CN110110257A

CN110110257A - Data processing method and its system, computer system and computer-readable medium

Info

Publication number: CN110110257A
Application number: CN201810088574.7A
Authority: CN
Inventors: 张尚志; 易朋; 王江洪
Original assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Current assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Priority date: 2018-01-30
Filing date: 2018-01-30
Publication date: 2019-08-09
Anticipated expiration: 2038-01-30
Also published as: CN110110257B

Abstract

Present disclose provides a kind of data processing methods, comprising: receives object query request, wherein object query request is for requesting inquiry target object；In response to object query request, obtain with the matched multiple objects of target object and for showing picture used in each object in multiple objects；The picture of acquisition is analyzed, effect is presented with the vision of each picture of determination；And effect is presented in the vision based on each picture, determines layout of each picture in object displayed page.In addition, the disclosure additionally provides a kind of data processing system, a kind of computer system and a kind of computer-readable medium.

Description

Data processing method and its system, computer system and computer-readable medium

Technical field

This disclosure relates to data processing field, more particularly, to a kind of data processing method and its system and one kind Computer system and a kind of computer-readable medium.

Background technique

With will be used wider and wider for network, user can be by the interested any target pair of network inquiry oneself As, for any one query behavior, there may be the hundreds of thousands of and matched object of target object to be returned and show user, The sequence for how realizing above-mentioned hundreds of thousands of objects has vital effect to user experience is promoted.By taking scene of doing shopping as an example, The affairs that sort occur after user carries out one query, and search engine shows the object of return in user interface (User Interface before).One query for user may have the multiple objects being consistent with querying condition to be returned, how real An existing ideal sequence, the commodity for favoring user appear in the head end of sequence with maximum probability, so as to shorten " search-point Hit-place an order " process chain, improve conversion ratio be of great importance.

However, at least there are the following problems in the related technology: phase for inventor's discovery during realizing disclosure design Pass technology provides a variety of sequencing schemes, but either traditional sequence based on artificial rule or emerging based on machine The sequence of device learning model is all based on what statistical data was ranked up, and sort less effective.

In view of the above problems in the related art, it does not put forward effective solutions also at present.

Summary of the invention

The first aspect of the disclosure provides a kind of data processing method, comprising: receives object query request, wherein Above-mentioned object query request is for requesting inquiry target object；In response to above-mentioned object query request, obtain and above-mentioned target pair As matched multiple objects and for showing picture used in each object in above-mentioned multiple objects；To the picture of acquisition into Effect is presented with the vision of each picture of determination in row analysis；And effect is presented in the vision based on above-mentioned each picture, determines above-mentioned each Layout of the picture in object displayed page.

In accordance with an embodiment of the present disclosure, the above method further include: in acquisition and the matched multiple objects of above-mentioned target object The operation data of each object；And effect is presented based on aforesaid operations data and the vision of above-mentioned each picture, it determines above-mentioned each Layout of the picture in above-mentioned object displayed page.

In accordance with an embodiment of the present disclosure, the picture of acquisition is analyzed, effect packet is presented with the vision of each picture of determination It includes: extracting above-mentioned each picture corresponding feature list under different scale space；To above-mentioned each picture under different scale space Corresponding features described above list is standardized, to obtain standardized feature list；And it is based on above-mentioned standard feature List determines that effect is presented in the vision of above-mentioned each picture.

In accordance with an embodiment of the present disclosure, extracting above-mentioned each picture corresponding feature list under different scale space includes: Gaussian Blur processing is carried out under different sizes to above-mentioned each picture, to obtain the height of different scale corresponding with above-mentioned each picture This pyramid picture；Each scale picture in gaussian pyramid picture corresponding with above-mentioned each picture is traversed, determines different scale The candidate point of corresponding characteristic point, wherein above-mentioned candidate point is the pixel for belonging to extreme value in this scale and adjacent scale；It is based on Above-mentioned candidate point carries out the screening and accurate positioning of characteristic point, wherein above-mentioned screening and accurate positioning are at least including fitting extreme value The offset and elimination skirt response of point；The characteristic point and its neighborhood obtained based on above-mentioned screening generates above-mentioned each characteristic point Feature Descriptor under its correspondence scale；And it is the Feature Descriptor of generation is right under different scale as above-mentioned each picture The feature list answered.

In accordance with an embodiment of the present disclosure, to above-mentioned each picture, corresponding features described above list carries out standard under different scale Change processing, includes: to choose in corresponding feature list under different scale from above-mentioned each picture to obtain standardized feature list At least one feature is as a feature class, wherein at least one above-mentioned characteristic storage is in initialization feature list；In calculating State the distance between other features and at least one the above-mentioned feature in feature list in addition at least one above-mentioned feature；It will be minimum Feature in corresponding features described above list is determined as corresponding feature normalization processing result；By above-mentioned corresponding feature Standardization result is updated in above-mentioned initialization feature list；And it will be all in updated initialization feature list The feature feature class new as one.

In accordance with an embodiment of the present disclosure, above-mentioned standard feature list includes multiple, is based on above-mentioned standard feature list, Determining that effect is presented in the vision of above-mentioned each picture includes: to determine each above-mentioned standard based on multiple above-mentioned standard feature lists Change active degree of each feature in this standard feature list in feature list；It determines in each above-mentioned standard feature list Popularity of each feature in all standardized feature lists；And it is based on above-mentioned active degree and above-mentioned popularity, really Effect is presented in the vision of fixed above-mentioned each picture.

In accordance with an embodiment of the present disclosure, effect is presented based on aforesaid operations data and the vision of above-mentioned each picture, in determination Stating layout of each picture in above-mentioned object displayed page includes: to be determined to above-mentioned each picture based on aforesaid operations data upper State the first layout information being laid out in object displayed page；Effect is presented based on above-mentioned vision, determines to above-mentioned each picture The second layout information and respective weights value being laid out in above-mentioned object displayed page；And believed according to above-mentioned first layout Breath and the second layout information and above-mentioned respective weights value, determine layout of the above-mentioned each picture in above-mentioned object displayed page.

In accordance with an embodiment of the present disclosure, the above method further include: determining above-mentioned each picture in above-mentioned object displayed page In layout after, determine whether the similarity of continuously arranged plurality of pictures in above-mentioned object displayed page meets preset condition； And in the case where satisfaction, layout of the above-mentioned continuously arranged plurality of pictures in above-mentioned object displayed page is adjusted It is whole.

The second aspect of the disclosure provides a kind of data processing system, comprising: receiving module is looked into for receiving object Ask request, wherein above-mentioned object query request is for requesting inquiry target object；First obtains module, in response to above-mentioned Object query request is obtained with the matched multiple objects of above-mentioned target object and each right in above-mentioned multiple objects for showing As used picture；Effect is presented for analyzing the picture of acquisition with the vision of each picture of determination in first determining module Fruit；And second determining module, effect is presented for the vision based on above-mentioned each picture, determines that above-mentioned each picture is shown in object Layout in the page.

In accordance with an embodiment of the present disclosure, above system further include: second obtains module, for obtaining and above-mentioned target object The operation data of each object in matched multiple objects；And third determining module, for based on aforesaid operations data and Effect is presented in the vision for stating each picture, determines layout of the above-mentioned each picture in above-mentioned object displayed page.

In accordance with an embodiment of the present disclosure, above-mentioned first determining module includes: extraction unit, is existed for extracting above-mentioned each picture Corresponding feature list under different scale space；Processing unit, for corresponding under different scale space to above-mentioned each picture Features described above list is standardized, to obtain standardized feature list；And first determination unit, for based on above-mentioned Standardized feature list determines that effect is presented in the vision of above-mentioned each picture.

In accordance with an embodiment of the present disclosure, said extracted unit includes: the first processing subelement, for existing to above-mentioned each picture Gaussian Blur processing is carried out under different sizes, to obtain the gaussian pyramid picture of different scale corresponding with above-mentioned each picture； First determines subelement, and for traverse each scale picture in gaussian pyramid picture corresponding with above-mentioned each picture, determination is not The candidate point of corresponding characteristic point with scale, wherein above-mentioned candidate point is the pixel for belonging to extreme value in this scale and adjacent scale Point；Second processing subelement carries out the screening and accurate positioning of characteristic point, wherein above-mentioned screening for being based on above-mentioned candidate point It at least include the offset and elimination skirt response for being fitted extreme point with being accurately positioned；Subelement is generated, for based on based on institute The characteristic point and its neighborhood that screening obtains are stated, Feature Descriptor of the above-mentioned each characteristic point under its correspondence scale is generated；And Third handles subelement, the corresponding characteristic series under different scale as above-mentioned each picture of the Feature Descriptor for that will generate Table.

In accordance with an embodiment of the present disclosure, above-mentioned processing unit includes: selection subelement, is used for from above-mentioned each picture in difference At least one feature is chosen under scale in corresponding feature list as a feature class, wherein at least one above-mentioned feature is deposited Storage is in initialization feature list；Computation subunit, for calculating in features described above list in addition at least one above-mentioned feature The distance between other features and at least one above-mentioned feature；Second determines subelement, for minimum range is corresponding above-mentioned Feature in feature list is determined as corresponding feature normalization processing result；Subelement is updated, is used for above-mentioned corresponding spy Sign standardization result is updated in above-mentioned initialization feature list；And fourth process subelement, being used for will be updated All features in initialization feature list feature class new as one.

In accordance with an embodiment of the present disclosure, above-mentioned first determination unit includes: that third determines subelement, for based on multiple Standardized feature list is stated, determines each feature enlivening in this standard feature list in each above-mentioned standard feature list Degree；4th determines subelement, arranges for determining in each above-mentioned standard feature list each feature in all standardized features Popularity in table；And the 5th determine subelement, for be based on above-mentioned active degree and above-mentioned popularity, determine above-mentioned Effect is presented in the vision of each picture.

In accordance with an embodiment of the present disclosure, above-mentioned second determining module includes: the second determination unit, for being based on aforesaid operations Data determine the first layout information being laid out in above-mentioned object displayed page to above-mentioned each picture；Third determination unit, For effect to be presented based on above-mentioned vision, the second cloth being laid out in above-mentioned object displayed page to above-mentioned each picture is determined Office's information and respective weights value；And the 4th determination unit, for according to above-mentioned first layout information and the second layout information With above-mentioned respective weights value, layout of the above-mentioned each picture in above-mentioned object displayed page is determined.

In accordance with an embodiment of the present disclosure, above system further include: third determining module, for determining that above-mentioned each picture exists After layout in above-mentioned object displayed page, determine that the similarity of continuously arranged plurality of pictures in above-mentioned object displayed page is It is no to meet preset condition；And adjustment module, it is used in the case where satisfaction, to above-mentioned continuously arranged plurality of pictures above-mentioned Layout in object displayed page is adjusted.

A kind of computer system is provided in terms of the third of the disclosure, comprising: one or more processors；Storage dress It sets, for storing one or more programs, wherein when one or more programs are executed by one or more processors, so that The data processing method of one or more processors realization any of the above-described.

4th aspect of the disclosure provides a kind of computer-readable medium, is stored thereon with executable instruction, this refers to Enable the data processing method for making processor realize any of the above-described when being executed by processor.

By embodiment of the disclosure, as being used to show used in each object in multiple objects using to acquisition Picture is analyzed, and effect is presented with the vision presentation effect of each picture of determination and the vision based on each picture, determines each figure The technical solution of layout of the piece in object displayed page can make up and determine that each picture is shown in object in the related technology Visual effect considers the blank of aspect when layout in the page, can at least partly overcome in the related technology due to based on statistical number According to determination each picture sort caused by the layout in object displayed page less effective the technical issues of, by estimating figure The visual effect of piece influences user behavior bring, adjusts layout of multiple objects in object displayed page, realizes and is promoted Picture sequence effect, and then realize the technical effect for being promoted and clicking conversion ratio.

Detailed description of the invention

In order to which the disclosure and its advantage is more fully understood, referring now to being described below in conjunction with attached drawing, in which:

Fig. 1 diagrammatically illustrates the exemplary system architecture suitable for data processing method according to the embodiment of the present disclosure；

Fig. 2 diagrammatically illustrates the flow chart of the data processing method according to the embodiment of the present disclosure；

Fig. 3 A diagrammatically illustrates the flow chart of the data processing method according to another embodiment of the disclosure；

Fig. 3 B diagrammatically illustrates analyzing the picture of acquisition according to the embodiment of the present disclosure, with each picture of determination The flow chart of vision presentation effect；

Fig. 3 C diagrammatically illustrates according to each picture of extraction of the embodiment of the present disclosure the corresponding spy under different scale space Levy the flow chart of list；

Fig. 3 D diagrammatically illustrates according to each picture of extraction of the embodiment of the present disclosure the corresponding spy under different scale space Levy the effect picture of list；

Fig. 3 E diagrammatically illustrates according to the embodiment of the present disclosure the corresponding feature list under different scale to each picture It is standardized, to obtain the flow chart of standardized feature list；

Fig. 3 F is diagrammatically illustrated according to the embodiment of the present disclosure based on standardized feature list, determines the vision of each picture The flow chart of effect is presented；

Fig. 3 G, which is diagrammatically illustrated, is presented effect based on operation data and the vision of each picture according to the embodiment of the present disclosure, Determine the flow chart of layout of each picture in object displayed page；

Fig. 3 H diagrammatically illustrates the flow chart of the data processing method of the another embodiment according to the embodiment of the present disclosure；

Fig. 4 diagrammatically illustrates the block diagram of the data processing system according to the embodiment of the present disclosure；

Fig. 5 A diagrammatically illustrates the block diagram of the data processing system according to another embodiment of the disclosure；

Fig. 5 B diagrammatically illustrates the block diagram of the first determining module according to the embodiment of the present disclosure；

Fig. 5 C diagrammatically illustrates the block diagram of the extraction unit according to the embodiment of the present disclosure；

Fig. 5 D diagrammatically illustrates the block diagram of the processing unit according to the embodiment of the present disclosure；

Fig. 5 E diagrammatically illustrates the block diagram of the first determination unit according to the embodiment of the present disclosure；

Fig. 5 F diagrammatically illustrates the block diagram of the second determining module according to the embodiment of the present disclosure；

Fig. 5 G diagrammatically illustrates the block diagram of the data processing system of the another embodiment according to the embodiment of the present disclosure；And

Fig. 6 diagrammatically illustrates the block diagram for being suitable for the computer system for being used to realize the embodiment of the present disclosure.

Specific embodiment

Hereinafter, will be described with reference to the accompanying drawings embodiment of the disclosure.However, it should be understood that these descriptions are only exemplary , and it is not intended to limit the scope of the present disclosure.In addition, in the following description, descriptions of well-known structures and technologies are omitted, with Avoid unnecessarily obscuring the concept of the disclosure.

Term as used herein is not intended to limit the disclosure just for the sake of description specific embodiment.It uses herein The terms "include", "comprise" etc. show the presence of the feature, step, operation and/or component, but it is not excluded that in the presence of Or add other one or more features, step, operation or component.

There are all terms (including technical and scientific term) as used herein those skilled in the art to be generally understood Meaning, unless otherwise defined.It should be noted that term used herein should be interpreted that with consistent with the context of this specification Meaning, without that should be explained with idealization or excessively mechanical mode.

It, in general should be according to this using statement as " at least one in A, B and C etc. " is similar to Field technical staff is generally understood the meaning of the statement to make an explanation (for example, " system at least one in A, B and C " Should include but is not limited to individually with A, individually with B, individually with C, with A and B, with A and C, have B and C, and/or System etc. with A, B, C).Using statement as " at least one in A, B or C etc. " is similar to, generally come Saying be generally understood the meaning of the statement according to those skilled in the art to make an explanation (for example, " having in A, B or C at least One system " should include but is not limited to individually with A, individually with B, individually with C, with A and B, have A and C, have B and C, and/or the system with A, B, C etc.).It should also be understood by those skilled in the art that substantially arbitrarily indicating two or more The adversative conjunction and/or phrase of optional project shall be construed as either in specification, claims or attached drawing A possibility that giving including one of these projects, either one or two projects of these projects.For example, phrase " A or B " should A possibility that being understood to include " A " or " B " or " A and B ".

Shown in the drawings of some block diagrams and/or flow chart.It should be understood that some sides in block diagram and/or flow chart Frame or combinations thereof can be realized by computer program instructions.These computer program instructions can be supplied to general purpose computer, The processor of special purpose computer or other programmable data processing units, so that these instructions are when executed by this processor can be with Creation is for realizing function/operation device illustrated in these block diagrams and/or flow chart.

Therefore, the technology of the disclosure can be realized in the form of hardware and/or software (including firmware, microcode etc.).Separately Outside, the technology of the disclosure can take the form of the computer program product on the computer-readable medium for being stored with instruction, should Computer program product uses for instruction execution system or instruction execution system is combined to use.In the context of the disclosure In, computer-readable medium, which can be, can include, store, transmitting, propagating or transmitting the arbitrary medium of instruction.For example, calculating Machine readable medium can include but is not limited to electricity, magnetic, optical, electromagnetic, infrared or semiconductor system, device, device or propagation medium. The specific example of computer-readable medium includes: magnetic memory apparatus, such as tape or hard disk (HDD)；Light storage device, such as CD (CD-ROM)；Memory, such as random access memory (RAM) or flash memory；And/or wire/wireless communication link.

Present disclose provides a kind of data processing methods, comprising: receives object query request, wherein object query request For requesting to inquire target object, in response to object query request, obtains and the matched multiple objects of target object and be used for It shows that picture used in each object analyzes the picture of acquisition in multiple objects, is presented with the vision of each picture of determination Effect；And effect is presented in vision based on each picture, determines layout of each picture in object displayed page, it can at least portion Decigram clothes are in the related technology as determining each picture row caused by the layout in object displayed page based on statistical data The technical issues of sequence less effective, reaches from vision angle presentation and promotes the technical effect of picture sequence effect.

Fig. 1 diagrammatically illustrates the exemplary system architecture suitable for data processing method according to the embodiment of the present disclosure 100.It should be noted that being only the example that can apply the system architecture of the embodiment of the present disclosure shown in Fig. 1, to help this field Technical staff understands the technology contents of the disclosure, but be not meant to the embodiment of the present disclosure may not be usable for other equipment, system, Environment or scene.

As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network according to this embodiment 104 and server 105.Network 104 between terminal device 101,102,103 and server 105 to provide communication link Medium.Network 104 may include various connection types, such as wired and or wireless communications link etc..

User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out Send message etc..Various telecommunication customer end applications, such as the application of shopping class, net can be installed on terminal device 101,102,103 (merely illustrative) such as the application of page browsing device, searching class application, instant messaging tools, mailbox client and/or social platform softwares.

Terminal device 101,102,103 can be the various electronic equipments with display screen and supported web page browsing, packet Include but be not limited to smart phone, tablet computer, pocket computer on knee and desktop computer etc..

Server 105 can be to provide the server of various services, such as utilize terminal device 101,102,103 to user The website browsed provides the back-stage management server (merely illustrative) supported.Back-stage management server can be to the use received The data such as family request analyze etc. processing, and by processing result (such as according to user's request or the webpage of generation, believe Breath or data etc.) feed back to terminal device.

It should be noted that data processing method provided by the embodiment of the present disclosure can generally be executed by server 105. Correspondingly, data processing system provided by the embodiment of the present disclosure generally can be set in server 105.The embodiment of the present disclosure Provided data processing method can also by be different from server 105 and can with terminal device 101,102,103 and/or clothes The server or server cluster that business device 105 communicates execute.Correspondingly, data processing system provided by the embodiment of the present disclosure It can be set in the service that is different from server 105 and can be communicated with terminal device 101,102,103 and/or server 105 In device or server cluster.

It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need It wants, can have any number of terminal device, network and server.

Fig. 2 diagrammatically illustrates the flow chart of the data processing method according to the embodiment of the present disclosure.

As shown in Fig. 2, the data processing method may include operation S210~S240, in which:

In operation S210, object query request is received.

It is obtained in response to object query request with the matched multiple objects of target object and for opening up in operation S220 Show picture used in each object in multiple objects.

In operation S230, the picture of acquisition is analyzed, effect is presented with the vision of each picture of determination.

In operation S240, the vision based on each picture is presented effect, determines layout of each picture in object displayed page.

In accordance with an embodiment of the present disclosure, object query request such as can be for requesting inquiry target object and pass through shopping Query site money or certain class commodity, can also be and inquire certain object by certain portal website, embodiment of the disclosure is to target Object is not specifically limited.In order to illustrate simplicity, below by by taking the commodity picture information that certain shopping website provides as an example to this Disclosed embodiment is illustrated, it should be understood that the Data Data method that the disclosure provides is not limited to this exemplary reality Example is applied, the data processing method of other scenes can do corresponding expansion according to the spirit of the disclosure, and details are not described herein again.

After receiving the object query request for requesting inquiry target object, search engine can be based on the Object Query Request, hit and return meet querying condition with the matched multiple objects of target object, also referred to as recall, it is multiple right to recall As that can be ranked up to above-mentioned multiple objects before UI showing interface, with cloth of each picture of determination in object displayed page Office.

It should be noted that being the master of multiple objects for showing picture used in each object in multiple objects Figure, so-called picture is master map below, before analyzing the picture of acquisition, the master map that shopping website can be provided Pictorial information processing be available format, weigh opening for influence and computing resource of the dimension of picture size to result precision Pin, this programme uniformly take 172*172 pixel as dimension of picture, but are not the limit to disclosure picture specific format and size It is fixed.

In accordance with an embodiment of the present disclosure, the picture of acquisition is analyzed, effect, base is presented with the vision of each picture of determination Effect is presented in the vision of each picture, determines layout of each picture in object displayed page.For visual experience, show The master map of the object of user is unique, which is easier to cause user behavior, such as browsing, purchase.Therefore, it can use Series of computation machine vision algorithm is based on Scale-space theory, detecting and the locality characteristic in description image, it is in space ruler Extreme point is found in degree, and extracts its position, size, rotational invariants, and effect is presented in the vision of predictive picture, if a certain figure The feature that piece shows and the characteristic difference of other pictures are bigger, mean that current image more can be different from other pictures, Corresponding goods are more intended to forward ranking.

By embodiment of the disclosure, as being used to show used in each object in multiple objects using to acquisition Picture is analyzed, and effect is presented with the vision presentation effect of each picture of determination and the vision based on each picture, determines each figure The technical solution of layout of the piece in object displayed page can make up and determine each picture in object displayed page in the related technology In layout when visual effect consider aspect blank, can at least partly overcome in the related technology due to based on statistical data it is true Fixed each picture sort caused by the layout in object displayed page less effective the technical issues of, pass through the vision of predictive picture Effect influences user behavior bring, adjusts layout of multiple objects in object displayed page, realizes and promotes picture sequence Effect, and then realize the technical effect for being promoted and clicking conversion ratio.

Below with reference to Fig. 3 A~Fig. 3 C and Fig. 3 E~Fig. 3 H, in conjunction with specific embodiments to data processing side shown in Fig. 2 Method is described further.

Fig. 3 A diagrammatically illustrates the flow chart of the data processing method according to another embodiment of the disclosure.

As shown in Figure 3A, the data processing method may include aforementioned operation S210~S240 and operation S311 and S312, wherein aforementioned operation S210~S240 is repeated no more, in which:

In operation S311, the operation data with each object in the matched multiple objects of target object is obtained.

In operation S312, effect is presented based on operation data and the vision of each picture, determines that each picture shows page in object Layout in face.

In accordance with an embodiment of the present disclosure, operation data can include but is not limited to be that numerous can measure two pieces commodity who is excellent Data spy may be implemented in who bad data characteristics (such as positive rating, price, businessman's prestige etc.), any of the relevant technologies offer Sign, by numerous visible or sightless rule, realize the method for final sequence output all in the protection scope of the disclosure, Such as traditional sort method based on artificial rule and the emerging sort method based on machine learning model are driven based on data Dynamic sequencing schemes, details are not described herein.

Sequencing schemes based on data characteristics more can objectively realize the sequence to commodity.But it is purchased in actual electric business During object is tested, visual stimulus is to influence another information that can not ignore of user's Shopping Behaviors.Existing sequencing schemes are to figure Considering for piece visual aspects is substantially zeroed, and having ignored commodity vision and dimension is presented is that ranking results bring upward mobility.

It is more reasonable simulation and the commodity selection experience of also original subscriber by embodiment of the disclosure, the disclosure is real It applies example the dimension that commodity picture feature sorts as electric business is used, two sides of effect is presented from data characteristics and vision Face, cooperation provide the ranking results of commodity.

Fig. 3 B diagrammatically illustrates analyzing the picture of acquisition according to the embodiment of the present disclosure, with each picture of determination The flow chart of vision presentation effect.

As shown in Figure 3B, the aforementioned operation S230 in Fig. 2 or Fig. 3 A (analyzes the picture of acquisition, with each figure of determination Effect is presented in the vision of piece) S321~S323 can be operated.Wherein:

In operation S321, each picture corresponding feature list under different scale space is extracted.

In operation S322, to each picture, corresponding feature list is standardized under different scale space, with To standardized feature list.

In operation S323, it is based on standardized feature list, determines that effect is presented in the vision of each picture.

In accordance with an embodiment of the present disclosure, the picture of acquisition is analyzed, effect is presented with the vision of each picture of determination, is The quantizing process that the intuitive visual signature of picture is converted to picture score, can use computer vision algorithms make to realize, This to specific algorithm without limitation, it is contemplated that the diversity of computer vision algorithms make will hereafter be enumerated a kind of relatively reasonable, easy It is used as explanation in the process of realization, it is not intended that present invention is limited only by the thinkings enumerated.The embodiment of the present disclosure With core ideas Scale invariant features transform algorithm (Scale-invariant feature transform, referred to as SIFT)+ For K-means+TF-IDF.It should be noted that similarity evaluation in K-means method can be referred to according to the actual situation Mark replaces with cosin distance by Euclidean distance, or K-means is replaced with another improved clustering algorithm, as long as in essence On, the thinking with the disclosure be it is identical, within the protection scope of the disclosure.

In accordance with an embodiment of the present disclosure, scale space is constructed first, is detected all existing under different scale spaces Characteristic point extracts each picture corresponding feature list under different scale space, it is contemplated that feature list is by pictorial information meter It obtains, identical feature hardly occurs.In order to calculate the similitude of feature list, need similar spy Unified representation is levied, therefore, is extracting each picture under different scale space after corresponding feature list, to each picture in difference Corresponding feature list is standardized under scale space, i.e., carries out cluster dimensionality reduction to the feature list under different scale, It realizes the cluster of feature, is finally based on cluster result, realize the quantization of picture uniqueness degree score, extended meeting after specific implementation process It is described in detail, details are not described herein again.

By embodiment of the disclosure, the extraction of feature list is carried out to each picture, and is standardized, to determine Effect is presented in the vision of each picture, has fully considered visual stimulus of the picture to user, and picture vision presentation effect is quantified as Picture uniqueness score provides intuitive data to layout of the picture in the page and supports, at least partly overcomes the relevant technologies In do not account for commodity vision present dimension to ranking results a possibility that.

Fig. 3 C diagrammatically illustrates according to each picture of extraction of the embodiment of the present disclosure the corresponding characteristic series under different scale The flow chart of table.

As shown in Figure 3 C, aforementioned operation S321 (extracting each picture corresponding feature list under different scale space) can be with Including operating S331~S335.Wherein:

In operation S331, Gaussian Blur processing is carried out under different sizes to each picture, it is corresponding with each picture to obtain The gaussian pyramid picture of different scale.

In operation S332, each scale picture in gaussian pyramid picture corresponding with each picture is traversed, determines different rulers Spend the candidate point of corresponding characteristic point, wherein candidate point is the pixel for belonging to extreme value in this scale and adjacent scale.

In operation S333, it is based on candidate point, carries out the screening and accurate positioning of characteristic point, wherein screening and accurate positioning Offset and elimination skirt response including at least fitting extreme point.

In operation S334, the characteristic point obtained based on screening and its neighborhood generate each characteristic point under its correspondence scale Feature Descriptor.

In operation S335, using the Feature Descriptor of generation as the feature list of each picture.

SIFT algorithm is a kind of algorithm for detecting local feature, and the algorithm is by seeking characteristic point in a width figure and its related Scaling, description rotated obtain feature and carry out Image Feature Point Matching.In accordance with an embodiment of the present disclosure, using SIFT algorithm Each picture corresponding feature list under different scale is extracted, realizes that detecting and the locality in the master map for describing multiple objects are special Sign, finds extreme point, and extract its position, size, rotational invariants in space scale.Dimension of picture size is weighed to knot The influence of fruit precision and the expense of computing resource, this programme uniformly take 172*172 pixel as dimension of picture.It is calculated based on SIFT Method, extracted from commodity master map SIFT feature the specific implementation process is as follows, Fig. 3 D is diagrammatically illustrated according to disclosure reality Apply the effect picture of each picture corresponding feature list under different scale of extraction of example.

A. it the generation of scale space: is sampled by the Gaussian Blur and dimensionality reduction that do different scale to image, creates picture Gaussian pyramid expression.

B. detect scale spatial extrema point: search for the picture position on all scale spaces, by gaussian derivative function come Identify the candidate point of the potentially characteristic point for scale and invariable rotary.

C. it is accurately positioned extreme point: on the position of each candidate point, position and ruler being determined by fitting refined model Degree, the degree of stability for being dependent on extreme point carry out selected characteristic point.

D. be each characteristic point assigned direction parameter: the gradient direction based on image local distributes to each feature point Set one or more directions.All subsequent operations to image data both relative to the direction of characteristic point, scale and position into Row transformation, to provide the invariance for these transformation.

E. the generation of feature point description: in the neighborhood around each characteristic point, image is measured on selected scale The gradient of part.These gradients are transformed into a kind of expression, this deformation and illumination for indicating to allow bigger local shape Variation.

The code of SIFT is realized to be completed using picture recognition Open-Source Tools packet.The disclosure is based on 4* in characteristic point scale space Feature Descriptor of the gradient information in 8 directions calculated in 4 window as characteristic point amounts to 4*4*8=128 dimensional vector table Sign.As the output of SIFT algorithm, the feature list of a commodity picture includes several each local features of picture, each feature Data structure be all one 128 dimension array.

By embodiment of the disclosure, the feature list of the corresponding each picture in different scale space is extracted, it can be accurately The feature that effect is presented in each picture vision can be represented by obtaining, to the greatest extent by the important information of picture and with discrimination Information is retained, while providing reliable data basis for next step standardized feature list.

Fig. 3 E diagrammatically illustrates according to the embodiment of the present disclosure the corresponding feature list under different scale to each picture It is standardized, to obtain the flow chart of standardized feature list.

Such as Fig. 3 E, aforementioned operation S322 (to each picture, corresponding feature list is standardized under different scale, To obtain standardized feature list) it may include operation S351~S355.Wherein:

In operation S351, at least one feature is chosen in corresponding feature list under different scale from each picture as one A feature class, wherein at least one characteristic storage is in initialization feature list.

In operation S352, calculate between other features and at least one feature in feature list in addition at least one feature Distance.

In operation S353, the feature in the corresponding feature list of minimum range is determined as corresponding feature normalization and is handled As a result.

In operation S354, corresponding feature normalization processing result is updated in initialization feature list.

In operation S355, using the feature class new as one of all features in updated initialization feature list.

In accordance with an embodiment of the present disclosure, each picture is being extracted under different scale after corresponding feature list, it can be right Each picture corresponding feature list under different scale is standardized, to obtain standardized feature list.The relevant technologies One kind is provided typically based on the clustering algorithm of distance, i.e. K-means cluster refers to using apart from the evaluation as similitude Mark, it is believed that the distance of two objects is closer, and similarity is bigger.The algorithm is compact and independent for clustering target object Several clusters.

It is clustered it should be noted that can use the set that K-means clustering algorithm forms SIFT feature, with reality The dimensionality reduction expression of existing feature list, the necessity of dimensionality reduction expression are: assuming that the feature list of commodity P1 be expressed as Fa1, Fb1, Fc1 ... }, the feature list of commodity P2 is expressed as { Fa2, Fb2, Fc2 ... }, these features are calculated by pictorial information, several It is not in identical feature.In order to calculate the similitude of feature list, need similar feature unifying table Show.Assuming that being a kind of (being denoted as Fa*) by Fa1 and Fa2 points by feature clustering, then it is assumed that after cluster dimensionality reduction, P1 is equal with P2 There are standard feature Fa*, thus have a degree of similitude.The quantity of cluster centre can by parameter optimization thinking, Minimum value is sought to evaluation index to find；But in view of cluster centre quantity has no significant effect SIFT feature dimensionality reduction, choose Cluster centre numberAfter cluster, cluster centre point conduct is taken The standard feature of the cluster completes the standardization to each picture feature list.So far, picture feature is expressed as (cluster by dimensionality reduction Num it) ties up.

The implementation process of K-means are as follows:

A. always collect from picture and arbitrarily select k a as initial cluster center in n feature of formation.

B. according to the mean value of each cluster feature (center position of Euclidean distance), each feature and these centers are calculated The distance of position；And individual features are divided again according to minimum range.

C. each mean value (central point) for changing cluster is recalculated

D. circulation B to C is until each cluster is no longer changed.

By embodiment of the disclosure, standardization is done to feature list, obtains the feature list of dimensionality reduction expression, so that There is similitude and comparability by the feature that different pictorial informations obtain, to be based further on standardized feature list, determine The vision of each picture is presented effect and feature list is made to provide reliable and effective data basis, improves the quantization of picture visual effect Confidence level.

Fig. 3 F is diagrammatically illustrated according to the embodiment of the present disclosure based on standardized feature list, determines the vision of each picture The flow chart of effect is presented.

As illustrated in Figure 3 F, aforementioned operation S323 (being based on standardized feature list, determine that effect is presented in the vision of each picture) It may include operation S361~S363.Wherein:

In operation S361, multiple standardized feature lists are based on, determine that each feature is at this in each standardized feature list Active degree in standardized feature list.

In operation S362, determine that each feature is universal in all standardized feature lists in each standardized feature list Degree.

In operation S363, it is based on active degree and popularity, determines that effect is presented in the vision of each picture.

In accordance with an embodiment of the present disclosure, after standardized feature list, standardized feature list can be quantified as weighing The score of spirogram piece uniqueness degree.Specifically, can be a text, description figure by the commodity picture information after standardization The feature of piece information is conceptualized as a lemma.It, should if some lemma (feature) generally comes across in each text (picture) The effect that lemma plays in terms of describing text characteristics it is inapparent；, whereas if an entry only occurs in current text, Then illustrate that the entry can preferably describe some characteristics of text.TF-IDF algorithm is to describe text spy for quantifying entry The method of property degree, to assess a certain words for the important of a copy of it file in a file set or a corpus Degree.But the disclosure is not limited to this, those skilled in the art can be using other any present inventive concepts of may be implemented Method, details are not described herein again.

For visual experience, unique commodity are easier to cause user behavior.If the TF-IDF of picture feature must divide it It is bigger, mean that current image more can be different from other pictures, corresponding goods are more intended to forward ranking.For current The feature list FeaSet of commodity picture, wherein the TF-IDF score of a certain feature feature X is expressed are as follows:

Wherein, S_frequenceIt indicates active degree of the feature feature X in feature list feature list Fea Set, is Measure the score of feature importance.The embodiment of the present disclosure takes:

S_popularIt indicates the degree that feature feature X is popularized in all texts, is the score for measuring feature universality. The embodiment of the present disclosure takes:

In addition, for balance characteristics quantity bring difference, the sum of above-mentioned picture feature TF-IDF score is only examined The feature of N before worry particular commodity picture TF-IDF score.(N default takes 10, or is set according to specific picture feature distributed number It sets).So far, quantization work of the picture visual signature to characterization picture uniqueness score is completed.

By embodiment of the disclosure, be based on standardized feature list, determine each feature active degree and universal journey Degree realizes that the conversion of effect is presented to picture for picture feature information, sorts for picture to determine that effect is presented in the vision of each picture Quantization basis is provided.

Fig. 3 G, which is diagrammatically illustrated, is presented effect based on operation data and the vision of each picture according to the embodiment of the present disclosure, Determine the flow chart of layout of each picture in object displayed page.

As shown in Figure 3 G, aforementioned operation S312 (is presented effect based on operation data and the vision of each picture, determines each picture Layout in object displayed page) it may include operation S371~S373.Wherein:

In operation S371, it is based on operation data, determines the first cloth being laid out in object displayed page to each picture Office's information.

In operation S372, effect is presented in view-based access control model, determines the be laid out in object displayed page to each picture Two layout informations and respective weights value.

Determine that each picture exists according to the first layout information and the second layout information and respective weights value in operation S373 Layout in object displayed page.

In accordance with an embodiment of the present disclosure, pictorial information is quantified as after vision presentation effect score, can be by picture score Sequencing feature is added, realizes on the basis of existing data characteristics, picture is added and obtains new feature of the fractional dimension as sequence.Specifically The priority of the weighted value of imparting, logic, needs to be adjusted according to the actual situation, herein without limitation.

The contribution that picture score feature sorts to commodity is mainly manifested in two aspects.

First aspect: picture score is higher, and commodity sequence is more tended to forward.

The second aspect: when the higher commodity continuous arrangement of picture similarity, sequence is tended to break up the sequence It operates (referring to the explanation of subsequent figure 3H).

For example, two commodity P1 and P2 are recalled in inquiry, according to original sequence logic determined based on data, the sequence of P1 S1, P2 score S2 is calculated in score；Picture visual signature is now added, P1, P2 score are respectively V1, V2 under this feature, and are enabled Picture visual signature weight is W_v, then new ranking score are as follows:

S1 '=S1+V1*W_v

S2 '=S2+V2*W_v

By S1 ' and S2 ' descending form final ranking results.

It is presented on the basis of determining each picture layout based on operation data in conjunction with vision by embodiment of the disclosure The layout of effect, to determine layout of each picture in object displayed page, so that the layout of picture considers picture to user's Visual experience fills up shortcoming of the object order in visual effect, so that the sequence effect of displaying of the object on the page is more Meet the direct feel of user, promote user experience, improves and click conversion ratio.

Fig. 3 H diagrammatically illustrates the flow chart of the data processing method of the another embodiment according to the embodiment of the present disclosure.

As shown in figure 3h, which may include operation S381~S382.Wherein:

In operation S381, determining that each picture after the layout in object displayed page, determines and connect in object displayed page Whether the similarity of the plurality of pictures of continuous arrangement meets preset condition.

In operation S382, in the case where satisfaction, to layout of the continuously arranged plurality of pictures in object displayed page It is adjusted.

As before, picture score feature is higher in picture similarity to the second aspect for the contribution performance that commodity sort When commodity continuous arrangement, sequence is tended to carry out breaing up operation to the sequence.It should be noted that it is based on each that sequence, which is broken up, Characteristic quantification score completes (minor sort) progress after sorting, and is a kind of optimization means of ranking results.For example go out in succession The commodity of existing more than 10 same businessmans, this is bad experience for a user.Picture breaks up purpose and is to optimize commodity picture The similar situation of appearance.

For example, a minor sort returns the sequence of commodity: P1, P2, P3 ..., compare the feature list (being denoted as A) of P1 picture With the similitude of the picture feature list (B) of P2, similitude is obtained by calculating the consin distance of vector:

Wherein, wherein all feature lists are formed by the dimension of feature space in the picture complete or collected works after n expression standardization Degree.(A, B are projected on the feature space of n dimension at this time, dimension having the same).

When similarity is higher than similarity threshold Ws, determine that the picture similarity degree of P1 and P2 is excessively high；As a sequence of P3, P4 Equal commodity are determined high with P1 picture similarity, and when quantity reaches similar amt threshold value Wn, are finally judged as similar quotient Product Pi is penalized.The measure of punishment is: finding a distance Pi recently and can terminate a quotient of current continuous similar situation Product Pj, replaces the position of Pi, and commodity Pi and thereafter commodity sequentially move back an exhibition position.

By embodiment of the disclosure, according to the similarity of plurality of pictures continuously arranged in object displayed page, to even Layout of the plurality of pictures of continuous arrangement in object displayed page is adjusted, and the optimization being laid out to picture may be implemented, so that Ranking results are more reasonable, and user experience is more preferable.

Fig. 4 diagrammatically illustrates the block diagram of the data processing system according to the embodiment of the present disclosure.

As shown in figure 4, the data processing equipment 400 may include that receiving module 410, first obtains module 420, first really Cover half block 430 and the second determining module 440.Wherein: receiving module 410 is for receiving object query request, wherein Object Query Request is for requesting inquiry target object.First, which obtains module 420, is used in response to object query request, acquisition and target object Matched multiple objects and for showing picture used in each object in multiple objects.First determining module 430 is used for The picture of acquisition is analyzed, effect is presented with the vision of each picture of determination.Second determining module 440 is used to be based on each picture Vision present effect, determine layout of each picture in object displayed page.

It is understood that receiving module 410, first obtains module 420, the first determining module 430 and second determines mould Block 440, which may be incorporated in a module, to be realized or any one module therein can be split into multiple modules.Or At least partly function of person, one or more modules in these modules can mutually be tied at least partly function of other modules It closes, and is realized in a module.It is determined according to an embodiment of the invention, receiving module 410, first obtains module 420, first At least one of module 430 and the second determining module 440 can at least be implemented partly as hardware circuit, such as scene can It programs gate array (FPGA), programmable logic array (PLA), system on chip, the system on substrate, the system in encapsulation, dedicated Integrated circuit (ASIC), or can be to carry out hardware or the firmwares such as any other rational method that is integrated or encapsulating to circuit It realizes, or is realized with software, the appropriately combined of hardware and firmware three kinds of implementations.Alternatively, receiving module 410, first Obtaining at least one of module 420, the first determining module 430 and second determining module 440 can at least be at least partially implemented The function of corresponding module can be executed when the program is run by computer for computer program module.

Fig. 5 A diagrammatically illustrates the block diagram of the data processing system according to another embodiment of the disclosure.

As shown in Figure 5A, which can also include the second acquisition module 450 and third determining module 460.Wherein: the second acquisition module 450 is used to obtain the operation data with each object in the matched multiple objects of target object. Third determining module 460 is used to that effect to be presented based on operation data and the vision of each picture, determines that each picture shows page in object Layout in face.

Fig. 5 B diagrammatically illustrates the block diagram of the first determining module according to the embodiment of the present disclosure.

As shown in Figure 5 B, the first determining module 430 may include that extraction unit 521, processing unit 522 and first are determining single Member 523.Wherein: extraction unit 521 is for extracting each picture corresponding feature list under different scale space.Processing unit 522 for each picture, corresponding feature list to be standardized under different scale space, to obtain standardized feature List.First determination unit 523 is used to be based on standardized feature list, determines that effect is presented in the vision of each picture.

Fig. 5 C diagrammatically illustrates the block diagram of the extraction unit according to the embodiment of the present disclosure.

As shown in Figure 5 C, extraction unit 521 may include that the first processing subelement 531, first determines subelement 532, the Two processing subelements 533 generate subelement 534 and third processing subelement 535.Wherein: first processing subelement 531 for pair Each picture carries out Gaussian Blur processing under different sizes, to obtain the gaussian pyramid figure of different scale corresponding with each picture Piece.First determines that subelement 532 is used to traverse each scale picture in gaussian pyramid picture corresponding with each picture, determines not The candidate point of corresponding characteristic point with scale, wherein candidate point is the pixel for belonging to extreme value in this scale and adjacent scale.The Two processing subelements 533 are used to be based on candidate point, carry out the screening and accurate positioning of characteristic point.It generates subelement 534 and is used for base In screening obtained characteristic point and its neighborhood, Feature Descriptor of each characteristic point under its correspondence scale is generated.Third processing The Feature Descriptor that subelement 535 is used to generate corresponding feature list under different scale as each picture.

By embodiment of the disclosure, the feature list of the corresponding each picture of different scale is extracted, can accurately be obtained The feature that effect is presented in each picture vision can be represented, ensures picture important information to the greatest extent and with discrimination information Retain, while providing reliable data basis for next step standardized feature list.

Fig. 5 D diagrammatically illustrates the block diagram of the processing unit according to the embodiment of the present disclosure.

As shown in Figure 5 D, processing unit 522 may include choosing the determining son of subelement 541, computation subunit 542, second Unit 543 updates subelement 544 and fourth process subelement 545.Wherein: choosing subelement 541 and be used for from each picture in difference At least one feature is chosen under scale in corresponding feature list as a feature class, wherein at least one characteristic storage exists In initialization feature list.Computation subunit 542 be used to calculate other features in feature list in addition at least one feature with The distance between at least one feature.Second determines that subelement 543 is used for the feature in the corresponding feature list of minimum range It is determined as corresponding feature normalization processing result.Subelement 544 is updated to be used for corresponding feature normalization processing result more Newly into initialization feature list.Fourth process subelement 545 is used for all spies in updated initialization feature list Levy the feature class new as one.

Fig. 5 E diagrammatically illustrates the block diagram of the first determination unit according to the embodiment of the present disclosure.

As shown in fig. 5e, the first determination unit 523 may include that third determines that subelement the 551, the 4th determines subelement 552 Subelement 553 is determined with the 5th.Wherein: third determines subelement 551 for determining each based on multiple standardized feature lists Active degree of each feature in this standard feature list in standardized feature list.4th determines subelement 552 for true Popularity of each feature in all standardized feature lists in fixed each standardized feature list.5th determines subelement 553, for being based on active degree and popularity, determine that effect is presented in the vision of each picture.

Fig. 5 F diagrammatically illustrates the block diagram of the second determining module according to the embodiment of the present disclosure.

As illustrated in figure 5f, the second determining module 440 may include the second determination unit 561, third determination unit 562 and Four determination units 563.Wherein: the second determination unit 561 is used to be based on operation data, determines to each picture in object displayed page In the first layout information for being laid out.Effect is presented for view-based access control model in third determination unit 562, determines to each picture right As the second layout information and respective weights value being laid out in displayed page.4th determination unit 563 is used for according to the first cloth Office's information and the second layout information and respective weights value, determine layout of each picture in object displayed page.

Fig. 5 G diagrammatically illustrates the block diagram of the data processing system of the another embodiment according to the embodiment of the present disclosure.

As depicted in fig. 5g, which can also include third determining module 470 and adjustment module 480.Wherein: Third determining module 470 is used to determine that each picture after the layout in object displayed page, determines and connect in object displayed page Whether the similarity of the plurality of pictures of continuous arrangement meets preset condition.It adjusts module 480 to be used in the case where satisfaction, to continuous Layout of the plurality of pictures of arrangement in object displayed page is adjusted.

Fig. 6 diagrammatically illustrates the computer for being adapted for carrying out data processing method and its system according to the embodiment of the present disclosure The block diagram of system.Computer system shown in Fig. 6 is only an example, function to the embodiment of the present disclosure and should not use model Shroud carrys out any restrictions.

As shown in fig. 6, include processor 601 according to the computer system 600 of the embodiment of the present disclosure, it can be according to storage It is loaded into random access storage device (RAM) 603 in the program in read-only memory (ROM) 602 or from storage section 608 Program and execute various movements appropriate and processing.Processor 601 for example may include general purpose microprocessor (such as CPU), refer to Enable set processor and/or related chip group and/or special microprocessor (for example, specific integrated circuit (ASIC)), etc..Processing Device 601 can also include the onboard storage device for caching purposes.Processor 601 may include for executing with reference to Fig. 2, Fig. 3 A The single treatment unit of the different movements of the method flow according to the embodiment of the present disclosure of~Fig. 3 C and Fig. 3 E~Fig. 3 H description Either multiple processing units.

In RAM 603, it is stored with system 600 and operates required various programs and data.Processor 601, ROM 602 with And RAM603 is connected with each other by bus 604.Processor 601 is executed by executing the program in ROM602 and/or RAM603 Above with reference to the various operations of Fig. 2, Fig. 3 A~Fig. 3 C and Fig. 3 E~Fig. 3 H data processing method described.It is noted that journey Sequence also can store in one or more memories in addition to ROM602 and RAM603.Processor 601 can also be by holding The program of row storage in one or more memories is retouched to execute above with reference to Fig. 2, Fig. 3 A~Fig. 3 C and Fig. 3 E~Fig. 3 H The various operations for the data processing method stated.

In accordance with an embodiment of the present disclosure, system 600 can also include input/output (I/O) interface 605, input/output (I/O) interface 605 is also connected to bus 604.System 600 can also include be connected to I/O interface 605 with one in lower component Item is multinomial: the importation 606 including keyboard, mouse etc.；Including such as cathode-ray tube (CRT), liquid crystal display (LCD) Deng and loudspeaker etc. output par, c 607；Storage section 608 including hard disk etc.；And including such as LAN card, modulatedemodulate Adjust the communications portion 609 of the network interface card of device etc..Communications portion 609 executes communication process via the network of such as internet. Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as disk, CD, magneto-optic disk, semiconductor Memory etc. is mounted on as needed on driver 610, in order to be pacified as needed from the computer program read thereon It is packed into storage section 608.

In accordance with an embodiment of the present disclosure, it may be implemented as computer software journey above with reference to the method for flow chart description Sequence.For example, embodiment of the disclosure includes a kind of computer program product comprising carry meter on a computer-readable medium Calculation machine program, the computer program include the program code for method shown in execution flow chart.In such embodiments, The computer program can be downloaded and installed from network by communications portion 609, and/or be pacified from detachable media 611 Dress.When the computer program is executed by processor 601, the above-mentioned function of limiting in the system of the embodiment of the present disclosure is executed.Root According to embodiment of the disclosure, system as described above, unit, module, unit etc. can by computer program module come It realizes.

It should be noted that computer-readable medium shown in the disclosure can be computer-readable signal media or meter Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device, Or above-mentioned any appropriate combination.In the disclosure, computer readable storage medium can be it is any include or storage journey The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this In open, computer-readable signal media may include in a base band or as the data-signal that carrier wave a part is propagated, Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned Any appropriate combination.In accordance with an embodiment of the present disclosure, computer-readable medium may include above-described ROM 602 And/or one or more memories other than RAM 603 and/or ROM 602 and RAM 603.

Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction It closes to realize.

As on the other hand, the disclosure additionally provides a kind of computer-readable medium, which can be Included in equipment described in above-described embodiment；It is also possible to individualism, and without in the supplying equipment.Above-mentioned calculating Machine readable medium carries one or more program, when said one or multiple programs are executed by the equipment, makes Obtain the equipment configuration for executing data processing, comprising: receive object query request, wherein object query request is for requesting inquiry Target object；In response to object query request, obtain with the matched multiple objects of target object and for showing multiple objects In picture used in each object；The picture of acquisition is analyzed, effect is presented with the vision of each picture of determination；And base Effect is presented in the vision of each picture, determines layout of each picture in object displayed page.

Embodiment of the disclosure is described above.But the purpose that these embodiments are merely to illustrate that, and It is not intended to limit the scope of the present disclosure.Although respectively describing each embodiment above, but it is not intended that each reality Use cannot be advantageously combined by applying the measure in example.The scope of the present disclosure is defined by the appended claims and the equivalents thereof.It does not take off From the scope of the present disclosure, those skilled in the art can make a variety of alternatives and modifications, these alternatives and modifications should all fall in this Within scope of disclosure.

Claims

1. a kind of data processing method, comprising:

Receive object query request, wherein the object query request is for requesting inquiry target object；

In response to the object query request, obtain and matched multiple objects of the target object and described more for showing Picture used in each object in a object；

The picture of acquisition is analyzed, effect is presented with the vision of each picture of determination；And

Effect is presented in vision based on each picture, determines layout of each picture in object displayed page.

2. according to the method described in claim 1, wherein, the method also includes:

Obtain the operation data with each object in the matched multiple objects of the target object；And

Effect is presented based on the operation data and the vision of each picture, determines that each picture shows page in the object Layout in face.

3. method according to claim 1 or 2, wherein analyze the picture of acquisition, with the vision of each picture of determination Effect, which is presented, includes:

Extract each picture corresponding feature list under different scale space；

To each picture, the corresponding feature list is standardized under different scale space, to be standardized Feature list；And

Based on the standardized feature list, determine that effect is presented in the vision of each picture.

4. according to the method described in claim 3, wherein, extracting each picture corresponding characteristic series under different scale space Table includes:

Gaussian Blur processing is carried out under different sizes to each picture, to obtain different scale corresponding with each picture Gaussian pyramid picture；

Each scale picture in gaussian pyramid picture corresponding with each picture is traversed, determines the corresponding feature of different scale The candidate point of point, wherein the candidate point is the pixel for belonging to extreme value in this scale and adjacent scale；

Based on the candidate point, the screening and accurate positioning of characteristic point are carried out, wherein the screening and accurate positioning at least include It is fitted the offset of extreme point and eliminates skirt response；

Obtained characteristic point and its neighborhood are screened based on described, feature of each characteristic point under its correspondence scale is generated and retouches State son；And

Using the Feature Descriptor of generation as the feature list of each picture.

5. according to the method described in claim 3, wherein, to each picture under different scale the corresponding feature list It is standardized, includes: to obtain standardized feature list

At least one feature is chosen in corresponding feature list under different scale from each picture as a feature class, In, at least one described characteristic storage is in initialization feature list；

It calculates between other features and at least one described feature in the feature list in addition at least one described feature Distance；

Feature in the corresponding feature list of minimum range is determined as corresponding feature normalization processing result；

The corresponding feature normalization processing result is updated in the initialization feature list；And

Using the feature class new as one of all features in updated initialization feature list.

6. according to the method described in claim 3, wherein, the standardized feature list include it is multiple, be based on the standardization Feature list determines that the vision of each picture is presented effect and includes:

Based on multiple standardized feature lists, determine that each feature is in this standardization spy in each standardized feature list Levy the active degree in list；

Determine popularity of each feature in all standardized feature lists in each standardized feature list；And

Based on the active degree and the popularity, determine that effect is presented in the vision of each picture.

7. according to the method described in claim 2, wherein, effect is presented based on the operation data and the vision of each picture Fruit determines that layout of each picture in the object displayed page includes:

Based on the operation data, determines and the first layout that each picture is laid out in the object displayed page is believed Breath；

Effect is presented based on the vision, determines the second cloth being laid out in the object displayed page to each picture Office's information and respective weights value；And

According to first layout information and the second layout information and the respective weights value, determine each picture described Layout in object displayed page.

8. according to the method described in claim 7, wherein, the method also includes:

Determining that each picture after the layout in the object displayed page, determines and continuously arrange in the object displayed page Whether the similarity of the plurality of pictures of column meets preset condition；And

In the case where satisfaction, layout of the continuously arranged plurality of pictures in the object displayed page is adjusted It is whole.

9. a kind of data processing system, comprising:

Receiving module, for receiving object query request, wherein the object query request is for requesting inquiry target object；

First obtains module, for obtaining and the matched multiple objects of the target object in response to the object query request And for showing picture used in each object in the multiple object；

Effect is presented for analyzing the picture of acquisition with the vision of each picture of determination in first determining module；And

Effect is presented for the vision based on each picture in second determining module, determines that each picture shows page in object Layout in face.

10. system according to claim 9, wherein the system also includes:

Second obtains module, for obtaining the operation data with each object in the matched multiple objects of the target object；With And

Third determining module determines each figure for effect to be presented based on the operation data and the vision of each picture Layout of the piece in the object displayed page.

11. system according to claim 9 or 10, wherein first determining module includes:

Extraction unit, for extracting each picture corresponding feature list under different scale space；

Processing unit, for the corresponding feature list to be standardized place under different scale space to each picture Reason, to obtain standardized feature list；And

First determination unit determines that effect is presented in the vision of each picture for being based on the standardized feature list.

12. system according to claim 11, wherein the extraction unit includes:

First processing subelement, for carrying out Gaussian Blur processing under different sizes to each picture, with acquisition with it is described The gaussian pyramid picture of the corresponding different scale of each picture；

First determines subelement, for traversing each scale picture in gaussian pyramid picture corresponding with each picture, really Determine the candidate point of the corresponding characteristic point of different scale, wherein the candidate point is to belong to extreme value in this scale and adjacent scale Pixel；

Second processing subelement carries out the screening and accurate positioning of characteristic point, wherein the sieve for being based on the candidate point Choosing and accurate positioning at least include the offset and elimination skirt response of fitting extreme point；

Subelement is generated, for screening obtained characteristic point and its neighborhood based on described, it is right at its to generate each characteristic point Answer the Feature Descriptor under scale；And

Third handles subelement, the corresponding feature under different scale as each picture of the Feature Descriptor for that will generate List.

13. system according to claim 11, wherein the processing unit includes:

Subelement is chosen, is made for choosing at least one feature in corresponding feature list under different scale from each picture For a feature class, wherein at least one described characteristic storage is in initialization feature list；

Computation subunit, for calculate other features in the feature list in addition at least one described feature and it is described at least The distance between one feature；

Second determines subelement, for the feature in the corresponding feature list of minimum range to be determined as corresponding feature mark Standardization processing result；

Subelement is updated, for the corresponding feature normalization processing result to be updated in the initialization feature list； And

Fourth process subelement, for the feature that all features in updated initialization feature list are new as one Class.

14. system according to claim 11, wherein first determination unit includes:

Third determines subelement, for being based on multiple standardized feature lists, determines each standardized feature list In active degree of each feature in this standard feature list；

4th determines subelement, for determining that each feature is in all standardized feature lists in each standardized feature list In popularity；And

5th determines subelement, for being based on the active degree and the popularity, determines that the vision of each picture is in Existing effect.

15. system according to claim 10, wherein second determining module includes:

Second determination unit, for be based on the operation data, determine to each picture in the object displayed page into First layout information of row layout；

Third determination unit is determined to each picture for effect to be presented based on the vision in the object displayed page In the second layout information for being laid out and respective weights value；And

4th determination unit is used for according to first layout information and the second layout information and the respective weights value, really Fixed layout of each picture in the object displayed page.

16. system according to claim 15, wherein the system also includes:

Third determining module, for determining each picture after the layout in the object displayed page, it is described right to determine As whether the similarity of plurality of pictures continuously arranged in displayed page meets preset condition；And

Module is adjusted, is used in the case where satisfaction, to the continuously arranged plurality of pictures in the object displayed page Layout be adjusted.

17. a kind of computer system, comprising:

One or more processors；And

Storage device, for storing one or more programs,

Wherein, when one or more programs are executed by one or more processors, so that one or more processors realize power Benefit requires any one of 1 to 8 data processing method.

18. a kind of computer-readable medium, is stored thereon with executable instruction, which makes processor real when being executed by processor The data processing method of existing any one of claims 1 to 8.