CN106445995A - Picture classification method and apparatus - Google Patents
Picture classification method and apparatus Download PDFInfo
- Publication number
- CN106445995A CN106445995A CN201610565611.XA CN201610565611A CN106445995A CN 106445995 A CN106445995 A CN 106445995A CN 201610565611 A CN201610565611 A CN 201610565611A CN 106445995 A CN106445995 A CN 106445995A
- Authority
- CN
- China
- Prior art keywords
- picture
- index information
- sorted
- target
- similarity value
- 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.)
- Pending
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
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a picture classification method and apparatus. The method comprises the steps of obtaining to-be-classified pictures of a terminal device; extracting feature information of a target object in the to-be-classified pictures, wherein the feature information is used for representing the type of the target object; searching for target index information matched with the feature information, wherein the target index information is used for representing a category which the target object belongs to; and when the target index information matched with the feature information of the target object is found, creating associative relationships between the to-be-classified pictures and the found target index information. According to the method and the apparatus, the technical problem of relatively low picture classification efficiency in the prior art is solved.
Description
Technical field
The present invention relates to data processing field, in particular to a kind of sorting technique of picture and device.
Background technology
In existing society, with the extensive application of intelligent terminal, user can store in intelligent terminal
Substantial amounts of picture or substantial amounts of picture is uploaded in high in the clouds or Dropbox by intelligent terminal is preserved, wherein,
The picture downloaded including photo and by internet.In order to user searches conveniently, in intelligent terminal or in high in the clouds
Picture can be classified and be managed, the processing method in prior art, picture classified and managed is to be built by user
Found corresponding file to realize that is to say, that user needs the classification according to picture to create corresponding file, then, will
Picture stores to the file creating.
For example, user by intelligent terminal, substantial amounts of picture is uploaded in high in the clouds or Dropbox and is preserved,
It is necessary first to consult the photo needing to upload before user's upper transmitting file in high in the clouds and Dropbox, sum up a collection of photo and comprise
Element, for example, place, the element such as time and scenery;Then, at least one file, example are set up according to the element extracting
As the title of the file of foundation can be " shooting in Beijing ", " being shot in 2016 " or " graduating photograph in 2016 " etc.;Connect
Get off, user passes through to select a file at least one file, photo upload in this document folder, or full dose
Corresponding catalogue is moved to after upload.
When realizing the classification of photo using said method, it is that completely manual the carrying out of user is classified, and the behaviour of photo
Make flow process loaded down with trivial details.If photo to be sorted is excessive, correspondingly classification time cost can be increased;Meanwhile, in substantial amounts of file
In go search and position a photo be a thing being extremely difficult.
For above-mentioned problem, effective solution is not yet proposed at present.
Content of the invention
Embodiments provide a kind of sorting technique of picture and device, divided with least solving picture in prior art
The less efficient technical problem of class.
A kind of one side according to embodiments of the present invention, there is provided sorting technique of picture, including:Obtain terminal unit
Picture to be sorted;Extract the characteristic information of destination object in described picture to be sorted, wherein, described characteristic information is used for characterizing
The classification of described destination object;Search the target index information matching with described characteristic information, wherein, described target index letter
Cease for characterizing the classification belonging to described destination object;Finding the target matching with the characteristic information of described destination object
During index information, create the incidence relation of described picture to be sorted and the described target index information finding.
Another aspect according to embodiments of the present invention, additionally provides a kind of sorter of picture, including:First obtains list
Unit, for obtaining the picture to be sorted of terminal unit;Extraction unit, for extracting the spy of destination object in described picture to be sorted
Reference ceases, and wherein, described characteristic information is used for characterizing the classification of described destination object;First searching unit, for lookup and institute
State the target index information that characteristic information matches, wherein, described target index information is used for characterizing belonging to described destination object
Classification;First creating unit, for finding the target index information matching with the characteristic information of described destination object
When, create the incidence relation of described picture to be sorted and the described target index information finding.
In embodiments of the present invention, using the picture to be sorted obtaining terminal unit;Extract mesh in described picture to be sorted
The characteristic information of mark object, wherein, described characteristic information is used for characterizing the classification of described destination object;Search and described feature letter
The target index information of manner of breathing coupling, wherein, described target index information is used for characterizing the classification belonging to described destination object;?
When finding the target index information matching with the characteristic information of described destination object, create described picture to be sorted and lookup
The mode of the incidence relation of described target index information arriving, extracts the characteristic information of destination object first, then, searches and spy
Levy the index information of information match, finally, when finding index information, by create picture to be sorted and index information it
Between incidence relation, with respect to the mode in prior art, manually classified for picture, reached according to picture
The characteristic information of middle destination object is automatically obtained the purpose of picture classification, it is achieved thereby that improving the technology effect of picture classification efficiency
Really, and then solve the less efficient technical problem of picture classification in prior art.
Brief description
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, this
Bright schematic description and description is used for explaining the present invention, does not constitute inappropriate limitation of the present invention.In the accompanying drawings:
The Organization Chart of the hardware environment that the terminal server that Fig. 1 is according to embodiments of the present invention is constituted;
Fig. 2 is a kind of flow chart of the sorting technique of picture according to embodiments of the present invention;
Fig. 3 is a kind of multiple index informations alternatively showing establishment in terminal unit according to embodiments of the present invention
Schematic diagram with the display interface associating picture to be sorted of each index information;
Fig. 4 is according to embodiments of the present invention a kind of alternatively to show the first object picture inquiring in terminal unit
The schematic diagram of display interface;
Fig. 5 is a kind of schematic diagram of the categorizing system of picture according to embodiments of the present invention;
Fig. 6 is a kind of schematic diagram of the sorter of picture according to embodiments of the present invention;And
Fig. 7 is the structure chart of terminal according to embodiments of the present invention.
Specific embodiment
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, is clearly and completely described to the technical scheme in the embodiment of the present invention it is clear that described embodiment is only
The embodiment of a present invention part, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained under the premise of not making creative work, all should belong to the model of present invention protection
Enclose.
It should be noted that term " first " in description and claims of this specification and above-mentioned accompanying drawing, "
Two " it is etc. for distinguishing similar object, without for describing specific order or precedence.It should be appreciated that such use
Data can exchange in the appropriate case so that embodiments of the invention described herein can with except here diagram or
Order beyond those of description is implemented.Additionally, term " comprising " and " having " and their any deformation are it is intended that cover
Cover non-exclusive comprising, for example, contain series of steps or process, method, system, product or the equipment of unit are not necessarily limited to
Those steps clearly listed or unit, but may include clearly not listing or for these processes, method, product
Or the intrinsic other steps of equipment or unit.
Embodiment 1
According to embodiments of the present invention, there is provided a kind of embodiment of the method that can be executed by the application device embodiment,
It should be noted that the step that illustrates of flow process in accompanying drawing can be in the department of computer science of such as one group of computer executable instructions
Execute in system, and although showing logical order in flow charts, but in some cases, can be with different from herein
The shown or described step of order execution.
According to embodiments of the present invention, there is provided a kind of sorting technique of picture.
Alternatively, in the present embodiment, the sorting technique of above-mentioned picture can apply to terminal 102 He as shown in Figure 1
In the hardware environment that server 104 is constituted.As shown in figure 1, terminal 102 is attached with server 104 by network, above-mentioned
Network includes but is not limited to:Mobile communications network, wide area network, Metropolitan Area Network (MAN) or LAN, terminal 102 can be mobile phone terminal,
Can be PC terminal, notebook terminal or panel computer terminal.The sorting technique of the picture of the embodiment of the present invention can be by servicing
Device 102, to execute it is also possible to be executed by terminal 102, can also be and jointly executed by server 104 and terminal 102.Wherein,
The sorting technique that terminal 102 executes the picture of the embodiment of the present invention can also be executed by client mounted thereto.
The main operational principle of the hardware environment system shown in Fig. 1 is:
Terminal 102 is passed through network and is uploaded onto the server in 104 by the picture to be sorted being stored in advance in terminal 102.Clothes
Business device 104, after the picture to be sorted getting terminal 102, will extract the characteristic information of destination object in picture to be sorted.
For example, the objects such as trees, personage, tables and chairs, blue sky and white clouds are comprised in pictures, destination object is in above-mentioned object extremely
One of few, characteristic information is the classification of above-mentioned target characteristic.Server 104, after extracting features described above information, will be looked into
Look for the target index information matching with characteristic information, for example, search the index information for storing the photo comprising trees, or
Person searches for storing the index information comprising personage.Server 104 is finding the target index matching with characteristic information
After information, picture to be sorted and the incidence relation of target index information can be created.After creating above-mentioned incidence relation, use
Arbitrary pictures just can be searched according to the incidence relation of foundation in family.It is possible in foundation during user input " trees "
In index information search include " tree " or " trees " index information, then, by with the inclusion " tree " finding or " set
Wood " index information be associated picture push to terminal 102, terminal 102 receive server 104 push picture it
Afterwards, this picture can be shown in terminal 102, so that user checks.
Fig. 2 is the flow chart of the sorting technique of picture according to embodiments of the present invention, below in conjunction with Fig. 2, the present invention is implemented
The sorting technique of the picture that example is provided does concrete introduction, as shown in Fig. 2 the sorting technique of this picture mainly comprises the steps
S202 is to step S208:
Step S202, obtains the picture to be sorted of terminal unit.
In embodiments of the present invention, terminal unit can be the terminal units such as mobile phone, panel computer and computer;To be sorted
Picture uploaded onto the server by internet in terminal unit for user in picture, specifically, picture to be sorted is permissible
The picture downloaded by network in terminal unit for user, can also be the photo that user is shot by terminal unit.
Step S204, extracts the characteristic information of destination object in picture to be sorted, wherein, characteristic information is used for characterizing target
The classification of object.
In embodiments of the present invention, picture to be sorted includes multiple objects, for example, the object such as " sky " and " personage ",
Destination object is at least one of multiple objects object, and this destination object is used for determining the affiliated classification of picture to be sorted.
In above-mentioned steps S204, server after getting the picture to be sorted that user is uploaded by terminal unit,
The characteristic information of destination object can be extracted, then, execute following step S206, that is, search the target matching with characteristic information
Index information, and after finding target index information, the association creating between picture to be sorted and target index information is closed
System.
Step S206, searches the target index information matching with characteristic information, wherein, target index information is used for characterizing
Classification belonging to destination object.
In embodiments of the present invention, index information is the information for representing the classification belonging to object, wherein, index information
It is properly termed as tag along sort, in the classification interface of picture as shown in Figure 3, " label " is above-mentioned tag along sort, classification again
Multiple female mark label, such as " building " and " article " can be included, each female mark is signed including multiple subtabs, example in label
If the subtab in " building " can be " museum ", " Stadium " and " classroom " etc.;Subtab in " article " can be
" picture ", " cartoon " and " automobile " etc..
Step S208, when finding the target index information being matched with characteristic information, is created picture to be sorted and looks into
The incidence relation of the target index information finding.
In embodiments of the present invention, extract the characteristic information of destination object first, then, search and match with characteristic information
Index information, finally, when finding index information, closed by creating the association between picture to be sorted and index information
System, it is possible to realize automatically carrying out classification storage for picture, wishes at the same time it can also search user according to the incidence relation set up
The picture searched, with respect to the mode in prior art, manually classified for picture, has reached according in picture
The characteristic information of destination object is automatically obtained the purpose of picture classification, it is achieved thereby that improving the technology effect of picture classification efficiency
Really, and then solve the less efficient technical problem of picture classification in prior art.
In embodiments of the present invention, the picture to be sorted to step S208 for above-mentioned steps S202 is to be stored in advance in terminal
The photo that picture in equipment, the picture downloaded by network including user and user are shot by terminal unit.Treating point
When class picture is classified, picture to be sorted can be uploaded to server by network.For example, terminal unit is installed " xx
The client of Dropbox ", then, uploads picture to be sorted by the user end to server of " xx Dropbox ".In the embodiment of the present invention
In, picture to be sorted can be one, can also be multiple.
Server get terminal unit by the picture to be sorted of client upload after, execution step S204, that is,
Extract the characteristic information of destination object in picture to be sorted, wherein, this feature information is used for characterizing the classification of destination object.Due to
The object comprising in one picture to be sorted can have many kinds, therefore, in embodiments of the present invention, can be in picture to be sorted
Selected one or more destination objects in the whole objects comprising, and obtain the characteristic information of this one or more destination object,
And then, search the target index information with this feature information match.Therefore, in an optional embodiment of the present invention,
During the characteristic information of the destination object in extracting picture to be sorted, above-mentioned steps S204 can be specially:Extract first and treat point
The characteristic information of whole objects in class picture, then, calculates the characteristic information of each object and pre-set categories mark in whole objects
Sign the Similarity value of the pre-set categories tag library in storehouse, obtain at least one Similarity value, finally, by least one Similarity value
As the characteristic information of destination object, wherein, target Similarity value is more than the corresponding pre-set categories label of middle target Similarity value
Or it is equal to default Similarity value.
For example, for example, arbitrary picture to be sorted, the whole objects in this picture to be sorted include " lamp ", " table
Son ", " bookshelf ", " books ", " flower " and " computer ".In embodiments of the present invention, image processing method or figure can be passed through
As the characteristic information of above-mentioned whole objects is extracted in recognition methodss, i.e. can be identified in picture to be sorted by image processing method
All view data of object, and then the characteristic information according to the whole object of the image data extraction recognizing, wherein, picture number
According to data messages such as the shape of inclusion object, color and sizes.After obtaining the characteristic information of above-mentioned whole objects, calculate every
The Similarity value of the characteristic information of individual object and pre-set categories label, obtains multiple Similarity value, for example, by " lamp ", " desk ",
" bookshelf ", " books ", " flower " and " computer " calculates the Similarity value with pre-set categories label respectively;Then, multiple similar
The target Similarity value searched more than or equal to default Similarity value in angle value, and will be corresponding for target Similarity value default
Class label is as the characteristic information of destination object.
By above-described embodiment, the quantity being calculated target similarity in multiple similarities may be multiple, such as
Fruit target similarity excessive, certain burden can be brought to terminal unit server, be not unfavorable for picture to be sorted and
The foundation of the incidence relation between target index information.Therefore, it can select the phase of predetermined number in multiple target similarities
Like spending, preferably, 1 to 3 target Similarity value can be chosen, then, by corresponding for this 1 to 3 target similarity default class
Distinguishing label is as the characteristic information of destination object.Specifically, when determining the target Similarity value of predetermined number, can set first
Put a predetermined number, for example, n, wherein, n could be arranged to 1,2 or 3, then, judges that the quantity of multiple target similarities is
No more than or equal to predetermined number, if it is judged that the quantity of multiple target similarity be more than or equal to predetermined number, then
Above-mentioned multiple target Similarity value can be ranked up according to order from high to low, obtain a sequence, and by this sequence
In the corresponding pre-set categories label of front n target Similarity value as destination object characteristic information.
For example, calculating object " lamp ", " desk ", " bookshelf ", " books ", " flower " and " computer " respectively and presetting
After the Similarity value of class label, obtain N number of Similarity value, wherein, N is the positive integer more than 1.By judging, in N
In individual Similarity value, it is A more than or equal to the quantity of the target Similarity value of default Similarity value, wherein, A is more than 1
Positive integer.If predetermined number is chosen for 3, judge A>Whether 3 set up, if it is judged that A>3 establishments, then by above-mentioned A
Target Similarity value is ranked up according to order from high to low or from low to high.Wherein, if according to from high to low
Order be ranked up, then using corresponding for front 3 target Similarity value pre-set categories label as destination object characteristic information;As
Fruit is to be ranked up according to order from low to high, then using corresponding for rear 3 target Similarity value pre-set categories label as mesh
The characteristic information of mark object.
After above-mentioned steps S204 extract the characteristic information of destination object in picture to be sorted, by execution step S206,
Search the target index information matching with characteristic information, if finding the target index letter matching with characteristic information
Breath, then the incidence relation creating picture to be sorted with the target index information finding.For example, some picture to be sorted
The characteristic information of the whole objects including is " lamp ", " desk ", " bookshelf ", " books ", " flower " and " computer ", by calculating
The characteristic information of above-mentioned object is understood with the similarity of pre-set categories label, the mesh that " lamp ", " desk ", " bookshelf " as extract
The characteristic information of mark object.Then, the characteristic information of above-mentioned destination object is mated with the index information creating, searched
The target index information that matches with the characteristic information of destination object is that is to say, that search and be in the index information of establishment
The no index tab matching with " lamp ", " desk ", " bookshelf ".If finding target index tab, for example, find with
The target index tab that " lamp " matches is " light fixture ", then create the pass between picture to be sorted and target index information " light fixture "
Connection relation.Without finding the target index information matching with the characteristic information of destination object, for example, do not find
The target index information matching with " desk ", then create target index information " desk ", and creates picture to be sorted and " table
The incidence relation of son ".
It should be noted that the mode creating picture to be sorted with the incidence relation of target index information has many kinds,
In the embodiment of the present invention, the incidence relation between the unique identification information of picture to be sorted and target index information can be created,
Wherein, unique identification information is used for this picture to be sorted of unique mark, for example, the id information of picture to be sorted.Treat point creating
After incidence relation between the unique identification information of class picture and target index information, user just can be closed according to above-mentioned association
Correspondingly picture is searched by system.
Further, in the above embodiment of the present invention, there is the target index information of incidence relation with picture to be sorted
Can be one, can also be multiple, the quantity of objectives index information depends on the quantity of the destination object setting.If
The quantity of target index information is one, then only need to set up the unique identification information of picture to be sorted and this target index information
Incidence relation;If the quantity of target index information is multiple, need to set up the unique identification information of picture to be sorted with every
Incidence relation between individual target index information.
Alternatively, in embodiments of the present invention, create picture to be sorted and the target index letter finding in step S208
After the incidence relation of breath, server can also obtain the query statement of user's transmission, and then, server is looked into according to query statement
Look for the first index information matching with key word, then search, according to incidence relation, the picture that the first index information associates, and will
As first object picture, finally, server can push first object picture to terminal unit to the picture of association, wherein, looks into
Ask the key word carrying in instruction for searching first object picture in multiple index informations.
For example, user input inquiry can instruct, for example, in the visitor of " xx Dropbox " in the client of terminal unit
Input " timbered picture " in the end of family, and this query statement is sent to server by internet;Server is obtaining
To after this query statement, search the first index information matching with the key word " timbered picture " inputting.For example, look into
The first index information " trees " finding, then can search and the first index letter according to the incidence relation of establishment in step S208
The picture of manner of breathing association, and using the picture finding as first object picture;Server after finding the first picture, to
Terminal unit pushes first object picture, to show first object picture in terminal unit, wherein, the number of first object picture
Amount can be multiple, acceptable one.It is a kind of signal inquiring about first object picture in terminal unit as shown in Figure 4
Figure, figure 4, it is seen that user can instruct in search column input inquiry, that is, inquires about " timbered picture ", server exists
After " timbered picture " is found according to query statement, push, to terminal unit, the first object picture finding, from Fig. 4
In as can be seen that Fig. 4 in display picture be server push " timbered picture " (that is, first object picture).
Alternatively, in embodiments of the present invention, create picture to be sorted and the target index letter finding in step S208
After the incidence relation of breath, server can also obtain the modification instruction of user's transmission, and wherein, modification instruction is for instruction modification
There is the target index information of incidence relation with picture to be sorted;Then, pass will be had with picture to be sorted according to modification instruction
The target index information of connection relation is revised as the second index information or the second index information and target index information.
For example, after creating picture to be sorted and the incidence relation of target index information in step S208, due to
Server can cause to identify inaccurate problem in data processing, there is provided a kind of user can manual modification and figure to be sorted
Piece has the index information mechanism of incidence relation.For example, there is the target index letter of incidence relation with arbitrary picture to be sorted
Cease for " blue sky " and " white clouds ", but user is actual to wish to associate this picture to be sorted with index information " sports ground " foundation
System, then can create and index information " fortune while retaining the incidence relation with target index information " blue sky " and " white clouds "
The incidence relation of dynamic field ";Or, delete the incidence relation with target index information " blue sky " and " white clouds ", create and index letter
The incidence relation of breath " sports ground ", wherein, above-mentioned index information " sports ground " is above-mentioned second index information.
In embodiments of the present invention, user can be uploaded to cloud photograph album and the substantial amounts of picture of cloud storage product is carried out soon
Speed, the classification of automatization, and shown in order on this basis and flexibly search for.Thus greatly save user uploading
Classification before picture, the time of filing.And in follow-up lookup picture it is only necessary to carry out spirit according to the feature things in impression
Live and quickly search for.
Fig. 5 is the schematic diagram of the categorizing system of picture according to embodiments of the present invention, and the categorizing system of this picture includes shining
Piece uploads subsystem 501, tag recognition subsystem 502, tab indexes subsystem 503 and customized label subsystem 504, its
In:
Photo upload subsystem 501, is arranged in terminal unit, for uploading picture to be sorted to server;
Tag recognition subsystem 502, arranges in the server, for being completed by image recognition or image processing method
The extraction of the characteristic information of destination object in picture to be sorted, and the characteristic information extracting is pushed to tab indexes subsystem
503;
Tab indexes subsystem 503, arranges in the server, for searching the spy pushing with tag recognition subsystem 502
Levy the target index information of information match, and when finding target index information, create picture to be sorted and target index
Incidence relation between information;Further, tab indexes subsystem 503 can also carry out the functions such as picture searching;
Customized label subsystem 504, setting in the server, the custom instruction for receive user or request,
For example, the modification instruction of the index information that modification is associated with picture to be sorted;Customized label subsystem 504 is on getting
After stating modification instruction, indicated content in modification instruction can be sent to tab indexes subsystem 503, so that label rope
The operation of the index information that directions subsystem 503 execution modification is associated with picture to be sorted.
In the sorting technique of picture provided in an embodiment of the present invention, user can be helped in personal cloud storage and photograph album etc.
Application software is achieved above to be sorted out, arranged and is searched for according to photo eigen sexual behavior thing.In embodiments of the present invention, right first
Picture photo to be sorted during user uploads onto the server carries out substantial amounts of pattern match, extracts to obtain 1 to 3 of category images
Characteristic information, and the incidence relation between picture to be sorted and target index information is created according to the characteristic information extracting.With
When, in terminal unit, client can be shown it is also possible to make to classified picture by pulling index information list
The classified picture in high in the clouds is scanned for search with input key word.The embodiment of the present invention, also support User Defined and
Edlin and modification are entered to the target index information of classified picture, on the one hand, can for the only a few of machine recognition not
Accurately target index information label is modified;On the other hand, user other can classify to individual according to the hobby of oneself
Picture creates the incidence relation and special target index information between.
In embodiments of the present invention, user's most classification demand can be met by server automatic identification, add
Self-defined can meet users ' individualized requirement and revise incorrect server recognition result on a small quantity.
It should be noted that for aforesaid each method embodiment, in order to be briefly described, therefore it is all expressed as a series of
Combination of actions, but those skilled in the art should know, the present invention is not limited by described sequence of movement because
According to the present invention, some steps can be carried out using other orders or simultaneously.Secondly, those skilled in the art also should know
Know, embodiment described in this description belongs to preferred embodiment, involved action and the module not necessarily present invention
Necessary.
Through the above description of the embodiments, those skilled in the art can be understood that according to above-mentioned enforcement
The method of example can be realized by the mode of software plus necessary general hardware platform naturally it is also possible to pass through hardware, but a lot
In the case of the former is more preferably embodiment.Based on such understanding, technical scheme is substantially in other words to existing
What technology contributed partly can be embodied in the form of software product, and this computer software product is stored in a storage
In medium (as ROM/RAM, magnetic disc, CD), including some instructions with so that a station terminal equipment (can be mobile phone, calculate
Machine, server, or network equipment etc.) execution each embodiment of the present invention described in method.
Embodiment 2
According to embodiments of the present invention, additionally provide a kind of classification dress of picture of the sorting technique for implementing above-mentioned picture
Put, the sorter of this picture is mainly used in executing the sorting technique of the picture that embodiment of the present invention the above is provided, with
Under the sorter of picture that the embodiment of the present invention is provided do concrete introduction:
Fig. 6 is the schematic diagram of the sorter of picture according to embodiments of the present invention, as shown in fig. 6, the classification of this picture
Device mainly includes:
First acquisition unit 61, for obtaining the picture to be sorted of terminal unit;
In embodiments of the present invention, terminal unit can be the terminal units such as mobile phone, panel computer and computer;To be sorted
Picture uploaded onto the server by internet in terminal unit for user in picture, specifically, picture to be sorted is permissible
The picture downloaded by network in terminal unit for user, can also be the photo that user is shot by terminal unit.
Extraction unit 63, for extracting the characteristic information of destination object in picture to be sorted, wherein, characteristic information is used for table
Levy the classification of destination object;
In embodiments of the present invention, picture to be sorted includes multiple objects, for example, the object such as " sky " and " personage ",
Destination object is at least one of multiple objects object, and this destination object is used for determining the affiliated classification of picture to be sorted.
In said extracted unit 63, server get the picture to be sorted that user uploaded by terminal unit it
Afterwards, the characteristic information of destination object can be extracted, then, by the first searching unit 65, that is, search and match with characteristic information
Target index information, and after finding target index information, create between picture to be sorted and target index information
Incidence relation.
First searching unit 65, for searching the target index information matching with characteristic information, wherein, target index letter
Cease for characterizing the classification belonging to destination object;
First creating unit 67, for finding the target index information matching with the characteristic information of destination object
When, create the incidence relation of picture to be sorted and the target index information finding.
In embodiments of the present invention, extract the characteristic information of destination object first, then, search and match with characteristic information
Index information, finally, when finding index information, closed by creating the association between picture to be sorted and index information
System, it is possible to realize automatically carrying out classification storage for picture, wishes at the same time it can also search user according to the incidence relation set up
The picture searched, with respect to the mode in prior art, manually classified for picture, has reached according in picture
The characteristic information of destination object is automatically obtained the purpose of picture classification, it is achieved thereby that improving the technology effect of picture classification efficiency
Really, and then solve the less efficient technical problem of picture classification in prior art.
In embodiments of the present invention, above-mentioned picture to be sorted is the picture being stored in advance in terminal unit, including user
The photo that the picture downloaded by network and user are shot by terminal unit.When treating category images and being classified, permissible
Picture to be sorted is uploaded to server by network.For example, the client of " xx Dropbox " is installed in terminal unit, then, leads to
The user end to server crossing " xx Dropbox " uploads picture to be sorted.In embodiments of the present invention, picture to be sorted can be
One, can also be multiple.
Alternatively, extraction unit includes:Extraction module, for extracting the characteristic information of whole objects in picture to be sorted;
Computing module, for calculating the pre-set categories label in the characteristic information of each object and pre-set categories tag library in whole objects
Similarity value, obtain at least one Similarity value;Determining module, for by target Similarity value at least one Similarity value
As the characteristic information of destination object, wherein, target Similarity value is more than or equal to default phase to corresponding pre-set categories label
Like angle value.
For example, for example, arbitrary picture to be sorted, the whole objects in this picture to be sorted include " lamp ", " table
Son ", " bookshelf ", " books ", " flower " and " computer ".In embodiments of the present invention, can be identified simultaneously by image processing method
Extract the characteristic information of above-mentioned whole object, i.e. whole objects in picture to be sorted can be identified by image processing method
View data, and then the characteristic information according to the whole object of the image data extraction recognizing, wherein, view data includes object
Shape, the data message such as color and size.After obtaining the characteristic information of above-mentioned whole objects, calculate the spy of each object
Reference breath and the Similarity value of pre-set categories label, obtain multiple Similarity value, for example, by " lamp ", " desk ", " bookshelf ", " book
Nationality ", " flower " and " computer " calculate the Similarity value with pre-set categories label respectively;Then, looking in multiple Similarity value
Look for the target Similarity value more than or equal to default Similarity value, and corresponding for target Similarity value pre-set categories label is made
Characteristic information for destination object.
Alternatively, the quantity of target similarity is multiple, and determining module includes:Judging submodule, for judging multiple mesh
Whether the quantity of mark Similarity value is more than or equal to predetermined number;Sorting sub-module, for judging that multiple targets are similar
In the case that the quantity of angle value is more than or equal to predetermined number, multiple target Similarity value are entered according to order from high to low
Row sequence;Determination sub-module, for by sequence after multiple target Similarity value in front predetermined number target Similarity value
Corresponding pre-set categories label is as the characteristic information of destination object.
By above-described embodiment, the quantity being calculated target similarity in multiple similarities may be multiple, such as
Fruit target similarity excessive, certain burden can be brought to terminal unit server, be not unfavorable for picture to be sorted and
The foundation of the incidence relation between target index information.Therefore, it can select the phase of predetermined number in multiple target similarities
Like spending, preferably, 1 to 3 target Similarity value can be chosen, then, by corresponding for this 1 to 3 target similarity default class
Distinguishing label is as the characteristic information of destination object.Specifically, when determining the target Similarity value of predetermined number, can set first
Put a predetermined number, for example, n, wherein, n could be arranged to 1,2 or 3, then, judges that the quantity of multiple target similarities is
No more than or equal to predetermined number, if it is judged that the quantity of multiple target similarity be more than or equal to predetermined number, then
Above-mentioned multiple target Similarity value can be ranked up according to order from high to low, obtain a sequence, and by this sequence
In the corresponding pre-set categories label of front n target Similarity value as destination object characteristic information.
For example, calculating object " lamp ", " desk ", " bookshelf ", " books ", " flower " and " computer " respectively and presetting
After the Similarity value of class label, obtain N number of Similarity value, wherein, N is the positive integer more than 1.By judging, in N
In individual Similarity value, it is A more than or equal to the quantity of the target Similarity value of default Similarity value, wherein, A is more than 1
Positive integer.If predetermined number is chosen for 3, judge A>Whether 3 set up, if it is judged that A>3 establishments, then by above-mentioned A
Target Similarity value is ranked up according to order from high to low or from low to high.Wherein, if according to from high to low
Order be ranked up, then using corresponding for front 3 target Similarity value pre-set categories label as destination object characteristic information;As
Fruit is to be ranked up according to order from low to high, then using corresponding for rear 3 target Similarity value pre-set categories label as mesh
The characteristic information of mark object.
Alternatively, device also includes:Second creating unit, matches for the characteristic information in lookup and destination object
After target index information, when not finding the target index information matching with the characteristic information of destination object, create mesh
Mark index information, and create picture to be sorted and the incidence relation of target index information.
After the characteristic information of destination object in picture to be sorted is extracted by said extracted unit, will be by above-mentioned the
The target index information that one searching unit lookup is matched with characteristic information, if find the target matching with characteristic information
Index information, then the incidence relation creating picture to be sorted with the target index information finding.For example, some is treated point
The characteristic information of whole objects that class picture includes is " lamp ", " desk ", " bookshelf ", " books ", " flower " and " computer ", leads to
The similarity crossing the characteristic information and pre-set categories label calculating above-mentioned object understands, " lamp ", " desk ", " bookshelf " are and extract
The characteristic information of the destination object arriving.Then, the characteristic information of above-mentioned destination object is carried out with the index information creating
Join, search the target index information matching with the characteristic information of destination object that is to say, that in the index information that creates
Search whether the index tab matching with " lamp ", " desk ", " bookshelf ".If finding target index tab, for example, search
It is " light fixture " to the target index tab matching with " lamp ", then create between picture to be sorted and target index information " light fixture "
Incidence relation.Without finding the target index information matching with the characteristic information of destination object, for example, do not look into
Find the target index information matching with " desk ", then create target index information " desk ", and create picture to be sorted
The incidence relation of " desk ".
It should be noted that the mode creating picture to be sorted with the incidence relation of target index information has many kinds,
In the embodiment of the present invention, the incidence relation between the unique identification information of picture to be sorted and target index information can be created,
Wherein, unique identification information is used for this picture to be sorted of unique mark, for example, the id information of picture to be sorted.Treat point creating
After incidence relation between the unique identification information of class picture and target index information, user just can be closed according to above-mentioned association
Correspondingly picture is searched by system.
Further, in the above embodiment of the present invention, there is the target index information of incidence relation with picture to be sorted
Can be one, can also be multiple, the quantity of objectives index information depends on the quantity of the destination object setting.If
The quantity of target index information is one, then only need to set up the unique identification information of picture to be sorted and this target index information
Incidence relation;If the quantity of target index information is multiple, need to set up the unique identification information of picture to be sorted with every
Incidence relation between individual target index information.
Alternatively, device also includes:Second acquisition unit, in the target index creating picture to be sorted with find
After the incidence relation of information, obtain the query statement that user sends, wherein, carry in query statement in multiple index letters
The key word of first object picture is searched in breath;Second searching unit, is matched with key word for being searched according to query statement
The first index information;3rd searching unit, for searching the picture of the first index information association according to incidence relation, and will close
The picture of connection is as first object picture;Push unit, for pushing first object picture to terminal unit.
For example, user input inquiry can instruct, for example, in the visitor of " xx Dropbox " in the client of terminal unit
Input " timbered picture " in the end of family, and this query statement is sent to server by internet;Server is obtaining
To after this query statement, search the first index information matching with the key word " timbered picture " inputting.For example, look into
The first index information " trees " finding, then can be searched and the first rope by the incidence relation of establishment in the first creating unit
The picture of fuse manner of breathing association, and using the picture finding as first object picture;Server find the first picture it
Afterwards, first object picture, to show first object picture in terminal unit, wherein, first object figure are pushed to terminal unit
The quantity of piece can be multiple, acceptable one.It is a kind of inquiry first object figure in terminal unit as shown in Figure 4 above
The schematic diagram of piece, figure 4, it is seen that user can instruct in search column input inquiry, that is, inquires about " timbered picture ",
Server, after find " timbered picture " according to query statement, pushes, to terminal unit, the first object figure finding
Piece, figure 4, it is seen that the picture of display is " the timbered picture " of server push (that is, the first mesh in Fig. 4
Mark on a map piece).
Alternatively, device includes:3rd acquiring unit, in the target index letter creating picture to be sorted with find
After the incidence relation of breath, obtain the modification instruction that user sends, wherein, modification instruction is for instruction modification and picture to be sorted
There is the target index information of incidence relation;Modification unit, is associated for being had with picture to be sorted according to modification instruction
The target index information of system is revised as the second index information or the second index information and target index information.
For example, after creating the incidence relation that picture to be sorted is with target index information, because server is counting
According to can cause in processing procedure to identify inaccurate problem, there is provided a kind of user manual modification can be had with picture to be sorted and associate
The index information mechanism of relation.For example, the target index information with arbitrary picture to be sorted with incidence relation is " blue sky "
" white clouds ", but user is actual to wish for this picture to be sorted and index information " sports ground " to set up incidence relation, then permissible
While retaining the incidence relation with target index information " blue sky " and " white clouds ", create the pass with index information " sports ground "
Connection relation;Or, delete the incidence relation with target index information " blue sky " and " white clouds ", create and index information " sports ground "
Incidence relation, wherein, above-mentioned index information " sports ground " be above-mentioned second index information.
In embodiments of the present invention, user can be uploaded to cloud photograph album and the substantial amounts of picture of cloud storage product is carried out soon
Speed, the classification of automatization, and shown in order on this basis and flexibly search for.Thus greatly save user uploading
Classification before picture, the time of filing.And in follow-up lookup picture it is only necessary to carry out spirit according to the feature things in impression
Live and quickly search for.
Embodiment 3
According to embodiments of the present invention, additionally provide a kind of terminal of the sorting technique for implementing above-mentioned picture, as Fig. 7 institute
Show, this terminal mainly includes processor 701, photographic head 702, display 703, data-interface 704, memorizer 705 and network and connects
Mouth 706, wherein:
Photographic head 702 is mainly used in being taken pictures, and obtains picture to be sorted.
Data-interface 704 then mainly by way of data transfer by the picture transfer to be sorted photographing to processor
701.
Memorizer 705 is mainly used in the target index letter storing picture to be sorted and having incidence relation with picture to be sorted
Breath.
Network interface 706 is mainly used in carrying out network service with server, and the classification for picture provides data to support.
Display 703 is mainly used in showing first object picture, can be also used for showing the target index information to create
And the picture to be sorted being associated with each target index information.
Processor 701 is mainly used in executing following operation:
Obtain the picture to be sorted of terminal unit;Extract the characteristic information of destination object in picture to be sorted, wherein, feature
Information is used for characterizing the classification of destination object;Search the target index information matching with characteristic information, wherein, target index letter
Cease for characterizing the classification belonging to destination object;Finding the target index information matching with the characteristic information of destination object
When, create the incidence relation of picture to be sorted and the target index information finding.
Processor 701 is additionally operable to extract the characteristic information of whole objects in picture to be sorted;Calculate each in whole objects
The Similarity value of the pre-set categories label in the characteristic information of object and pre-set categories tag library, obtains at least one similarity
Value;Using target Similarity value corresponding pre-set categories label at least one Similarity value as the characteristic information of destination object,
Wherein, target Similarity value is more than or equal to default Similarity value.
Processor 701 is additionally operable to judge whether the quantity of multiple target Similarity value is more than or equal to predetermined number;As
Fruit judge multiple target Similarity value quantity be more than or equal to predetermined number, then by multiple target Similarity value according to by
High to Low order is ranked up;By front predetermined number target Similarity value pair in the multiple target Similarity value after sequence
The pre-set categories label answered is as the characteristic information of destination object.
If processor 701 is additionally operable to not find the target index information matching with the characteristic information of destination object,
Create target index information, and create picture to be sorted and the incidence relation of target index information.
Processor 701 is additionally operable to obtain the query statement that user sends, and wherein, carries in multiple ropes in query statement
The key word of first object picture is searched in fuse breath;Search the first index letter matching with key word according to query statement
Breath;Search the picture of the first index information association according to incidence relation, and using the picture of association as first object picture;To end
End equipment pushes first object picture.
Processor 701 is additionally operable to obtain the modification instruction that user sends, wherein, modification instruction for instruction modification with treat point
Class picture has the target index information of incidence relation;The target of incidence relation will be had according to modification instruction with picture to be sorted
Index information is revised as the second index information or the second index information and target index information.
Alternatively, the specific example in the present embodiment may be referred to showing described in above-described embodiment 1 and embodiment 2
Example, the present embodiment will not be described here.
Embodiment 4
Embodiments of the invention additionally provide a kind of storage medium.Alternatively, in the present embodiment, above-mentioned storage medium can
For storing the program code of the sorting technique of the picture of the embodiment of the present invention.
Alternatively, in the present embodiment, above-mentioned storage medium may be located at mobile communications network, wide area network, Metropolitan Area Network (MAN) or
At least one of multiple network equipments in the network of the LAN network equipment.
Alternatively, in the present embodiment, storage medium is arranged to store the program code for executing following steps:
S1, obtains the picture to be sorted of terminal unit;
S2, extracts the characteristic information of destination object in picture to be sorted, and wherein, characteristic information is used for characterizing destination object
Classification;
S3, searches the target index information matching with characteristic information, wherein, target index information is used for characterizing target pair
As affiliated classification;
S4, when finding the target index information matching with the characteristic information of destination object, creates picture to be sorted
Incidence relation with the target index information finding.
Alternatively, in the present embodiment, above-mentioned storage medium can include but is not limited to:USB flash disk, read only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), portable hard drive, magnetic disc or
CD etc. is various can be with the medium of store program codes.
Alternatively, the specific example in the present embodiment may be referred to showing described in above-described embodiment 1 and embodiment 2
Example, the present embodiment will not be described here.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
If the integrated unit in above-described embodiment is realized and as independent product using in the form of SFU software functional unit
When selling or using, can be stored in the storage medium that above computer can read.Based on such understanding, the skill of the present invention
The part that art scheme substantially contributes to prior art in other words or all or part of this technical scheme can be with soft
The form of part product embodies, and this computer software product is stored in storage medium, including some instructions with so that one
Platform or multiple stage computers equipment (can be personal computer, server or network equipment etc.) execution each embodiment institute of the present invention
State all or part of step of method.
In the above embodiment of the present invention, the description to each embodiment all emphasizes particularly on different fields, and does not have in certain embodiment
The part describing in detail, may refer to the associated description of other embodiment.
It should be understood that disclosed client in several embodiments provided herein, other sides can be passed through
Formula is realized.Wherein, device embodiment described above is only the schematically division of for example described unit, and only one
Kind of division of logic function, actual can have other dividing mode when realizing, for example multiple units or assembly can in conjunction with or
It is desirably integrated into another system, or some features can be ignored, or do not execute.Another, shown or discussed mutual it
Between coupling or direct-coupling or communication connection can be by some interfaces, the INDIRECT COUPLING of unit or module or communication link
Connect, can be electrical or other form.
The described unit illustrating as separating component can be or may not be physically separate, show as unit
The part showing can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On NE.The mesh to realize this embodiment scheme for some or all of unit therein can be selected according to the actual needs
's.
In addition, can be integrated in a processing unit in each functional unit in each embodiment of the present invention it is also possible to
It is that unit is individually physically present it is also possible to two or more units are integrated in a unit.Above-mentioned integrated list
Unit both can be to be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
The above is only the preferred embodiment of the present invention it is noted that ordinary skill people for the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (12)
1. a kind of sorting technique of picture is it is characterised in that include:
Obtain the picture to be sorted of terminal unit;
Extract the characteristic information of destination object in described picture to be sorted, wherein, described characteristic information is used for characterizing described target
The classification of object;
Search the target index information that matches with described characteristic information, wherein, described in described target index information is used for characterizing
Classification belonging to destination object;
When finding the target index information matching with the characteristic information of described destination object, create described picture to be sorted
Incidence relation with the described target index information finding.
2. method according to claim 1 is it is characterised in that extract the feature letter of destination object in described picture to be sorted
Breath includes:
Extract the characteristic information of whole objects in described picture to be sorted;
Calculate the phase of the pre-set categories label in the characteristic information of each object and pre-set categories tag library in described whole object
Like angle value, obtain at least one Similarity value;
Using target Similarity value corresponding pre-set categories label at least one Similarity value described as described destination object
Characteristic information, wherein, described target Similarity value is more than or equal to default Similarity value.
3. method according to claim 2 it is characterised in that described target similarity quantity be multiple, by described extremely
In a few Similarity value, target Similarity value corresponding pre-set categories label includes as the characteristic information of described destination object:
Judge whether the quantity of multiple described target Similarity value is more than or equal to predetermined number;
If it is judged that the quantity of the plurality of described target Similarity value is more than or equal to described predetermined number, then will be described
Multiple described target Similarity value are ranked up according to order from high to low;
Will be corresponding default for front predetermined number target Similarity value in the plurality of described target Similarity value after sequence
Class label is as the characteristic information of described destination object.
4. method according to claim 1 is it is characterised in that searching the target index matching with described characteristic information
After information, methods described also includes:
If not finding the target index information matching with the characteristic information of described destination object, create described target index
Information, and create described picture to be sorted and the incidence relation of described target index information.
5. method according to claim 1 is it is characterised in that creating described picture to be sorted and the described mesh finding
After the incidence relation of mark index information, methods described also includes:
Obtain the query statement that user sends, wherein, carry in described query statement for searching the in multiple index informations
The key word of one Target Photo;
The first index information matching with described key word is searched according to described query statement;
Search the picture of described first index information association according to described incidence relation, and using the picture of association as described first
Target Photo;
Push described first object picture to described terminal unit.
6. method according to claim 1 is it is characterised in that creating described picture to be sorted and the described mesh finding
After the incidence relation of mark index information, methods described includes:
Obtain the modification instruction that user sends, wherein, described modification is instructed to be had with picture to be sorted for instruction modification and associates
The described target index information of relation;
According to described modification instruction, the described target index information with described picture to be sorted with incidence relation is revised as the
Two index informations or described second index information and described target index information.
7. a kind of sorter of picture is it is characterised in that include:
First acquisition unit, for obtaining the picture to be sorted of terminal unit;
Extraction unit, for extracting the characteristic information of destination object in described picture to be sorted, wherein, described characteristic information is used for
Characterize the classification of described destination object;
First searching unit, for searching the target index information matching with described characteristic information, wherein, described target index
Information is used for characterizing the classification belonging to described destination object;
First creating unit, for when finding the target index information matching with the characteristic information of described destination object,
The incidence relation creating described picture to be sorted with the described target index information finding.
8. device according to claim 7 is it is characterised in that described extraction unit includes:
Extraction module, for extracting the characteristic information of whole objects in described picture to be sorted;
Computing module, default in the characteristic information of each object and pre-set categories tag library in described whole object for calculating
The Similarity value of class label, obtains at least one Similarity value;
Determining module, for using target Similarity value corresponding pre-set categories label at least one Similarity value described as institute
State the characteristic information of destination object, wherein, described target Similarity value is more than or equal to default Similarity value.
9. device according to claim 8 is it is characterised in that the quantity of described target similarity is multiple, described determination
Module includes:
Judging submodule, whether the quantity for judging multiple described target Similarity value is more than or equal to predetermined number;
Sorting sub-module, described default for being more than or equal in the quantity judging the plurality of described target Similarity value
In the case of quantity, the plurality of described target Similarity value is ranked up according to order from high to low;
Determination sub-module, similar for front predetermined number target in the plurality of described target Similarity value after sorting
Angle value corresponding pre-set categories label is as the characteristic information of described destination object.
10. device according to claim 7 is it is characterised in that described device also includes:
Second creating unit, for after searching the target index information matching with the characteristic information of described destination object,
When not finding the target index information matching with the characteristic information of described destination object, create described target index letter
Breath, and create described picture to be sorted and the incidence relation of described target index information.
11. devices according to claim 7 are it is characterised in that described device also includes:
Second acquisition unit, in the incidence relation creating described picture to be sorted and the described target index information finding
Afterwards, obtain the query statement that user sends, wherein, carry in described query statement for searching the in multiple index informations
The key word of one Target Photo;
Second searching unit, for searching, according to described query statement, the first index information matching with described key word;
3rd searching unit, for searching the picture of described first index information association according to described incidence relation, and will associate
Picture as described first object picture;
Push unit, for pushing described first object picture to described terminal unit.
12. devices according to claim 7 are it is characterised in that described device includes:
3rd acquiring unit, in the incidence relation creating described picture to be sorted and the described target index information finding
Afterwards, obtain the modification instruction that user sends, wherein, described modification is instructed to be had with picture to be sorted for instruction modification and associates
The described target index information of relation;
Modification unit, for having the described target index of incidence relation according to described modification instruction with described picture to be sorted
Information is revised as the second index information or described second index information and described target index information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610565611.XA CN106445995A (en) | 2016-07-18 | 2016-07-18 | Picture classification method and apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610565611.XA CN106445995A (en) | 2016-07-18 | 2016-07-18 | Picture classification method and apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106445995A true CN106445995A (en) | 2017-02-22 |
Family
ID=58184133
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610565611.XA Pending CN106445995A (en) | 2016-07-18 | 2016-07-18 | Picture classification method and apparatus |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106445995A (en) |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107590291A (en) * | 2017-10-30 | 2018-01-16 | 广东欧珀移动通信有限公司 | A kind of searching method of picture, terminal device and storage medium |
CN107657051A (en) * | 2017-10-16 | 2018-02-02 | 广东欧珀移动通信有限公司 | A kind of generation method of picture tag, terminal device and storage medium |
CN108021654A (en) * | 2017-12-01 | 2018-05-11 | 北京奇安信科技有限公司 | A kind of photograph album image processing method and device |
CN108052563A (en) * | 2017-12-05 | 2018-05-18 | 广东小天才科技有限公司 | A kind of picture searching method and mobile terminal |
CN108121816A (en) * | 2017-12-28 | 2018-06-05 | 广东欧珀移动通信有限公司 | Picture classification method, device, storage medium and electronic equipment |
CN108416003A (en) * | 2018-02-27 | 2018-08-17 | 百度在线网络技术(北京)有限公司 | A kind of picture classification method and device, terminal, storage medium |
CN108596280A (en) * | 2018-04-27 | 2018-09-28 | 佛山市日日圣科技有限公司 | Image identification method for literature search |
CN108628869A (en) * | 2017-03-16 | 2018-10-09 | 富士施乐实业发展(中国)有限公司 | A kind of method and apparatus that category division is carried out to electronic document |
CN109033237A (en) * | 2018-07-04 | 2018-12-18 | 奇酷互联网络科技(深圳)有限公司 | Image searching method, device, readable storage medium storing program for executing and terminal |
CN109635153A (en) * | 2018-11-22 | 2019-04-16 | 腾讯音乐娱乐科技(深圳)有限公司 | Migration path generation method, device and storage medium |
CN109766052A (en) * | 2018-11-27 | 2019-05-17 | 浙江口碑网络技术有限公司 | Vegetable picture upload method, device, computer equipment and readable storage medium storing program for executing |
CN110019913A (en) * | 2018-06-01 | 2019-07-16 | 平安好房(上海)电子商务有限公司 | Picture match method, user equipment, storage medium and device |
CN110019894A (en) * | 2017-07-21 | 2019-07-16 | 北京搜狗科技发展有限公司 | A kind of location lookup method and device |
CN110162645A (en) * | 2019-05-28 | 2019-08-23 | 广东三维家信息科技有限公司 | Image search method, device and electronic equipment based on index |
CN110703972A (en) * | 2019-09-29 | 2020-01-17 | 维沃移动通信有限公司 | File control method and electronic equipment |
CN111061900A (en) * | 2018-10-17 | 2020-04-24 | 丽宝大数据股份有限公司 | Searching method for personal wearing record |
CN111368111A (en) * | 2020-02-28 | 2020-07-03 | Oppo(重庆)智能科技有限公司 | Picture retrieval method and device for mobile equipment and mobile equipment |
CN111434118A (en) * | 2017-11-10 | 2020-07-17 | 三星电子株式会社 | Apparatus and method for generating user interest information |
CN111491136A (en) * | 2020-04-23 | 2020-08-04 | 盈多伙伴(北京)科技有限公司 | Image transmission system and method |
CN111625672A (en) * | 2020-07-30 | 2020-09-04 | 腾讯科技(深圳)有限公司 | Image processing method, image processing device, computer equipment and storage medium |
CN113837286A (en) * | 2021-09-26 | 2021-12-24 | 展讯通信(天津)有限公司 | Image classification method and related equipment |
CN115687673A (en) * | 2022-11-08 | 2023-02-03 | 杭州晶彩数字科技有限公司 | Picture archiving method and device, electronic equipment and readable storage medium |
CN116302293A (en) * | 2023-05-18 | 2023-06-23 | 荣耀终端有限公司 | Picture display method and device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103995889A (en) * | 2014-06-03 | 2014-08-20 | 广东欧珀移动通信有限公司 | Method and device for classifying pictures |
CN104504109A (en) * | 2014-12-30 | 2015-04-08 | 百度在线网络技术(北京)有限公司 | Image search method and device |
CN105518679A (en) * | 2015-03-26 | 2016-04-20 | 北京旷视科技有限公司 | Image management method and image synchronization method |
CN105630915A (en) * | 2015-12-21 | 2016-06-01 | 山东大学 | Method and device for classifying and storing pictures in mobile terminals |
-
2016
- 2016-07-18 CN CN201610565611.XA patent/CN106445995A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103995889A (en) * | 2014-06-03 | 2014-08-20 | 广东欧珀移动通信有限公司 | Method and device for classifying pictures |
CN104504109A (en) * | 2014-12-30 | 2015-04-08 | 百度在线网络技术(北京)有限公司 | Image search method and device |
CN105518679A (en) * | 2015-03-26 | 2016-04-20 | 北京旷视科技有限公司 | Image management method and image synchronization method |
CN105630915A (en) * | 2015-12-21 | 2016-06-01 | 山东大学 | Method and device for classifying and storing pictures in mobile terminals |
Cited By (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108628869A (en) * | 2017-03-16 | 2018-10-09 | 富士施乐实业发展(中国)有限公司 | A kind of method and apparatus that category division is carried out to electronic document |
CN110019894A (en) * | 2017-07-21 | 2019-07-16 | 北京搜狗科技发展有限公司 | A kind of location lookup method and device |
CN110019894B (en) * | 2017-07-21 | 2022-12-06 | 北京搜狗科技发展有限公司 | Position searching method and device |
CN107657051A (en) * | 2017-10-16 | 2018-02-02 | 广东欧珀移动通信有限公司 | A kind of generation method of picture tag, terminal device and storage medium |
CN107657051B (en) * | 2017-10-16 | 2020-03-17 | Oppo广东移动通信有限公司 | Picture label generation method, terminal device and storage medium |
CN107590291A (en) * | 2017-10-30 | 2018-01-16 | 广东欧珀移动通信有限公司 | A kind of searching method of picture, terminal device and storage medium |
US11678012B2 (en) | 2017-11-10 | 2023-06-13 | Samsung Electronics Co., Ltd. | Apparatus and method for user interest information generation |
CN111434118B (en) * | 2017-11-10 | 2022-10-21 | 三星电子株式会社 | Apparatus and method for generating user interest information |
CN111434118A (en) * | 2017-11-10 | 2020-07-17 | 三星电子株式会社 | Apparatus and method for generating user interest information |
CN108021654A (en) * | 2017-12-01 | 2018-05-11 | 北京奇安信科技有限公司 | A kind of photograph album image processing method and device |
CN108052563A (en) * | 2017-12-05 | 2018-05-18 | 广东小天才科技有限公司 | A kind of picture searching method and mobile terminal |
CN108121816A (en) * | 2017-12-28 | 2018-06-05 | 广东欧珀移动通信有限公司 | Picture classification method, device, storage medium and electronic equipment |
CN108121816B (en) * | 2017-12-28 | 2020-09-08 | Oppo广东移动通信有限公司 | Picture classification method and device, storage medium and electronic equipment |
CN108416003A (en) * | 2018-02-27 | 2018-08-17 | 百度在线网络技术(北京)有限公司 | A kind of picture classification method and device, terminal, storage medium |
CN108596280A (en) * | 2018-04-27 | 2018-09-28 | 佛山市日日圣科技有限公司 | Image identification method for literature search |
CN110019913A (en) * | 2018-06-01 | 2019-07-16 | 平安好房(上海)电子商务有限公司 | Picture match method, user equipment, storage medium and device |
WO2020007201A1 (en) * | 2018-07-04 | 2020-01-09 | 奇酷互联网络科技(深圳)有限公司 | Picture search method and device, readable storage medium, and terminal |
CN109033237A (en) * | 2018-07-04 | 2018-12-18 | 奇酷互联网络科技(深圳)有限公司 | Image searching method, device, readable storage medium storing program for executing and terminal |
CN111061900A (en) * | 2018-10-17 | 2020-04-24 | 丽宝大数据股份有限公司 | Searching method for personal wearing record |
CN109635153B (en) * | 2018-11-22 | 2023-01-13 | 腾讯音乐娱乐科技(深圳)有限公司 | Migration path generation method, device and storage medium |
CN109635153A (en) * | 2018-11-22 | 2019-04-16 | 腾讯音乐娱乐科技(深圳)有限公司 | Migration path generation method, device and storage medium |
CN113296680A (en) * | 2018-11-27 | 2021-08-24 | 浙江口碑网络技术有限公司 | Dish picture uploading method and device, computer equipment and readable storage medium |
CN109766052A (en) * | 2018-11-27 | 2019-05-17 | 浙江口碑网络技术有限公司 | Vegetable picture upload method, device, computer equipment and readable storage medium storing program for executing |
CN110162645A (en) * | 2019-05-28 | 2019-08-23 | 广东三维家信息科技有限公司 | Image search method, device and electronic equipment based on index |
CN110703972A (en) * | 2019-09-29 | 2020-01-17 | 维沃移动通信有限公司 | File control method and electronic equipment |
CN111368111A (en) * | 2020-02-28 | 2020-07-03 | Oppo(重庆)智能科技有限公司 | Picture retrieval method and device for mobile equipment and mobile equipment |
CN111491136A (en) * | 2020-04-23 | 2020-08-04 | 盈多伙伴(北京)科技有限公司 | Image transmission system and method |
CN111625672A (en) * | 2020-07-30 | 2020-09-04 | 腾讯科技(深圳)有限公司 | Image processing method, image processing device, computer equipment and storage medium |
CN113837286A (en) * | 2021-09-26 | 2021-12-24 | 展讯通信(天津)有限公司 | Image classification method and related equipment |
CN115687673A (en) * | 2022-11-08 | 2023-02-03 | 杭州晶彩数字科技有限公司 | Picture archiving method and device, electronic equipment and readable storage medium |
CN116302293A (en) * | 2023-05-18 | 2023-06-23 | 荣耀终端有限公司 | Picture display method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106445995A (en) | Picture classification method and apparatus | |
US10586108B2 (en) | Photo processing method and apparatus | |
US10956784B2 (en) | Neural network-based image manipulation | |
Ashraf et al. | MDCBIR-MF: multimedia data for content-based image retrieval by using multiple features | |
CN103069415B (en) | Computer-implemented method, computer program and computer system for image procossing | |
CN103631819B (en) | A kind of method and system of picture name | |
WO2020007201A1 (en) | Picture search method and device, readable storage medium, and terminal | |
US20180025002A1 (en) | Hue-based color naming for an image | |
WO2019127832A1 (en) | Intelligent search method and apparatus, terminal, server, and storage medium | |
TWI781554B (en) | Method of determining item name of object, device, computer equipment and storage medium | |
CN104765891A (en) | Searching shopping method based on pictures | |
CN102831200A (en) | Commodity propelling method and device based on image character recognition | |
CN102317929A (en) | Method and system for image matching | |
CN107590267B (en) | Information-pushing method and device, terminal and readable storage medium storing program for executing based on picture | |
CN103988202A (en) | Image attractiveness based indexing and searching | |
CN108038506A (en) | A kind of library automatic classification method | |
US9286325B2 (en) | Methods and systems for ranking images using semantic and aesthetic models | |
CN105117399B (en) | Image searching method and device | |
CN105631051A (en) | Character recognition based mobile augmented reality reading method and reading system thereof | |
TWI651673B (en) | Seraching method for product photograph and searching system for product photograph | |
CN107153838A (en) | A kind of photo automatic grading method and device | |
JP2011253424A (en) | Image recognition device and image recognition method and information processing system | |
CN105956051A (en) | Information finding method, device and system | |
CN105260385B (en) | A kind of picture retrieval method | |
CN108431829A (en) | System and method for searching for product in catalogue |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170222 |