equipment 800 can also include photographic head, bluetooth module etc., does not repeats them here.Specifically existIn the present embodiment, the display unit of electronic equipment 800 is touch-screen display.
Each embodiment in this specification all uses the mode gone forward one by one to describe, what each embodiment stressed is withThe difference of other embodiments, between each embodiment, identical similar part sees mutually.
Those skilled in the art are it should be appreciated that the embodiment of the embodiment of the present invention can be provided as method, device or calculateMachine program product.Therefore, the embodiment of the present invention can use complete hardware embodiment, complete software implementation or combine software andThe form of the embodiment of hardware aspect.And, the embodiment of the present invention can use one or more wherein include computer canWith in the computer-usable storage medium (including but not limited to disk memory, CD-ROM, optical memory etc.) of program codeThe form of the computer program implemented.
The embodiment of the present invention is with reference to method, terminal unit (system) and computer program according to embodiments of the present inventionThe flow chart of product and/or block diagram describe.It should be understood that can be by computer program instructions flowchart and/or block diagramIn each flow process and/or the flow process in square frame and flow chart and/or block diagram and/or the combination of square frame.These can be providedComputer program instructions sets to general purpose computer, special-purpose computer, Embedded Processor or other programmable data processing terminalsStandby processor is to produce a machine so that held by the processor of computer or other programmable data processing terminal equipmentThe instruction of row produces for realizing in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frameThe device of the function specified.
These computer program instructions may be alternatively stored in and can guide computer or other programmable data processing terminal equipmentIn the computer-readable memory worked in a specific way so that the instruction being stored in this computer-readable memory produces bagIncluding the manufacture of command device, this command device realizes in one flow process of flow chart or multiple flow process and/or one side of block diagramThe function specified in frame or multiple square frame.
These computer program instructions also can be loaded on computer or other programmable data processing terminal equipment so thatOn computer or other programmable terminal equipment, execution sequence of operations step is to produce computer implemented process, thusThe instruction performed on computer or other programmable terminal equipment provides for realizing in one flow process of flow chart or multiple flow processAnd/or the step of the function specified in one square frame of block diagram or multiple square frame.
Although having been described for the preferred embodiment of the embodiment of the present invention, but those skilled in the art once knowing baseThis creativeness concept, then can make other change and amendment to these embodiments.So, claims are intended to be construed toThe all changes including preferred embodiment and falling into range of embodiment of the invention and amendment.
Finally, in addition it is also necessary to explanation, in this article, the relational terms of such as first and second or the like be used merely to byOne entity or operation separate with another entity or operating space, and not necessarily require or imply these entities or operationBetween exist any this reality relation or order.And, term " includes ", " comprising " or its any other variant meaningContaining comprising of nonexcludability, so that include that the process of a series of key element, method, article or terminal unit not only wrapInclude those key elements, but also include other key elements being not expressly set out, or also include for this process, method, articleOr the key element that terminal unit is intrinsic.In the case of there is no more restriction, by wanting that statement " including ... " limitsElement, it is not excluded that there is also other identical element in including the process of described key element, method, article or terminal unit.
Above to a kind of method provided by the present invention and a kind of device, it is described in detail, tool used hereinPrinciple and the embodiment of the present invention are set forth by body example, and the explanation of above example is only intended to help and understands thisBright method and core concept thereof；Simultaneously for one of ordinary skill in the art, according to the thought of the present invention, concrete realExecuting and all will change in mode and range of application, in sum, this specification content should not be construed as the limit to the present inventionSystem.