Specific embodiment
In order to make those skilled in the art better understand the technical solutions in the application, below in conjunction with the application realityThe attached drawing in mode is applied, the technical solution in the application embodiment is clearly and completely described, it is clear that describedEmbodiment is only a part of embodiment of the application, rather than whole embodiments.Based on the embodiment party in the applicationFormula, every other embodiment obtained by those of ordinary skill in the art without making creative efforts, is all answeredWhen the range for belonging to the application protection.
The application provides a kind of recognition methods of video quality, and the method can be applied to the service of video playback websiteIn device.Fig. 1 and Fig. 2 are please referred to, the method may include following steps.
S1: the broadcast information of target video is obtained, and is based on the broadcast information, judges whether the target video is doubtfulLike low quality video.
In the present embodiment, the target video can be the video of quality to be assessed.The target video is uploadingTo video playback website, can be watched by user.During user's viewing, the server of video playback websiteIt can recorde the broadcast information of the target video.It may include time that the target video is played in the broadcast informationNumber that the played average duration of several, the described target video, the target video are completely played, the target video are by pointPraise the number etc. that/forwarding/is reported/downloaded.
In the present embodiment, the broadcast information of the target video can be with the mark of the target video in serverMiddle associated storage.The mark of the target video can be digital coding of the target video in video playback website.InstituteThe mark and broadcast information for stating target video can be associated storage by way of key-value pair (key-value).Wherein, instituteStating mark can be used as key, and the broadcast information can be used as value, in this way, mark of the server by inquiry target video, justThe broadcast information of the target video can be read from storage equipment.
It in the present embodiment, can be preliminary to judge according to the broadcast information after getting the broadcast informationWhether the target video is doubtful low quality video.Specifically, it is contemplated that the lower video of run-of-the-mill, user viewing whenBetween all will not be too long, and the lower video of quality is by also higher a possibility that user's report.In consideration of it, in an embodimentIn, it can be for the average playing duration for the target video that the broadcast information includes and the ratio of total duration to judgeState whether target video is doubtful low quality video.Wherein, the average playing duration of the target video can be to the targetPlaying duration averaged of the video in each playing process obtains.For example, when the average broadcasting of the target videoA length of 1 minute, and the total duration of the target video is 10 minutes, then the ratio of the average playing duration and total duration is justIt can be 0.1.In the present embodiment, it can preset for judging whether the ratio calculated is in the finger of normal range (NR)Periodically long proportion threshold value.The specified duration proportion threshold value can be the ratio-dependent for normal video.For example, normal viewThe frequency averagely ratio of playing duration and total duration is often 0.6 or more, then the specified duration proportion threshold value can be set as0.6, or the numerical value more slightly lower than 0.6, it can specifically be adjusted flexibly according to the actual situation.Specified duration proportion threshold value is being determinedLater, the ratio being calculated can be compared with the specified duration proportion threshold value.When being averaged for the target videoWhen the ratio of playing duration and total duration is lower than the specified duration proportion threshold value, it is possible to determine that the target video is doubtful lowQuality video.
In another embodiment, it is contemplated that a possibility that low quality video is completely watched by user is also relatively low, becauseThis can according to number that the target video for including in broadcast information is completely watched with by the ratio of viewing total degree comeJudge whether the target video is doubtful low quality video.Specifically, it can equally preset for judging that the ratio isThe no predetermined number of times proportion threshold value in normal range (NR).In this way, the number completely watched when the target video with seenWhen seeing the ratio of total degree lower than predetermined number of times proportion threshold value, it is possible to determine that the target video is doubtful low quality video.
In another embodiment, the number that can also be reported from the target video judges the target videoIt whether is doubtful low quality video.Specifically, when the target video is higher than specified report frequency threshold value by report number,It can be determined that the target video is doubtful low quality video.The specified report frequency threshold value can be previously according to normal viewThe frequency number reported counts.For example, the average time that normal video is reported is 10.4 times, then the specified actReport frequency threshold value can be set to 11 times or get Geng Gao be arranged according to the actual situation.
Certainly, the above-mentioned data in broadcast information carry out the step of doubtful low quality video judges only for thisThe limited of application scheme enumerates, and in practical application scene, can also be judged for more data, the skill of the applicationArt scheme is not limited to enumerated several situations.
S3: after determining that the target video is doubtful low quality video, to the audio-frequency information and picture of the target videoAt least one of face information is identified, to judge that the target video whether there is specified qualitative character.
It in the present embodiment, can be further after being determined that the target video is doubtful low quality videoThe actual content of the target video is judged.Specifically, the actual content of the target video may include audio letterBreath and image information, then can for it is therein at least one further identified.Audio-frequency information and/or picture are believedIt ceases the purpose identified and is to identify wherein whether include the content for causing video quality to reduce.In the present embodiment, thisThe content that will lead to video quality reduction a bit may be collectively referred to as specified qualitative character.In this way, to the audio-frequency information of target videoAnd/or after image information is identified, it can judge in the target video with the presence or absence of specified qualitative character.
It in the present embodiment, may include background sound in video, monologue, the dialogue of personage etc. in the audio-frequency informationAudio-frequency information.In low quality video, machine sound mechanically broadcast voice information is often used.For example, in certain videos,It can read out pre-prepd manuscript is mechanically bright by way of machine synthesized voice, to constitute the audio letter of videoBreath.If there are a large amount of such machine sounds in video, bad viewing experience is often caused to user.Therefore, at thisApply in an embodiment, can be identified for the machine sound in audio-frequency information, which can be as wait knowOther specified qualitative character.Specifically, when identifying to audio-frequency information, the audio-frequency information can be divided into multiple soundsFrequency frame, and extract the audio feature vector of the audio frame.Specifically, the audio frame can by when window divided,The window length of window can be with flexible setting when described.This when window can be translated on the time shaft of audio-frequency information, every time can willIn when window in audio-frequency information as an audio frame.After audio-frequency information is divided into multiple audio frames, it can extractThe audio feature vector of each audio frame.The audio feature vector can carry out Fourier by the data to the audio frameTransformation obtains.After Fourier transformation, available a series of discrete value, these discrete values can be used as element vectorElement, to constitute the audio feature vector.Certainly, in practical applications, the discrete value quantity after Fourier transformation is more, isThe dimension for reducing audio feature vector, the discrete value after Fourier transformation can be grouped, and every group by calculating energyThe mode of value obtains an energy value.Obtained energy value may finally be constituted audio feature vector, so as to greatlyReduce the dimension of audio feature vector.
In the present embodiment, the audio feature vector of each audio frame can be obtained by above-mentioned method, in this way,The audio feature vector based on extraction, can construct the characteristic sequence of the audio-frequency information.It specifically, can will be eachThe audio feature vector of audio frame is arranged successively, to constitute the characteristic sequence.That is, every in the characteristic sequenceA sequential value is an audio feature vector in fact.Since the characteristic sequence is obtained based on audio information,The characteristic sequence can characterize the audio-frequency information.Similar audio-frequency information, the obtained similarity between characteristic sequenceIt is often higher.In the present embodiment, the voice of machine sound for identification can be trained in a manner of first passing through machine learning in advanceIdentification model.Specifically, a large amount of audio-frequency informations being made of machine sound can be prepared in advance, then can extract these sounds respectivelyThe characteristic sequence of frequency information, then the characteristic sequence of extraction is inputted in the speech recognition modeling.The speech recognition modeling can be seenA neural network is done, has various neurons in the neural network, these neurons can have respective operational parameter.It is logicalAfter neural network is crossed to the characteristic sequence calculating of input, an available prediction probability value, which can be with tableProbability of the characteristic sequence of sign input as the characteristic sequence of machine sound.In this way, can will the obtained result of prediction with it is actualAs a result it is compared, and seeks difference.The parameter in speech recognition modeling can be corrected by the difference, so thatAgain when the input corresponding characteristic sequence of machine sound, prediction result is consistent with actual result.In this way, passing through a large amount of machine soundAfter audio-frequency information is trained, the speech recognition modeling can accurately differentiate input characteristic sequence whether be machine sound spyLevy sequence.Therefore, in the present embodiment, the characteristic sequence of building can be inputted to above-mentioned default speech recognition modelingIn, so as to obtain the corresponding prediction probability value of the characteristic sequence.It is specified when the prediction probability value is more than or equal toWhen probability threshold value, it can determine that the target video has the specified qualitative character of characterization machine sound.The specified probability thresholdValue can be according to presetting what training result of the speech recognition modeling in the training stage counted, if the training stage shows machineThe corresponding prediction probability value of the characteristic sequence of device sound is 0.8 or more, then can be using 0.8 or slightly lower value as described inSpecified probability threshold value.
In one embodiment, it advertising information can also identify present in the image information to target video.For example, can in image information advertisement link, promote the information such as text and identify.In this way, the specified qualitative characterIt can be advertising information.Specifically, in the present embodiment, optical character identification (Optical Character can be passed throughRecognition, OCR) etc. identify the text information for including in the image information in each frame image of character identifying methods.It, can be to text in order to distinguish advertising information and normal subtitle or literal with special effect after obtaining the text informationWord information is further identified.Specifically, the text information can be segmented, to obtain the text informationAt least one corresponding vocabulary.It, can be using pre-set lexicon to the word in the text information when being segmentedRemittance is identified, so as to identify to obtain multiple vocabulary in the text information.In practical applications, it can use variousSegmenter segments text information.The segmenter for example can be friso segmenter, Jcseg segmenter, MMSEG4JSegmenter etc..Further, in order to improve the accuracy to advertising information identification, building can be remitted based on common advertising wordsThe dictionary of segmenter, so that the result of segmenter output can be more in line with the speech habits of advertisement vocabulary.In this implementationIn mode, after obtaining each vocabulary in text information, the vocabulary that participle can be obtained is in default lexiconIt is matched.Various advertisement common words can be converged in the default lexicon, in this way, if the number for the vocabulary that matching obtainsWhen measuring shared ratio in the total quantity for the vocabulary that participle obtains more than or equal to designated ratio threshold value, comment informationIn include advertisement vocabulary it is quite a lot of, at this point it is possible to determine in the target video exist characterization advertising information specified qualityFeature.
In one embodiment, the blank screen frame occurred in target video can also be identified.Specifically, it is identifyingWhen blank screen frame, it can be identified by the gray value of image frame.If the average gray value of image frame is higher, illustrate image frameCompare dark, as blank screen frame a possibility that is higher.Therefore, in the present embodiment, can calculate in the image information whenThe average gray value of previous frame picture.Specifically, the gray value that can count each pixel in the present frame picture, is then askedThe average value of these gray values is taken, to obtain the average gray value.It is specified when the average gray value is more than or equal toWhen gray threshold, it is possible to determine that the target video has the specified qualitative character of characterization blank screen frame.Wherein, the specified gray scaleThreshold value, which can be according to the average gray value of image frame in normal video, to be determined.In determining method and above embodimentThe method of elaboration is similar, just repeats no more here.
In one embodiment, picture static in target video can also be identified.In certain videos, meetingIn the presence of a large amount of static pictures, the viewing experience of spectators also will affect in this way.Specifically, phase in the image information can be calculatedA possibility that similarity between adjacent two frame pictures, similarity is higher, and adjacent two frames picture is as still picture, is then bigger.It is countingWhen calculating the similarity between adjacent two frames picture, the feature vector of each frame picture can be extracted respectively.In described eigenvectorVector element can be the gray value of each pixel in picture.For example, each frame picture includes 1000 in target videoThen a pixel can draw this 1000 pixels then the gray value of this 1000 pixels can be determined respectivelyThe putting in order as gray value that put in order in face, to obtain the feature vector being made of gray value.Adjacent two frame is drawnSimilarity between face can be obtained by calculating two feature vectors in the distance of vector space.Closer, the similarity of distanceIt is bigger.When the similarity of calculating is more than or equal to two frame picture of specified similarity threshold, can determine describedThere is the specified qualitative character of characterization still picture in target video.
S5: obtaining decision threshold associated with the specified qualitative character, and based on the decision threshold, described in judgementWhether specified qualitative character belongs to anomalous mass feature.
In the present embodiment, it identifies after specifying qualitative character present in the target video, can not directly sentenceThe fixed target video is low quality video, if because the frequency for specifying qualitative character to occur in target video is not high,The quality of video can't be seriously affected.It in the present embodiment can be different to accurately assess the quality of videoDifferent decision thresholds is arranged in specified qualitative character.Specified qualitative character and associated decision threshold can be pre-stored within serviceIn device, when needing to specify qualitative character to judge some, server can be read specifies qualitative character associated with thisDecision threshold.
In the present embodiment, the decision threshold can be what the data based on normal video were analyzed.The judgementThreshold value can be used as the critical value for measuring normal quality feature and anomalous mass feature.Therefore, it is based on the decision threshold, it can be withJudge whether the specified qualitative character belongs to anomalous mass feature.
Specifically, for the specified qualitative character of machine sound, decision threshold associated with the specified qualitative characterThe maximum time limit value that value can occur in the target video for the specified qualitative character of characterization machine sound.For example, described sentenceDetermine threshold value to show only to allow the machine sound within appearance 20 seconds in a video, then 20 seconds can be as in the durationLimit value.In this way, when the total duration that the specified qualitative character occurs in the target video is greater than the maximum time limit value,The specified qualitative character can be determined for anomalous mass feature.
In another embodiment, related to the specified qualitative character for the specified qualitative character of advertising informationThe decision threshold of connection can be the maximum number of times value that the specified qualitative character occurs in the target video.For example, describedDecision threshold, which is limited in a video, only to be allowed 3 advertising informations occur, then 3 times can be as above-mentioned maximum number of timesValue.So, when the total degree that the specified qualitative character of characterization advertising information occurs in the target video is greater than the numberWhen upper limit value, the specified qualitative character can be determined for anomalous mass feature.
In another embodiment, associated with the specified qualitative character for the specified qualitative character of blank screenDecision threshold can be the frame number upper limit value that the specified qualitative character occurs in the target video.For example, the judgementThreshold value, which is limited in a video, only to be allowed 10 frame blank screens occur, then 10 frames can be as above-mentioned frame number upper limit value.That, can in the target video statistical average gray value be more than or equal to the specified gray threshold image frame it is totalFrame number can determine that the specified qualitative character is exception when the totalframes of statistics is greater than the frame number upper limit valueQualitative character.
In another embodiment, related to the specified qualitative character for the specified qualitative character of static imageThe decision threshold of connection may be the frame number upper limit value that the specified qualitative character occurs in the target video.For example, instituteStating decision threshold and being limited in a video only allows 10 frame static images occur, then 10 frames can be as above-mentioned frame numberUpper limit value.It is possible to which the two frame pictures that similarity is more than or equal to the specified similarity threshold are drawn labeled as targetFace, and count the totalframes of target picture described in the image information of the target video.When the totalframes is greater than the frameWhen number upper limit value, the specified qualitative character can be determined for anomalous mass feature.
S7: based on the judging result for being directed to the specified qualitative character, determine whether the target video is low quality viewFrequently.
In the present embodiment, it after judging the specified qualitative character in the target video, can countThe total quantity of anomalous mass feature present in the target video out.In this way, when the total quantity of statistics is greater than or waitsWhen specified quantity threshold value, the target video can be determined for low quality video.The specified quantity threshold value can basisActual conditions are adjusted, if quality audit is stringenter, which can be set relatively small.
In one embodiment of the application, it is contemplated that the duration of entire video may be long, to the sound of entire videoFrequency information and image information, which carry out identification, may expend considerable resource and considerable time.So in this embodiment partyIt, can be in conjunction with operation behavior data of the user when watching target video, to determine the playing duration area for needing to identify in formulaBetween.Specifically, the operation behavior data can characterize user when watching target video, in the broadcasting page of the target videoThe operation behavior of upper execution.In practical applications, the operation behavior data of user can be obtained by the way of burying a little.SpecificallyGround, the part control in the page of video playback website can be bound with program coding in advance.When the control is by userWhen triggering, the program coding of binding can execute automatically, to record this operation behavior of user.Wherein, the controlIt can be the page elements that can be interacted with user.For example, the control can be the broadcasting button of video, can also beThe button that can be dragged on progress bar can also be the hyperlink etc. in the page.For example bright, it can in the broadcasting button of videoThe program coding of behavior is clicked for obtaining user with preparatory binding, in this way, when the broadcasting button is clicked by user, programCoding can be performed automatically, to record this click behavior and can recorde the timing node of click behavior generation.RecordThe terminal device that can be used by target user of these information be automatically sent in the server of video playback website.In this way,These information being sent in server can be as the operation behavior data of user.
It in the present embodiment, can be according to the operation behavior data after obtaining the operation behavior dataDetermine corresponding playing duration section when the target video is terminated.Specifically, user close the target video orWhen from the page jump of the target video to another page, when can recorde corresponding broadcasting at this time by way of burying a littleIntermediate node.It is then possible to select certain duration range to constitute the playing duration area centered on the play time nodeBetween.For example, user closes target video when target video is played to 3 points and 10 seconds, then server can be automatically by 3 minutesTo 3 points of 20 seconds playing duration sections as duration section to be identified.It is subsequent, it can be broadcast by above-mentioned mode describedAt least one of the audio-frequency information and image information put in duration section are identified.In this way, can save considerableComputing resource and calculating duration.
It in another embodiment, can also can to obtain in comment area or the comment information in barrage area for userThere can be the timing node of mass defect.Specifically, the available user comment information for the target video, and from instituteThe vocabulary identified in user comment information for characterizing the specified qualitative character is stated, and is mentioned from the user comment informationTake timing node associated with the vocabulary.Wherein, for characterize the vocabulary of the specified qualitative character for example can be it is " blackScreen ", " noise ", vocabulary as " motionless ".These vocabulary can characterize corresponding mass defect.For example, a userComment is " why just blocking to 3 points of 10 seconds pictures motionless? ", " motionless " can be recognized in this way from the comment informationVocabulary, by semantic analysis, timing node associated with the vocabulary is " 3 points and 10 seconds ", then 3 points of 10 seconds or so mesh of explanationThere is mass defect in mark video.In this way, the playing duration section comprising the timing node can be determined, and broadcast describedAt least one of the audio-frequency information and image information put in duration section are identified.When determining the playing duration section,Certain duration range can be selected to constitute the playing duration area centered on from the timing node identified in comment informationBetween.For example, can be using 3 minutes to 3 points 20 seconds playing duration sections as the duration section for identifying specified qualitative character.
Referring to Fig. 3, the application also provides a kind of identifying system of video quality, the system comprises memories and processingDevice stores computer program in the memory and performs the steps of when the computer program is executed by the processor
S1: the broadcast information of target video is obtained, and is based on the broadcast information, judges whether the target video is doubtfulLike low quality video；
S3: after determining that the target video is doubtful low quality video, to the audio-frequency information and picture of the target videoAt least one of face information is identified, to judge that the target video whether there is specified qualitative character；
S5: obtaining decision threshold associated with the specified qualitative character, and based on the decision threshold, described in judgementWhether specified qualitative character belongs to anomalous mass feature；
S7: based on the judging result for being directed to the specified qualitative character, determine whether the target video is low quality viewFrequently.
In the present embodiment, the broadcast information includes the number and watched always that the target video is completely watchedThe ratio of number；Correspondingly, it when the computer program is executed by the processor, also performs the steps of
When the number that the target video is completely watched is lower than predetermined number of times ratio threshold with by the ratio of viewing total degreeWhen value, determine that the target video is doubtful low quality video.
In the present embodiment, it when the computer program is executed by the processor, also performs the steps of
The audio-frequency information is divided into multiple audio frames, and extracts the audio feature vector of the audio frame；
The audio feature vector based on extraction, constructs the characteristic sequence of the audio-frequency information；
The characteristic sequence of building is inputted in default speech recognition modeling, the corresponding prediction of the characteristic sequence is obtainedProbability value；
When the prediction probability value is more than or equal to specified probability threshold value, determining the target video, there are the fingersDetermine qualitative character.
In the present embodiment, it when the computer program is executed by the processor, also performs the steps of
Operation behavior data when user watches the target video are obtained, and determine institute according to the operation behavior dataState corresponding playing duration section when target video is terminated；
At least one of audio-frequency information and image information in the playing duration section are identified.
In the present embodiment, the memory may include the physical unit for storing information, usually by informationIt is stored again with the media using the methods of electricity, magnetic or optics after digitlization.Memory described in present embodiment again may be usedTo include: to store the device of information, such as RAM, ROM in the way of electric energy；The device of information is stored in the way of magnetic energy, it is such as hardDisk, floppy disk, tape, core memory, magnetic bubble memory, USB flash disk；Using the device of optical mode storage information, such as CD or DVD.Certainly, there are also memories of other modes, such as quantum memory, graphene memory etc..
In the present embodiment, the processor can be implemented in any suitable manner.For example, the processor can be withTake such as microprocessor or processor and storage can by (micro-) processor execute computer readable program code (such asSoftware or firmware) computer-readable medium, logic gate, switch, specific integrated circuit (Application SpecificIntegrated Circuit, ASIC), programmable logic controller (PLC) and the form etc. for being embedded in microcontroller.
The specific function that the identifying system for the video quality that this specification embodiment provides, memory and processor are realizedCan, explanation can be contrasted with the aforementioned embodiments in this specification, and the technical effect of aforementioned embodiments can be reached,Here it just repeats no more.
Therefore technical solution provided by the present application can preliminary root when the quality to target video identifiesAccording to the broadcast information of target video, judge whether target video is doubtful low quality video.The broadcast information can be video and broadcastIt is laid flat what platform was counted for the broadcasting situation of target video.For example, to can be video complete for the broadcast informationThe number or video of broadcasting are averaged the information such as the duration of viewing.Determined target video be doubtful low quality video itAfterwards, further the audio-frequency information of target video and/or image information can be identified.In identification, can be directed toCurrent common several factors for causing video quality to reduce, respectively detect target video.These lead to video qualityReduced factor can be used as specified qualitative character.For example, qualitative character is specified to can be the machine sound in target video, advertisementInformation, blank screen, still image etc..After there is specified qualitative character in recognizing target video, for different specified matterMeasure feature can have different decision thresholds.In this way, available decision threshold associated with specified qualitative character, andJudge whether specified qualitative character has exceeded normal tolerance according to the decision threshold, to judge the specified qualitative characterIt whether is anomalous mass feature.Finally, in conjunction with the judging result for being directed to each specified qualitative character, the target video can be determinedIt whether is low quality video.Therefore technical solution provided by the present application, can detect in target video from many aspects isNo there are anomalous mass features, so as to assess more fully hereinafter the quality of target video.
In the 1990s, the improvement of a technology can be distinguished clearly be on hardware improvement (for example,Improvement to circuit structures such as diode, transistor, switches) or software on improvement (improvement for method flow).SoAnd with the development of technology, the improvement of current many method flows can be considered as directly improving for hardware circuit.Designer nearly all obtains corresponding hardware circuit by the way that improved method flow to be programmed into hardware circuit.CauseThis, it cannot be said that the improvement of a method flow cannot be realized with hardware entities module.For example, programmable logic device(Programmable Logic Device, PLD) (such as field programmable gate array (Field Programmable GateArray, FPGA)) it is exactly such a integrated circuit, logic function determines device programming by user.By designerVoluntarily programming comes a digital display circuit " integrated " on a piece of PLD, designs and makes without asking chip makerDedicated IC chip.Moreover, nowadays, substitution manually makes IC chip, this programming is also used instead mostly " is patrolledVolume compiler (logic compiler) " software realizes that software compiler used is similar when it writes with program development,And the source code before compiling also write by handy specific programming language, this is referred to as hardware description language(Hardware Description Language, HDL), and HDL is also not only a kind of, but there are many kind, such as ABEL(Advanced Boolean Expression Language)、AHDL(Altera Hardware DescriptionLanguage)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(RubyHardware Description Language) etc., VHDL (Very-High-Speed is most generally used at presentIntegrated Circuit Hardware Description Language) and Verilog2.Those skilled in the artIt will be apparent to the skilled artisan that only needing method flow slightly programming in logic and being programmed into integrated circuit with above-mentioned several hardware description languagesIn, so that it may it is readily available the hardware circuit for realizing the logical method process.
It is also known in the art that the identification in addition to realizing video quality in a manner of pure computer readable program codeOther than system, completely can by by method and step carry out programming in logic come so that video quality identifying system with logic gate,Switch, specific integrated circuit, programmable logic controller (PLC) and insertion microcontroller etc. form realize identical function.Therefore thisThe identifying system of kind of video quality is considered a kind of hardware component, and to including for realizing various functions in itDevice can also be considered as the structure in hardware component.Or even, both can may be used being considered as realizing the device of various functionsTo be that the software module of implementation method can be the structure in hardware component again.
As seen through the above description of the embodiments, those skilled in the art can be understood that the application canIt realizes by means of software and necessary general hardware platform.Based on this understanding, the technical solution essence of the applicationOn in other words the part that contributes to existing technology can be embodied in the form of software products, the computer software productIt can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer equipment(can be personal computer, server or the network equipment etc.) executes each embodiment of the application or embodimentMethod described in certain parts.
Each embodiment in this specification is described in a progressive manner, same and similar between each embodimentPart may refer to each other, what each embodiment stressed is the difference with other embodiments.In particular, needleFor the embodiment of the identifying system of video quality, it is referred to the introduction control solution of the embodiment of preceding methodIt releases.
The application can describe in the general context of computer-executable instructions executed by a computer, such as programModule.Generally, program module includes routines performing specific tasks or implementing specific abstract data types, programs, objects, groupPart, data structure etc..The application can also be practiced in a distributed computing environment, in these distributed computing environments, byTask is executed by the connected remote processing devices of communication network.In a distributed computing environment, program module can be withIn the local and remote computer storage media including storage equipment.
Although depicting the application by embodiment, it will be appreciated by the skilled addressee that there are many deformations by the applicationWith variation without departing from spirit herein, it is desirable to which the attached claims include these deformations and change without departing from the applicationSpirit.