CN110209854A - Picture determines method and device - Google Patents
Picture determines method and device Download PDFInfo
- Publication number
- CN110209854A CN110209854A CN201910373208.0A CN201910373208A CN110209854A CN 110209854 A CN110209854 A CN 110209854A CN 201910373208 A CN201910373208 A CN 201910373208A CN 110209854 A CN110209854 A CN 110209854A
- Authority
- CN
- China
- Prior art keywords
- picture
- candidate
- end article
- default
- recommendation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
- G06F16/538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0641—Shopping interfaces
- G06Q30/0643—Graphical representation of items or shoppers
Abstract
The disclosure is directed to a kind of pictures to determine method and device.The described method includes: obtaining multiple candidate pictures, the first picture in multiple described candidate pictures is default picture, and candidate's picture is for showing end article；It is determined according to the quality information of the degree of correlation and each Zhang Suoshu candidate picture between candidate's picture and the end article described in every and recommends picture；Show the selection interface of the recommendation picture and the default picture；The recommendation picture or the default picture are determined as the master map of the end article by the selection instruction received according to the selection interface.The disclosure can promote merchandise display effect, to promote the clicking rate of commodity.
Description
Technical field
This disclosure relates to which field of computer technology more particularly to a kind of picture determine method and device.
Background technique
In e-commerce system, be better understood by merchandise news in order to facilitate buyer user, seller user often on
It passes plurality of pictures and is used as description of commodity, current commonplace way is the first picture that directly uploads seller user as quotient
The master map of product is to show in items list.But first picture be often not necessarily most can be noticeable or most convincing
Picture, therefore may result in buyer user using first picture as master map and can not select at the first time the commodity admired,
To which good bandwagon effect can not be obtained, the problem that thus commodity clicking rate will be caused not high.
Summary of the invention
To overcome the problems in correlation technique, the embodiment of the present disclosure provides a kind of picture and determines method and device.Institute
It is as follows to state technical solution:
According to the first aspect of the embodiments of the present disclosure, a kind of picture is provided and determines method, comprising:
Multiple candidate pictures are obtained, the first picture in multiple described candidate pictures is default picture, candidate's picture
For showing end article；
According between candidate's picture and the end article described in every the degree of correlation and each Zhang Suoshu candidate picture
Quality information, which determines, recommends picture；
Show the selection interface of the recommendation picture and the default picture；
The recommendation picture or the default picture are determined as by the selection instruction received according to the selection interface
The master map of the end article.
In one embodiment, the degree of correlation according between candidate's picture and the end article described in every and
The quality information of each Zhang Suoshu candidate picture determines that recommendation picture includes:
Extract the semantic information of each Zhang Suoshu candidate picture, institute's semantic information include the candidate picture classification and
Target object；
Every time is determined according to the summary info of the semantic information of candidate's picture and the end article described in every
Select the degree of correlation between picture and the end article；
According to the degree of correlation and the determining candidate pictures of preset threshold between the candidate picture and the end article
It closes；
It is highest described that quality information based on the candidate picture chooses mass fraction in the candidate picture set
Candidate picture is as recommendation picture.
In one embodiment, the degree of correlation according between the candidate picture and the end article and default
Threshold value determines that candidate picture set includes:
Judge whether the degree of correlation between presently described candidate picture and the end article is more than the preset threshold；
When the degree of correlation is more than the preset threshold, presently described candidate picture is divided into the candidate pictures
It closes.
In one embodiment, the quality information based on the candidate picture is chosen in the candidate picture set
The highest candidate picture of mass fraction includes: as recommending picture
Calculate or extract the mass fraction of every candidate picture in the candidate picture set；
The highest candidate picture of the mass fraction is chosen as the recommendation picture.
In one embodiment, the method also includes:
Judge whether the candidate picture set is empty set；
When the candidate pictures are combined into empty set, the default picture is determined as to the master map of the end article.
In one embodiment, the method also includes:
Judge whether the recommendation picture is the default picture；
When the recommendation picture is the default picture, the default picture is determined as to the master of the end article
Figure.
According to the second aspect of an embodiment of the present disclosure, a kind of picture determining device is provided, comprising:
Module is obtained, for obtaining multiple candidate pictures, the first picture in multiple described candidate pictures is default picture,
Candidate's picture is for showing end article；
Recommending module, for according between candidate's picture and the end article described in every the degree of correlation and each institute
The quality information for stating candidate picture, which determines, recommends picture；
Display module, for showing the selection interface of the recommendation picture and the default picture；
First determining module, selection instruction for being received according to the selection interface is by the recommendation picture or institute
State the master map that default picture is determined as the end article.
In one embodiment, the recommending module includes:
Extracting sub-module, for extracting the semantic information of each Zhang Suoshu candidate picture, institute's semantic information includes the time
Select the classification and target object of picture；
First determines submodule, for the abstract according to the semantic information of candidate's picture and the end article described in every
Information determines the degree of correlation between every candidate picture and the end article；
Second determines submodule, for according to the degree of correlation between the candidate picture and the end article and default
Threshold value determines candidate picture set；
Recommend submodule, chooses quality in the candidate picture set for the quality information based on the candidate picture
The highest candidate picture of score is as recommendation picture.
In one embodiment, described second determine that submodule includes:
Judging unit, for judging whether the degree of correlation between presently described candidate picture and the end article is more than institute
State preset threshold；
Division unit, for when the degree of correlation is more than the preset threshold, presently described candidate picture to be divided into institute
State candidate picture set.
In one embodiment, the recommendation submodule includes:
Computing unit, for calculating or extracting the quality point of every candidate picture in the candidate picture set
Number；
Selection unit, for choosing the highest candidate picture of the mass fraction as the recommendation picture.
In one embodiment, described device further include:
First determination module, for judging whether the candidate picture set is empty set；
Second determining module, for the default picture being determined as described when the candidate pictures are combined into empty set
The master map of end article.
In one embodiment, described device further include:
Second determination module, for judging whether the recommendation picture is the default picture；
Third determining module, for when the recommendation picture is the default picture, the default picture to be determined
For the master map of the end article.
According to the third aspect of an embodiment of the present disclosure, a kind of picture determining device device is provided, comprising:
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to the step of executing first aspect any embodiment the method.
According to a fourth aspect of embodiments of the present disclosure, a kind of computer readable storage medium is provided, calculating is stored thereon with
Machine instruction, when which is executed by processor the step of realization first aspect any embodiment the method.
The technical scheme provided by this disclosed embodiment can include the following benefits:
The embodiment of the present disclosure can be determined as the first picture in multiple candidate pictures to default picture, and according to candidate picture
The quality information of the degree of correlation and candidate picture between end article is determined to recommend picture, silent so as to provide for seller
Recognize picture and recommend two kinds of picture selections, and determine the master map of end article according to the final selection of seller, so can be
Seller recommends more particularly suitable picture as master map, therefore helps to promote merchandise display effect, and then promote the click of commodity
Rate.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
The disclosure can be limited.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.
Fig. 1 is the flow chart one that the picture shown accoding to exemplary embodiment determines method；
Fig. 2 is the flowchart 2 that the picture shown accoding to exemplary embodiment determines method；
Fig. 3 a is the schematic diagram one of the master map selection interface shown accoding to exemplary embodiment；
Fig. 3 b is the schematic diagram two of the master map selection interface shown accoding to exemplary embodiment；
Fig. 4 is the flow chart 3 that the picture shown accoding to exemplary embodiment determines method；
Fig. 5 a is the module map one of the picture determining device shown accoding to exemplary embodiment；
Fig. 5 b is the module map two of the picture determining device shown accoding to exemplary embodiment；
Fig. 5 c is the module map three of the picture determining device shown accoding to exemplary embodiment；
Fig. 5 d is the module map four of the picture determining device shown accoding to exemplary embodiment；
Fig. 5 e is the module map five of the picture determining device shown accoding to exemplary embodiment；
Fig. 5 f is the module map six of the picture determining device shown accoding to exemplary embodiment；
Fig. 6 is the block diagram for picture determining device shown accoding to exemplary embodiment.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all implementations consistent with this disclosure.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the disclosure.
Technical solution provided by the embodiment of the present disclosure is related to a kind of picture and determines method, is applied to server, purpose
It is to determine that most suitable master map shows buyer from multiple candidate pictures that seller uploads.In the related technology, more general
Time way be directly using seller upload first picture as the master map of commodity to show in items list, but head picture
Often be not necessarily most can noticeable or most convictive picture, therefore may result in and buy as master map using first picture
Family can not select at the first time the commodity admired, so that good bandwagon effect can not be obtained, thereby result in commodity clicking rate
Not high problem.Based on this, the embodiment of the present disclosure can be determined as the first picture in multiple candidate pictures to default picture, and root
Determine to recommend picture according to the quality information of the degree of correlation and candidate picture between candidate picture and end article, so as to for
Seller provides default picture and recommends two kinds of picture selections, and the master map of end article is determined according to the final selection of seller,
So more particularly suitable picture can be recommended as master map for seller, therefore help to promote merchandise display effect, and then promoted
The clicking rate of commodity.
Fig. 1 illustrates the flow chart that picture provided by the embodiment of the present disclosure determines method.According to Fig. 1,
The picture determines that method specifically comprises the following steps S100 to S400:
In the step s 100, multiple candidate pictures are obtained, the first picture in multiple candidate pictures is default picture, institute
Candidate picture is stated for showing end article；
In step s 200, according to the degree of correlation and each candidate picture between every candidate picture and end article
Quality information is determined to recommend picture；
In step S300, the selection interface recommended picture and default picture is shown；
In step S400, the selection instruction received according to selection interface will recommend picture or default picture to be determined as
The master map of end article.
Exemplary, when showing end article, seller can sequentially upload multiple candidate pictures about end article, service
Device, can be using first candidate picture as default picture after receiving these candidate pictures, which refers to unselected
It will be confirmed as the picture of master map when more suitable picture out, while can also obtain every time according to the associated description of end article
The degree of association between picture and end article is selected, and obtains the quality information of each candidate picture, thus according to candidate picture
The quality information of the degree of association and candidate picture between end article is determined to recommend picture.Based on this, server may be used also
A selection interface is provided, which shows about the option for recommending picture and default picture, in order to which seller selects to need
Master map of the picture wanted as end article.When seller chooses the picture as master map, server can receive correspondence
Selection instruction, to be determined as the master map of end article in response to the selection instruction will recommend picture or default picture.
Based on above-mentioned steps S100-S400 it is found that disclosed technique scheme can be by the first picture in multiple candidate pictures
It is determined as defaulting picture, while true according to the quality information of the degree of correlation and candidate picture between candidate picture and end article
Recommendation picture is made, to provide default picture for seller and recommend two kinds of picture selections, and is determined according to the selection of seller
The master map of end article so can recommend more particularly suitable picture as master map, therefore help to promote commodity exhibition for seller
Show effect, and then promotes the clicking rate of commodity.
Each step of disclosed technique scheme is described in detail combined with specific embodiments below.
In the step s 100, multiple candidate pictures are obtained, the first picture in multiple candidate pictures is default picture, institute
Candidate picture is stated for showing end article.
Exemplary, seller can sequentially upload multiple candidate pictures about end article when showing end article, and every
Candidate picture can show end article from different perspectives.Server, can be by first Zhang Houxuan after receiving these candidate pictures
Picture is as default picture, in order to which the default picture is determined as master map when not selecting more suitable picture.
In step s 200, according to the degree of correlation and each candidate picture between every candidate picture and end article
Quality information is determined to recommend picture.
Exemplary, server also needs the degree of correlation according to candidate picture and end article after determining default picture
And the quality information of candidate picture is determined to recommend picture, in order to provide more particularly suitable picture as master map for seller
Alternative picture.Wherein, Fig. 2 illustrates the determination method flow diagram for recommending picture in the embodiment of the present disclosure.It can with reference to Fig. 2
Know, is pushed away according to the quality information determination of the degree of correlation and each candidate picture between candidate's picture and end article described in every
Recommending picture specifically comprises the following steps S201 to S204:
In step s 201, the semantic information of each candidate picture is extracted, which includes the classification of candidate picture
And target object；
Every time is determined in step S202, according to the semantic information of every candidate picture and the summary info of end article
Select the degree of correlation between picture and end article；
In step S203, according between candidate picture and end article the degree of correlation and preset threshold determine candidate figure
Piece set；
It is highest that quality information in step S204, based on candidate picture chooses mass fraction in candidate picture set
Candidate picture is as recommendation picture.
Exemplary, server can extract respectively its semantic information based on each candidate picture, and semantic information here refers to
Being able to reflect picture classification and its various information comprising object, specific extracting method includes but is not limited to be based on deep learning
Picture classification sorting technique.By taking a candidate picture is to wear the female model of one-piece dress and leather boots as an example, server is based on should
Candidate picture can at least obtain the semantic information about female model, one-piece dress and leather boots.At this point, server can also be further
Obtain seller provide the summary info about end article, which includes the classification and others of end article
Associated description information, for example, end article classification be women's shoes, associated description information include spring and autumn money, in, hollow out, pointed shoes
With ankle boots etc..Based on this, server can determine time according to the summary info of the semantic information of candidate picture and end article
The degree of correlation between picture and end article is selected, the determination principle of the degree of correlation can highlight end article based on candidate picture
Depending on degree.
After having determined the degree of correlation between every candidate picture and end article, server can be transferred and be preset
Preset threshold, and the degree of correlation between each candidate picture and end article is compared with the preset threshold, so as to general
The degree of correlation is more than that the candidate picture of preset threshold is divided into candidate picture set.Specifically, server more candidate picture with
When the size relation of the degree of correlation and preset threshold between end article, it can determine whether between current candidate picture and end article
The degree of correlation whether be more than preset threshold, if so, current candidate picture is divided into candidate picture set, under judging again later
One candidate picture；If it is not, then continuing to judge next candidate picture, until traversing all candidate pictures.It should be understood that
: candidate's picture set is formed by meeting the preset condition i.e. degree of correlation more than or equal to the candidate picture of preset threshold
, i.e., the picture number in candidate picture set needs according to the degree of correlation and depending on the size relation of preset threshold, therefore also
Empty situation is combined into there may be candidate pictures.
Based on this, after having determined candidate picture set, the embodiment of the present disclosure also needs further to judge candidate's picture
Whether set is empty set.When candidate's pictures are combined into empty set, the candidate picture that not can satisfy preset condition is indicated, by
It is to be selected from candidate picture set, therefore it is not select suitably that candidate pictures, which are combined into empty result, in recommendation picture
Recommend picture, default picture is directly determined as to the master map of end article at this time.In candidate's picture set non-empty, clothes
Business device can also obtain the quality information of each candidate picture in candidate's picture set, which can pass through mass fraction
To reflect.Wherein, server can calculate the mass fraction of candidate picture when receiving the candidate picture of seller's upload, at this time
Directly extract the mass fraction calculated；Alternatively, server can also be counted again after determining candidate picture set
The mass fraction for calculating candidate picture, only needs the mass fraction for calculating the candidate picture in candidate's picture set, therefore energy at this time
Enough effective saving computing resources.In getting candidate picture set after the quality information of each candidate picture, server
The recommendation picture that the highest candidate picture of mass fraction is provided as system can be chosen, which has with end article
The relatively high degree of correlation, and also there is best picture quality, therefore be more suitable for the master map of end article.
In step S300, the selection interface recommended picture and default picture is shown.
Exemplary, server can show that a selection interface, the selection interface are aobvious after determining to recommend picture to seller
It is shown with about the option for recommending picture and default picture, in order to which seller selects to recommend picture as needed or defaults picture
Master map as end article.In one embodiment, as shown in Figure 3a, server can will recommend picture and default picture to show
Show in the region of left and right two at same interface, and the corresponding position below picture shows the option for choosing, at this time seller
Two pictures can be compared, intuitively to make final selection.In another embodiment, as shown in Figure 3b, server
Also directly picture will can be recommended show in interface, while the inquiry request of offer one " whether by the picture be set as commodity master map "
And the option of "Yes" and "No", seller directly clicks "Yes" or "No" as needed at this time can be completed selection.
In step 400, the selection instruction received according to selection interface will recommend picture or default picture to be determined as
The master map of end article.
Exemplary, seller can automatically generate corresponding selection instruction and send after selection interface completes selection operation
To server, server after receiving the selection instruction, may be in response to the selection instruction with by corresponding recommendation picture or
Person defaults the master map that picture is determined as end article.Certainly, after determining the master map of end article, server can also be popped up
One prompt information, the prompt information can show in the form of a popup window, for showing finally determining master map to seller.In addition, this
The option of replacement master map can also be provided in embodiment, and when seller clicks the option, server can provide about default picture, recommend
The option information of picture and other pictures, in order to which seller changes the master map of end article.
Based on above-mentioned steps S100 to S400, it is contemplated that recommend picture and default picture to be likely to be identical picture, be
Such situation bring non-productive work is avoided, after determining to recommend picture, as shown in connection with fig. 4, the embodiment of the present disclosure is also
Can further comprise following steps S500 to S600:
In step S500, judge to recommend whether picture is default picture；
If so, thening follow the steps S600: default picture is determined as to the master map of end article；
If it is not, thening follow the steps S300: showing the selection interface recommended picture and default picture.
The embodiment of the present disclosure no longer needs to execute and recommends picture and default figure after determining that recommending picture is to default picture
The selection course of piece, but default picture is directly determined as to the master map of end article, it so can avoid subsequent nothing
Work is imitated, so that promoting picture determines efficiency.
Following is embodiment of the present disclosure, can be used for executing embodiments of the present disclosure.
Fig. 5 a is the structural schematic diagram of the picture determining device shown accoding to exemplary embodiment, which can be by soft
Part, hardware or both are implemented in combination with as some or all of of electronic equipment.According to Fig. 5 a, which determines dress
It sets including obtaining module 501, recommending module 502, display module 503 and the first determining module 504.Wherein, module 501 is obtained to use
In obtaining multiple candidate pictures, the first picture in multiple described candidate pictures is default picture, and candidate's picture is for opening up
Show end article；Recommending module 502 be used for according between candidate's picture and the end article described in every the degree of correlation and
The quality information of each Zhang Suoshu candidate picture, which determines, recommends picture；Display module 503 is for showing the recommendation picture and described
Default the selection interface of picture；The selection instruction that first determining module 504 is used to be received according to the selection interface will be described
Picture or the default picture is recommended to be determined as the master map of the end article.
In one embodiment, with reference to shown in Fig. 5 b, the recommending module 502 includes that extracting sub-module 5021, first is true
Stator modules 5022, second determine submodule 5023 and recommend submodule 5024.Wherein, extracting sub-module 5021 is each for extracting
The semantic information of Zhang Suoshu candidate's picture, institute's semantic information include the classification and target object of the candidate picture；First
Determine submodule 5022 for being determined according to the summary info of the semantic information of candidate's picture and the end article described in every
The degree of correlation between every candidate picture and the end article；Second determines that submodule 5023 is used for according to the candidate
The degree of correlation and preset threshold between picture and the end article determine candidate picture set；Submodule 5024 is recommended to be used for
Quality information based on the candidate picture chooses the highest candidate picture of mass fraction in the candidate picture set
As recommendation picture.
In one embodiment, with reference to shown in Fig. 5 c, described second determines that submodule 5023 includes determining whether 50231 He of unit
Division unit 50232.Wherein, judging unit 50231 is for judging between presently described candidate picture and the end article
Whether the degree of correlation is more than the preset threshold；Division unit 50232 is used for when the degree of correlation is more than the preset threshold, will
Presently described candidate's picture is divided into the candidate picture set.
In one embodiment, with reference to shown in Fig. 5 d, the recommendation submodule 5024 includes computing unit 50241 and chooses
Unit 50242.Wherein, computing unit 50241 is used to calculate or extract every candidate figure in the candidate picture set
The mass fraction of piece；Selection unit 50242 is for choosing the highest candidate picture of the mass fraction as the recommendation
Picture.
In one embodiment, with reference to shown in Fig. 5 e, the picture determining device further includes the first determination module 505 and
Two determining modules 506.Wherein, whether the first determination module 505 is empty set for judging the candidate picture set；Second determines
Module 506 is used to that the default picture to be determined as to the master map of the end article when the candidate pictures are combined into empty set.
In one embodiment, with reference to shown in Fig. 5 f, the picture determining device further includes the second determination module 507 and
Three determining modules 508.Wherein, the second determination module 507 is for judging whether the recommendation picture is the default picture；Third
Determining module 508 is used for when the recommendation picture is the default picture, and the default picture is determined as the target
The master map of commodity.
First picture in multiple candidate pictures can be determined as writing from memory by picture determining device provided by the embodiment of the present disclosure
Recognize picture, and recommendation figure is determined according to the quality information of the degree of correlation and candidate picture between candidate picture and end article
Piece so as to provide default picture for seller and recommend two kinds of picture selections, and determines target according to the final selection of seller
The master map of commodity so can recommend more particularly suitable picture as master map for seller, therefore help to promote merchandise display effect
Fruit, and then promote the clicking rate of commodity.
About the device in above-described embodiment, modules execute the concrete mode of operation in related this method
It is described in detail in embodiment, no detailed explanation will be given here.
The embodiment of the present disclosure also provides a kind of picture determining device, which includes:
Processor；
Memory for storage processor executable instruction；
Wherein, processor is configured as executing:
Multiple candidate pictures are obtained, the first picture in multiple described candidate pictures is default picture, candidate's picture
For showing end article；
According between candidate's picture and the end article described in every the degree of correlation and each Zhang Suoshu candidate picture
Quality information, which determines, recommends picture；
Show the selection interface of the recommendation picture and the default picture；
The recommendation picture or the default picture are determined as by the selection instruction received according to the selection interface
The master map of the end article.
In one embodiment, above-mentioned processor is also configured to: the semantic information of each Zhang Suoshu candidate picture is extracted,
Institute's semantic information includes the classification and target object of the candidate picture；
Every time is determined according to the summary info of the semantic information of candidate's picture and the end article described in every
Select the degree of correlation between picture and the end article；
According to the degree of correlation and the determining candidate pictures of preset threshold between the candidate picture and the end article
It closes；
It is highest described that quality information based on the candidate picture chooses mass fraction in the candidate picture set
Candidate picture is as recommendation picture.
In one embodiment, above-mentioned processor is also configured to: judging presently described candidate picture and the target
Whether the degree of correlation between commodity is more than the preset threshold；
When the degree of correlation is more than the preset threshold, presently described candidate picture is divided into the candidate pictures
It closes.
In one embodiment, above-mentioned processor is also configured to: being calculated or is extracted in the candidate picture set
The mass fraction of every candidate picture；
The highest candidate picture of the mass fraction is chosen as the recommendation picture.
In one embodiment, above-mentioned processor is also configured to: judging whether the candidate picture set is empty set；
When the candidate pictures are combined into empty set, the default picture is determined as to the master map of the end article.
In one embodiment, above-mentioned processor is also configured to: judging whether the recommendation picture is the default
Picture；
When the recommendation picture is the default picture, the default picture is determined as to the master of the end article
Figure.
Fig. 6 is the structural block diagram for picture determining device shown accoding to exemplary embodiment.For example, device 60 can quilt
It is provided as a server.Device 60 includes processing component 602, further comprises one or more processors, and by storing
Memory resource representated by device 604, can be by the instruction of the execution of processing component 602, such as application program for storing.Storage
The application program stored in device 604 may include it is one or more each correspond to one group of instruction module.In addition,
Processing component 602 is configured as executing instruction, to execute the above method.
Device 60 can also include the power management that a power supply module 606 is configured as executive device 60, and one wired
Or radio network interface 608 is configured as device 60 being connected to network and input/output (I/O) interface 610.Device
60 can operate based on the operating system for being stored in memory 604, such as Windows ServerTM, Mac OS XTM,
UnixTM, LinuxTM, FreeBSDTM or similar.
The embodiment of the present disclosure also provides a kind of non-transitorycomputer readable storage medium, when the instruction in storage medium by
When the processor of device 60 executes, so that device 60 is able to carry out above-mentioned picture and determines method, which comprises
Multiple candidate pictures are obtained, the first picture in multiple described candidate pictures is default picture, candidate's picture
For showing end article；
According between candidate's picture and the end article described in every the degree of correlation and each Zhang Suoshu candidate picture
Quality information, which determines, recommends picture；
Show the selection interface of the recommendation picture and the default picture；
The recommendation picture or the default picture are determined as by the selection instruction received according to the selection interface
The master map of the end article.
In one embodiment, the degree of correlation according between candidate's picture and the end article described in every and
The quality information of each Zhang Suoshu candidate picture determines that recommendation picture includes:
Extract the semantic information of each Zhang Suoshu candidate picture, institute's semantic information include the candidate picture classification and
Target object；
Every time is determined according to the summary info of the semantic information of candidate's picture and the end article described in every
Select the degree of correlation between picture and the end article；
According to the degree of correlation and the determining candidate pictures of preset threshold between the candidate picture and the end article
It closes；
It is highest described that quality information based on the candidate picture chooses mass fraction in the candidate picture set
Candidate picture is as recommendation picture.
In one embodiment, the degree of correlation according between the candidate picture and the end article and default
Threshold value determines that candidate picture set includes:
Judge whether the degree of correlation between presently described candidate picture and the end article is more than the preset threshold；
When the degree of correlation is more than the preset threshold, presently described candidate picture is divided into the candidate pictures
It closes.
In one embodiment, the quality information based on the candidate picture is chosen in the candidate picture set
The highest candidate picture of mass fraction includes: as recommending picture
Calculate or extract the mass fraction of every candidate picture in the candidate picture set；
The highest candidate picture of the mass fraction is chosen as the recommendation picture.
In one embodiment, the method also includes:
Judge whether the candidate picture set is empty set；
When the candidate pictures are combined into empty set, the default picture is determined as to the master map of the end article.
In one embodiment, the method also includes:
Judge whether the recommendation picture is the default picture；
When the recommendation picture is the default picture, the default picture is determined as to the master of the end article
Figure.
Those skilled in the art will readily occur to its of the disclosure after considering specification and practicing disclosure disclosed herein
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Person's adaptive change follows the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by appended
Claim is pointed out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure should be limited by the accompanying claims.
Claims (14)
1. a kind of picture determines method characterized by comprising
Multiple candidate pictures are obtained, the first picture in multiple described candidate pictures is default picture, and candidate's picture is used for
Show end article；
According to the quality of the degree of correlation and each Zhang Suoshu candidate picture between candidate's picture and the end article described in every
Information, which determines, recommends picture；
Show the selection interface of the recommendation picture and the default picture；
The recommendation picture or the default picture are determined as described by the selection instruction received according to the selection interface
The master map of end article.
2. the method according to claim 1, wherein described according to candidate's picture described in every and the target quotient
The quality information of the degree of correlation and each Zhang Suoshu candidate picture between product determines that recommendation picture includes:
The semantic information of each Zhang Suoshu candidate picture is extracted, institute's semantic information includes the classification and target of the candidate picture
Object；
Every candidate figure is determined according to the summary info of the semantic information of candidate's picture and the end article described in every
The degree of correlation between piece and the end article；
According to the degree of correlation and the determining candidate picture set of preset threshold between the candidate picture and the end article；
Quality information based on the candidate picture chooses the highest candidate of mass fraction in the candidate picture set
Picture is as recommendation picture.
3. according to the method described in claim 2, it is characterized in that, it is described according to the candidate picture and the end article it
Between the degree of correlation and preset threshold determine that candidate picture set includes:
Judge whether the degree of correlation between presently described candidate picture and the end article is more than the preset threshold；
When the degree of correlation is more than the preset threshold, presently described candidate picture is divided into the candidate picture set.
4. according to the method described in claim 2, it is characterized in that, the quality information based on the candidate picture is described
The highest candidate picture of mass fraction is chosen in candidate picture set as recommendation picture includes:
Calculate or extract the mass fraction of every candidate picture in the candidate picture set；
The highest candidate picture of the mass fraction is chosen as the recommendation picture.
5. according to the method described in claim 2, it is characterized by further comprising:
Judge whether the candidate picture set is empty set；
When the candidate pictures are combined into empty set, the default picture is determined as to the master map of the end article.
6. method according to claim 1-5, which is characterized in that further include:
Judge whether the recommendation picture is the default picture；
When the recommendation picture is the default picture, the default picture is determined as to the master map of the end article.
7. a kind of picture determining device characterized by comprising
Module is obtained, for obtaining multiple candidate pictures, the first picture in multiple described candidate pictures is default picture, described
Candidate picture is for showing end article；
Recommending module, for according to the degree of correlation and each Zhang Suoshu time between candidate's picture and the end article described in every
It selects the quality information of picture to determine and recommends picture；
Display module, for showing the selection interface of the recommendation picture and the default picture；
First determining module, selection instruction for being received according to the selection interface is by the recommendation picture or described silent
Recognize the master map that picture is determined as the end article.
8. device according to claim 7, which is characterized in that the recommending module includes:
Extracting sub-module, for extracting the semantic information of each Zhang Suoshu candidate picture, institute's semantic information includes the candidate figure
The classification and target object of piece；
First determines submodule, for the summary info according to the semantic information of candidate's picture and the end article described in every
Determine the degree of correlation between every candidate picture and the end article；
Second determines submodule, for according to the degree of correlation and preset threshold between the candidate picture and the end article
Determine candidate's picture set；
Recommend submodule, chooses mass fraction in the candidate picture set for the quality information based on the candidate picture
The highest candidate picture is as recommendation picture.
9. device according to claim 8, which is characterized in that described second determines that submodule includes:
Judging unit, for judging whether the degree of correlation between presently described candidate picture and the end article is more than described pre-
If threshold value；
Division unit, for when the degree of correlation is more than the preset threshold, presently described candidate picture to be divided into the time
Select picture set.
10. device according to claim 8, which is characterized in that the recommendation submodule includes:
Computing unit, for calculating or extracting the mass fraction of every candidate picture in the candidate picture set；
Selection unit, for choosing the highest candidate picture of the mass fraction as the recommendation picture.
11. device according to claim 8, which is characterized in that further include:
First determination module, for judging whether the candidate picture set is empty set；
Second determining module, for when the candidate pictures are combined into empty set, the default picture to be determined as the target
The master map of commodity.
12. according to the described in any item devices of claim 7-11, which is characterized in that further include:
Second determination module, for judging whether the recommendation picture is the default picture；
Third determining module, for when the recommendation picture is the default picture, the default picture to be determined as institute
State the master map of end article.
13. a kind of picture determining device characterized by comprising
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to the step of perform claim requires any one of 1-6 the method.
14. a kind of computer readable storage medium, is stored thereon with computer instruction, which is characterized in that the instruction is by processor
The step of any one of claim 1-6 the method is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910373208.0A CN110209854B (en) | 2019-05-06 | 2019-05-06 | Picture determination method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910373208.0A CN110209854B (en) | 2019-05-06 | 2019-05-06 | Picture determination method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110209854A true CN110209854A (en) | 2019-09-06 |
| CN110209854B CN110209854B (en) | 2021-08-31 |
Family
ID=67785521
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910373208.0A Active CN110209854B (en) | 2019-05-06 | 2019-05-06 | Picture determination method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110209854B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112866387A (en) * | 2021-01-20 | 2021-05-28 | 金蝶蝶金云计算有限公司 | Picture uploading method and device and computer storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103092861A (en) * | 2011-11-02 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Method and system for selecting commodity representative picture |
| KR20140023079A (en) * | 2012-08-16 | 2014-02-26 | 삼성전자주식회사 | Method for displaying for image and an electronic device thereof |
| CN104484090A (en) * | 2014-12-12 | 2015-04-01 | 深圳市财富之舟科技有限公司 | Method for displaying pictures |
| CN105045818A (en) * | 2015-06-26 | 2015-11-11 | 腾讯科技（深圳）有限公司 | Picture recommending method, apparatus and system |
| US20160328785A1 (en) * | 2015-05-07 | 2016-11-10 | Alibaba Group Holding Limited | Device and method for targeted display of information |
| CN106547767A (en) * | 2015-09-18 | 2017-03-29 | 北京国双科技有限公司 | Determine the method and device of video front cover picture |
| CN106547808A (en) * | 2015-09-23 | 2017-03-29 | 阿里巴巴集团控股有限公司 | Picture update method, classification sort method and device |
| WO2017166748A1 (en) * | 2016-04-01 | 2017-10-05 | 乐视控股（北京）有限公司 | Calendar background setting method and apparatus, and electronic device |
| CN107239473A (en) * | 2016-03-29 | 2017-10-10 | 阿里巴巴集团控股有限公司 | A kind of exhibiting pictures system of selection and device |
| CN107330750A (en) * | 2017-05-26 | 2017-11-07 | 北京三快在线科技有限公司 | A kind of recommended products figure method and device, electronic equipment |
| CN108268488A (en) * | 2016-12-30 | 2018-07-10 | 百度在线网络技术（北京）有限公司 | The recognition methods of webpage master map and device |
| CN108345700A (en) * | 2018-03-29 | 2018-07-31 | 百度在线网络技术（北京）有限公司 | Article represents the choosing method, device and computer equipment of picture |
-
2019
- 2019-05-06 CN CN201910373208.0A patent/CN110209854B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103092861A (en) * | 2011-11-02 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Method and system for selecting commodity representative picture |
| KR20140023079A (en) * | 2012-08-16 | 2014-02-26 | 삼성전자주식회사 | Method for displaying for image and an electronic device thereof |
| CN104484090A (en) * | 2014-12-12 | 2015-04-01 | 深圳市财富之舟科技有限公司 | Method for displaying pictures |
| US20160328785A1 (en) * | 2015-05-07 | 2016-11-10 | Alibaba Group Holding Limited | Device and method for targeted display of information |
| CN105045818A (en) * | 2015-06-26 | 2015-11-11 | 腾讯科技（深圳）有限公司 | Picture recommending method, apparatus and system |
| CN106547767A (en) * | 2015-09-18 | 2017-03-29 | 北京国双科技有限公司 | Determine the method and device of video front cover picture |
| CN106547808A (en) * | 2015-09-23 | 2017-03-29 | 阿里巴巴集团控股有限公司 | Picture update method, classification sort method and device |
| CN107239473A (en) * | 2016-03-29 | 2017-10-10 | 阿里巴巴集团控股有限公司 | A kind of exhibiting pictures system of selection and device |
| WO2017166748A1 (en) * | 2016-04-01 | 2017-10-05 | 乐视控股（北京）有限公司 | Calendar background setting method and apparatus, and electronic device |
| CN108268488A (en) * | 2016-12-30 | 2018-07-10 | 百度在线网络技术（北京）有限公司 | The recognition methods of webpage master map and device |
| CN107330750A (en) * | 2017-05-26 | 2017-11-07 | 北京三快在线科技有限公司 | A kind of recommended products figure method and device, electronic equipment |
| CN108345700A (en) * | 2018-03-29 | 2018-07-31 | 百度在线网络技术（北京）有限公司 | Article represents the choosing method, device and computer equipment of picture |
Non-Patent Citations (1)
| Title |
|---|
| 林在韵: "人体体型因素及界面展示方式对线上服装购买成功率的影响", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112866387A (en) * | 2021-01-20 | 2021-05-28 | 金蝶蝶金云计算有限公司 | Picture uploading method and device and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110209854B (en) | 2021-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10049466B2 (en) | Color name generation from images and color palettes | |
| US20190073710A1 (en) | Color based social networking recommendations | |
| KR101972285B1 (en) | Image evaluation | |
| US8583685B2 (en) | Determination of category information using multiple stages | |
| US10262068B2 (en) | System, method, and non-transitory computer-readable storage media for displaying an optimal arrangement of facets and facet values for a search query on a webpage | |
| KR102125119B1 (en) | Data handling method and device | |
| US10162868B1 (en) | Data mining system for assessing pairwise item similarity | |
| EP3564870A1 (en) | Causality for machine learning systems | |
| US20210390365A1 (en) | Aspect Pre-selection using Machine Learning | |
| CN106570714B (en) | Recommendation method for matching object picture, and mapping relation establishment method and device | |
| JP2018509684A (en) | Information disclosure method and apparatus | |
| JP6989485B2 (en) | Multi-label data learning support device, multi-label data learning support method and multi-label data learning support program | |
| CN110992124A (en) | House resource recommendation method and system | |
| CN110209854A (en) | Picture determines method and device | |
| US20210357553A1 (en) | Apparatus and method for option data object performance prediction and modeling | |
| JP5945206B2 (en) | Product recommendation device, method and program | |
| CN112905937A (en) | Service content updating and generating method based on big data and cloud computing service system | |
| EP3216266B1 (en) | Generating network resource | |
| CN114730435A (en) | Management server and product search method | |
| CN110717095A (en) | Service item pushing method and device | |
| KR102495011B1 (en) | Method for recommending product using regression analysis according to product code and purchase pattern and apparatus for the same | |
| CN114596113A (en) | Commodity push strategy generation method and device | |
| US20190213613A1 (en) | Segmenting market data | |
| CN110196796B (en) | Effect evaluation method and device for recommendation algorithm | |
| JP2018101245A (en) | Selection device, selection method, and selection program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |