CN106708940A - Method and device used for processing pictures - Google Patents
Method and device used for processing pictures Download PDFInfo
- Publication number
- CN106708940A CN106708940A CN201611042535.0A CN201611042535A CN106708940A CN 106708940 A CN106708940 A CN 106708940A CN 201611042535 A CN201611042535 A CN 201611042535A CN 106708940 A CN106708940 A CN 106708940A
- Authority
- CN
- China
- Prior art keywords
- picture
- search
- index
- vision
- keyword set
- 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
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method and a device used for processing pictures. One specific implementation way of the method comprises the following steps that: obtaining a picture and text information related to the picture; analyzing the above picture to obtain a first keyword set, analyzing the above text information to obtain a second keyword set, and generating a text index used for indicating the above picture according to the above first keyword set and the above second keyword set; selecting a keyword used for representing the visual feature of the above picture from the above first keyword set to generate a visual index used for indicating the above picture; and storing the above text index and the above visual index. By use of the implementation way, high-accuracy index establishment and storage can be realized.
Description
Technical field
The application is related to field of computer technology, and in particular to picture Processing Technique field, more particularly, to treatment figure
The method and apparatus of piece.
Background technology
Existing picture searching platform, picture searching is greatly enhanced frequently by the index related to picture is set up
Efficiency.
However, the search text that existing picture searching platform is generally submitted to the Search Results that user is presented with user
Or search lacks correlation, its reason with picture it is usually because an index degree of accuracy set up and store is relatively low.
The content of the invention
In a first aspect, this application provides a kind of method for processing picture, the method includes：Obtain picture and with it is upper
State the related text message of picture；Above-mentioned picture is analyzed to obtain the first keyword set, above-mentioned text message is entered
Row analysis is generated and is used for obtain the second keyword set according to above-mentioned first keyword set and above-mentioned second keyword set
The text index indicated above-mentioned picture；The vision for characterizing above-mentioned picture is chosen from above-mentioned first keyword set
The keyword of feature, is indexed with the vision generated for being indicated above-mentioned picture；To above-mentioned text index and above-mentioned vision
Index is stored.
In certain embodiments, the above method also includes：Search picture in response to receiving user terminal transmission, then
Above-mentioned search picture is analyzed to extract visual signature, by the visual signature of above-mentioned search picture and above-mentioned vision rope
Row matching is introduced, the indicated picture of the vision index that will be matched is shown as Search Results.
In certain embodiments, the above method also includes：Search text in response to receiving user terminal transmission, then
Above-mentioned search text is analyzed to generate the 3rd keyword set, by above-mentioned 3rd keyword set and above-mentioned text rope
Row matching is introduced, the picture indicated by text index that will be matched is shown as Search Results.
In certain embodiments, above-mentioned index above-mentioned search with above-mentioned vision with the visual signature of picture is matched,
Including：Calculate the matching degree that each vision in the visual signature of above-mentioned search picture and above-mentioned vision index is indexed；Will
The vision for being not less than first threshold with the matching degree of the visual signature of above-mentioned search picture in above-mentioned vision index indexes signified
The picture for showing as with above-mentioned search picture identical picture, will be low with the matching degree of the visual signature of above-mentioned search picture
In above-mentioned first threshold and be not less than the indicated picture of vision index of Second Threshold as with above-mentioned search picture phase
As picture.
In certain embodiments, the indicated picture of the above-mentioned vision index that will be matched is opened up as Search Results
Show, including：To be shown as Search Results to above-mentioned search picture identical picture and/or similar picture.
In certain embodiments, above-mentioned visual signature include it is following at least one：Color, texture, shape.
Second aspect, this application provides a kind of device for processing picture, said apparatus include：Acquiring unit, matches somebody with somebody
Put for obtaining picture and the text message related to above-mentioned picture；Text index generation unit, is configured to above-mentioned picture
It is analyzed to obtain the first keyword set, above-mentioned text message is analyzed to obtain the second keyword set, according to
Above-mentioned first keyword set and above-mentioned second keyword set generate the text index for being indicated above-mentioned picture；Depending on
Feel index generation unit, be configured to be chosen from above-mentioned first keyword set visual signature for characterizing above-mentioned picture
Keyword, is indexed with the vision generated for being indicated above-mentioned picture；Memory cell, is configured to above-mentioned text index
Stored with above-mentioned vision index.
In certain embodiments, said apparatus also include：First processing units, are configured in response to receiving user's end
The search picture for sending is held, then above-mentioned search picture is analyzed to extract visual signature, by above-mentioned search picture
Visual signature and above-mentioned vision index matched, the indicated picture of the vision index that will be matched enters as Search Results
Row displaying.
In certain embodiments, said apparatus also include：Second processing unit, is configured in response to receiving user's end
The search text for sending is held, then above-mentioned search text is analyzed to generate the 3rd keyword set, by the above-mentioned 3rd
Keyword set is matched with above-mentioned text index, and the picture indicated by text index that will be matched enters as Search Results
Row displaying.
In certain embodiments, above-mentioned first processing units include：Computation subunit, is configured to calculate above-mentioned search use
The visual signature of picture and above-mentioned vision index in the matching degree that indexes of each vision；Treatment subelement, be configured to by
The vision for being not less than first threshold with the matching degree of the visual signature of above-mentioned search picture in above-mentioned vision index indexes signified
The picture for showing as with above-mentioned search picture identical picture, will be low with the matching degree of the visual signature of above-mentioned search picture
In above-mentioned first threshold and be not less than the indicated picture of vision index of Second Threshold as with above-mentioned search picture phase
As picture.
In certain embodiments, above-mentioned first processing units include：Displaying subelement, is configured to be used with above-mentioned search
Picture identical picture and/or similar picture are shown as Search Results.
In certain embodiments, above-mentioned visual signature include it is following at least one：Color, texture, shape.
The method and apparatus for processing picture that the application is provided, are analyzed to obtain by acquired picture
First keyword set, is analyzed to obtain the second keyword set by the acquired text information related to picture
Close, afterwards according to the first keyword set and the second keyword set to generate the text index for being indicated picture,
Then the keyword of the visual signature for characterizing picture is chosen from the first keyword set to generate for carrying out picture
The vision index of instruction, finally stores, to text index and vision index so as to be effectively utilized the first keyword set
With the second keyword set, realize the index foundation of high accuracy and store.
Brief description of the drawings
By the detailed description made to non-limiting example made with reference to the following drawings of reading, the application other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is that the application can apply to exemplary system architecture figure therein；
Fig. 2 is the flow chart of one embodiment of the method for processing picture according to the application；
Fig. 3 is the flow chart of another embodiment of the method for processing picture according to the application；
Fig. 4 is the flow chart of the further embodiment of the method for processing picture according to the application；
Fig. 5 is the structural representation of one embodiment of the device for processing picture according to the application；
Fig. 6 is adapted for the structural representation of the computer system of the server for realizing the embodiment of the present application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that, in order to
Be easy to description, be illustrate only in accompanying drawing to about the related part of invention.
It should be noted that in the case where not conflicting, the feature in embodiment and embodiment in the application can phase
Mutually combination.Describe the application in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 shows can apply the method for processing picture of the application or the implementation for processing the device of picture
The exemplary system architecture 100 of example.
As shown in figure 1, system architecture 100 can include terminal device 101,102,103, network 104,106, server
105 and database 107.Network 104 is to the offer communication link between terminal device 101,102,103 and server 105
Medium.Network 106 is used to be provided between server 105 and database 107 medium of communication link.Network 104,106 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be interacted by network 104 with using terminal equipment 101,102,103 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed on terminal device 101,102,103, such as picture transfer application,
Picture searching instrument, web browser applications, JICQ etc..
Terminal device 101,102,103 can be the various electronic equipments with display screen, including but not limited to intelligent hand
Machine, panel computer, pocket computer on knee and desktop computer etc..
Database 107 can be the picture database of an at least pictures of being stored with.
Server 105 can be to provide the server of various services.For example server 105 can from terminal device 101,
102nd, 103 and/or database 107 obtain picture and the text message related to picture；Can be to acquired picture and text
Information such as is analyzed at the treatment, is indexed with the text index and vision that generate for being indicated picture；Can also be to giving birth to
Into text index and vision index stored.
It should be noted that the method for processing picture that the embodiment of the present application is provided typically is held by server 105
OK, correspondingly, it is generally positioned in server 105 for processing the device of picture.
It should be understood that the number of the terminal device, network, server and database in Fig. 1 is only schematical.According to
Realize needing, can have any number of terminal device, network, server and database.In server 105 from local acquisition
In the case of picture and the text message related to picture, terminal device 101,102,103 and database 107 can.
With continued reference to Fig. 2, Fig. 2 shows the flow of one embodiment of the method for processing picture according to the application
200.This is used for the method for processing picture, comprises the following steps：
Step 201, obtains picture and the text message related to picture.
In the present embodiment, electronic equipment (such as service shown in Fig. 1 thereon is run for processing the method for picture
Device 105) picture and the text message related to picture can be obtained by wired connection mode or radio connection.This
In, above-mentioned picture can be the picture for processing the arbitrary format that the method for picture is supported, such as PNG, JPEG, GIF etc.
The picture of form.
In some optional implementations of the present embodiment, above-mentioned electronic equipment can be from the data for communicating with connection
Storehouse (such as the database 107 shown in Fig. 1) and/or user terminal (such as the terminal device 101,102,103 shown in Fig. 1) are obtained
Above-mentioned picture and above-mentioned text message；Above-mentioned electronic equipment can also locally obtain above-mentioned picture and above-mentioned text message.Here,
Above-mentioned electronic equipment can be by acquired above-mentioned picture and above-mentioned text information storage to internal memory or hard disk.
Step 202, is analyzed to picture to obtain the first keyword set, and text message is analyzed to obtain
Two keyword sets, the text rope for being indicated picture is generated according to the first keyword set and the second keyword set
Draw.
In the present embodiment, above-mentioned electronic equipment can be analyzed using image recognition technology to picture, so as to extract
Go out one or more keywords, the keyword for being then based on extracting generates the first keyword set.As an example, a display
There is the picture of night scene, comprising star and bent moon, background colour is black；Above-mentioned electronic equipment is by image recognition technology to the picture
It is analyzed, keyword " star ", " moon ", " reaping hook ", " black ", " yellow ", " curved ", " five points ", " evening can be obtained
On ", " banana ", above-mentioned electronic equipment can be by above-mentioned keyword " star ", " moon ", " reaping hook ", " black ", " yellow ", " curved
", " five points ", " evening " and " banana " be used as the keyword in the first keyword set.
In the present embodiment, above-mentioned electronic equipment can be analyzed using various analysis means to above-mentioned text message,
So as to extract one or more keywords, the extracted crucial set of keyword generation second is then based on.
In the present embodiment, above-mentioned electronic equipment can be used according to the first keyword set and the generation of the second keyword set
In the text index indicated picture.As an example, above-mentioned electronic equipment can by merge the first keyword set and
Second keyword set, to generate above-mentioned text index.
In some optional implementations of the present embodiment, above-mentioned electronic equipment can be by merging the first keyword set
Close and the second keyword set merges the keyword that removal is repeated, to generate the text index for being indicated picture.
In some optional implementations of the present embodiment, in the first keyword set and the second keyword set
In the case that each keyword is not provided with weight, above-mentioned electronic equipment can be by the first keyword set and the second keyword set
Merge to obtain crucial phrase, above-mentioned electronic equipment can remove the keyword of the repetition in above-mentioned crucial phrase afterwards,
Then above-mentioned electronic equipment can be based on the correlation of each keyword in above-mentioned crucial phrase and above-mentioned picture, be above-mentioned
Each keyword in crucial phrase sets weight, and then above-mentioned electronic equipment can be not low based on weight in stating crucial phrase
In the keyword of preset value generating above-mentioned text index.
In some optional implementations of the present embodiment, the analysis mode to above-mentioned text message can be statistical
Analysis mode.For example, the frequency of occurrences of each word present in above-mentioned text message can be counted and sorted, afterwards,
The frequency of occurrences is chosen again sorts one or more forward words as keyword.
In some optional implementations of the present embodiment, the analysis mode to above-mentioned text message can also be semantic
Analysis mode.As an example, full cutting method etc. can be carried out with above-mentioned text message process, the content segmentation of text message into
Word；Importance calculating is carried out to resulting word again (for example with word frequency-reverse document-frequency method (Term Frequency-
Inverse Document Frequency, TF-IDF)), keyword is obtained based on the result of importance calculating.
Using full cutting method, all possible word matched with language dictionary can be first syncopated as, then with statistics
Language model determines optimal cutting result.By taking text " star and the moon in the night sky " as an example, language word can be first carried out
Storehouse matches, and finds all words --- the night of matching, the night sky, in the air, star, the moon；These words are with word grid (word
Lattices) form represents, is next based on word grid and does route searching, then based on statistical language model (such as N-Gram moulds
Type) find optimal path.If result shows the language model scores highest of " night sky star moon ", " night sky star month
It is bright " it is the optimal cutting of " star and the moon in the night sky ".Above-mentioned N-Gram models are a kind of conventional language moulds herein
Type, for Chinese, can be referred to as Chinese language model (Chinese Language Model, CLM).The N-Gram models
Based on such a it is assumed that the appearance of n-th word is only related to above N-1 word, and, whole sentence all uncorrelated to other any words
Probability be exactly each word probability of occurrence product, and these probability can be by directly counting N number of word while going out from language material
Existing number of times is obtained.
Using full cutting method by content segmentation into after word, can be using word frequency-reverse document-frequency method to these
Word carries out importance calculating, then carries out prominence score as keyword or to these words choosing word based on importance.Word
The main thought of frequently-reverse document-frequency method is, if the frequency (Term that certain word or phrase occur in an article
Frequency, TF) it is high, and seldom occur in other articles, then it is assumed that this word or phrase have good class discrimination
Ability, is adapted to classification.And reverse document-frequency (Inverse Document Frequency, IDF) is primarily referred to as, if
Document comprising certain word or phrase is fewer, then IDF is bigger, then illustrate that the word or phrase have good class discrimination ability.
Thus, using word frequency-reverse document-frequency method, the importance of certain word or phrase inside certain article can be calculated.
It should be noted that the various methods of above-mentioned semantic analysis mode are widely studied at present and application known skills
Art, will not be repeated here.
In some optional implementations of the present embodiment, above-mentioned electronic equipment can be by image analysis technology point
Separate out the keyword related to above-mentioned picture, and analyzed using various analysis means it is related to above-mentioned text message
Keyword sets weight, and the keyword that can be not less than threshold value based on weight generates the first keyword set and the second keyword set
Close.Wherein, the size and keyword of weight are relevant with the correlation of image content, and correlation is higher, and weighted value is higher, related
The lower weighted value of property is lower.As an example, above-mentioned electronic equipment passes through image recognition technology to the above-mentioned figure for showing night scene
Piece is analyzed, and it is " star ", " moon ", " reaping hook ", " black ", " yellow ", " curved ", " five points ", " evening to obtain keyword
On ", " banana "；Wherein, the weight of " star " and " moon " can be 0.5, and the weight of " black " and " yellow " can be 0.4,
The weight of " curved " can be 0.3, and the weight in " evening " and " reaping hook " can be 0.2, and the weight of " five points " can be 0.1, " fragrant
The weight of any of several broadleaf plants " can be 0；If threshold value is 0.2, above-mentioned electronic equipment can be based on keyword " star ", " moon ", " black
Color ", " yellow ", " curved ", " evening " and " reaping hook " generate the first keyword set.
Step 203, chooses the keyword of the visual signature for characterizing picture, to generate use from the first keyword set
Indexed in the vision indicated picture.
In the present embodiment, above-mentioned electronic equipment can choose the vision for characterizing picture from the first keyword set
The keyword of feature (such as yellow, black, curved etc.), is indexed with the vision generated for being indicated picture.
In some optional implementations of the present embodiment, above-mentioned visual signature can include it is following at least one：Face
Color, texture, shape.
Step 204, stores to text index and vision index.
In the present embodiment, above-mentioned electronic equipment can be stored to above-mentioned electronic equipment text index and vision index
In internal memory or hard disk, can also store into the database communicated to connect with above-mentioned electronic equipment.
The method that above-described embodiment of the application is provided is crucial to obtain first by being analyzed to acquired picture
Set of words, is analyzed to obtain the second keyword set, Zhi Hougen by the acquired text message related to picture
According to the first keyword set and the second keyword set to generate the text index for being indicated picture, then from first
The keyword of the visual signature for characterizing picture is chosen in keyword set to generate the vision for indicating picture
Index, finally stores to text index and vision index, so as to be effectively utilized the first keyword set and the second key
Set of words, the index for realizing high accuracy is set up and is stored.
With further reference to Fig. 3, it illustrates the flow 300 of another embodiment of the method for processing picture, the use
In the flow 300 of the method for the treatment of picture, comprise the following steps：
Step 301, obtains picture and the text message related to picture.
In the present embodiment, electronic equipment (such as service shown in Fig. 1 thereon is run for processing the method for picture
Device 105) picture and the text message related to picture can be obtained by wired connection mode or radio connection.This
In, above-mentioned picture can be the picture for processing the arbitrary format that the method for picture is supported, such as PNG, JPEG, GIF etc.
The picture of form.
Step 302, is analyzed to picture to obtain the first keyword set, and text message is analyzed to obtain
Two keyword sets, the text rope for being indicated picture is generated according to the first keyword set and the second keyword set
Draw.
In the present embodiment, above-mentioned electronic equipment can be analyzed using image recognition technology to picture, so as to extract
Go out one or more keywords, the keyword for being then based on extracting generates the first keyword set.And above-mentioned electronic equipment
Above-mentioned text message can be analyzed using various analysis means, so as to extract one or more keywords, Ran Houji
In the crucial set of keyword generation second for being extracted.And above-mentioned electronic equipment can be by merging the first keyword set
With the second keyword set, to generate for the text index that is indicated picture.
Step 303, chooses the keyword of the visual signature for characterizing picture, to generate use from the first keyword set
Indexed in the vision indicated picture.
In the present embodiment, above-mentioned electronic equipment can choose the vision for characterizing picture from the first keyword set
The keyword of feature (such as yellow, black, curved etc.), is indexed with the vision generated for being indicated picture.
Step 304, stores to text index and vision index.
In the present embodiment, above-mentioned electronic equipment can be stored to above-mentioned electronic equipment text index and vision index
In internal memory or hard disk, can also store into the database communicated to connect with above-mentioned electronic equipment.
Step 305, in response to receive user terminal transmission search picture, then search picture is analyzed with
Extract visual signature.
In the present embodiment, the search picture of user terminal transmission is received in response to above-mentioned electronic equipment, then it is above-mentioned
Electronic equipment can be analyzed to search by visual search technology with picture, so as to extract at least one visual signature
(such as visual signature such as color, texture, shape).
It should be noted that above-mentioned visual search technology is normally based on the visual search technology of image content analysis.On
State visual search technology with correct understanding image content, and to realize scheming to search figure, picture searching effect can be effectively improved
Rate and search quality, are conducive to auxiliary to recommend more to meet the Search Results of user's request, and improve user satisfaction.
Step 306, search is indexed with the visual signature of picture with vision and is matched.
In the present embodiment, above-mentioned electronic equipment can be by calculating the visual signature and above-mentioned vision rope of search picture
The matching degree of each vision index in drawing, search is indexed with the visual signature of picture with vision and is matched.As
The visual signature of search picture can be converted to search keyword set by example, above-mentioned electronic equipment, can be by meter
The matching degree that each vision during search keyword set is indexed with above-mentioned vision is indexed is calculated, to obtain search picture
Visual signature and above-mentioned vision index in the matching degree that indexes of each vision.If for example, search keyword set bag
Containing 10 keywords, some vision index in above-mentioned vision index includes 9 keywords in search keyword set,
The vision rope of 9 keywords in the visual signature of search picture and the above-mentioned keyword set comprising search can then be determined
The matching degree drawn is that 9 and 10 ratio, i.e. matching degree are 90%.
In some optional implementations of the present embodiment, above-mentioned electronic equipment can by above-mentioned vision index in search
The matching degree of the visual signature of rope picture is not less than the picture conduct indicated by the vision index of first threshold (such as 95%)
With search picture identical picture, above-mentioned first threshold can will be less than simultaneously with the matching degree of the visual signature of search picture
And the indicated picture of the vision index of Second Threshold (such as 70%) is not less than as the picture similar to search picture.Need
It is noted that above-mentioned first threshold and Second Threshold can be artificial settings, can be modified according to needs are realized.
Step 307, the picture indicated by vision index that will be matched is shown as Search Results.
In the present embodiment, the matching degree with the visual signature of search picture that above-mentioned electronic equipment will can be matched
Order of the picture indicated by vision index being not zero according to matching degree from high to low is ranked up, then by the figure after sequence
Piece is shown to user terminal.
In some optional implementations of the present embodiment, above-mentioned electronic equipment will can be matched schemes with search
Piece identical picture is shown as Search Results to user terminal.
In some optional implementations of the present embodiment, above-mentioned electronic equipment will can be matched schemes with search
Order of the identical and similar picture of piece according to matching degree from high to low is ranked up, then by sequence after with search scheme
The identical and similar picture of piece is shown to user terminal.
From figure 3, it can be seen that compared with the corresponding embodiments of Fig. 2, the method for processing picture in the present embodiment
Flow 300 highlight step 305, step 306 and step 307.Thus, the scheme of the present embodiment description can realize high precision
The index of degree is set up and is stored；Can realize to scheme to search figure, optimization picture search process, and also picture searching effect can be improved
Rate.
With further reference to Fig. 4, it illustrates the flow 400 of the further embodiment of the method for processing picture.The use
In the flow 400 for the treatment of picture, comprise the following steps：
Step 401, obtains picture and the text message related to picture.
In the present embodiment, electronic equipment (such as service shown in Fig. 1 thereon is run for processing the method for picture
Device 105) picture and the text message related to picture can be obtained by wired connection mode or radio connection.This
In, above-mentioned picture can be the picture for processing the arbitrary format that the method for picture is supported, such as PNG, JPEG, GIF etc.
The picture of form.
Step 402, is analyzed to picture to obtain the first keyword set, and text message is analyzed to obtain
Two keyword sets, the text rope for being indicated picture is generated according to the first keyword set and the second keyword set
Draw.
In the present embodiment, above-mentioned electronic equipment can be analyzed using image recognition technology to picture, so as to extract
Go out one or more keywords, the keyword for being then based on extracting generates the first keyword set.And above-mentioned electronic equipment
Above-mentioned text message can be analyzed using various analysis means, so as to extract one or more keywords, Ran Houji
In the crucial set of keyword generation second for being extracted.And above-mentioned electronic equipment can be by merging the first keyword set
With the second keyword set, to generate for the text index that is indicated picture.
Step 403, chooses the keyword of the visual signature for characterizing picture, to generate use from the first keyword set
Indexed in the vision indicated picture.
In the present embodiment, above-mentioned electronic equipment can choose the vision for characterizing picture from the first keyword set
The keyword of feature (such as yellow, black, curved etc.), is indexed with the vision generated for being indicated picture.
Step 404, stores to text index and vision index.
In the present embodiment, above-mentioned electronic equipment can be stored to above-mentioned electronic equipment text index and vision index
In internal memory or hard disk, can also store into the database communicated to connect with above-mentioned electronic equipment.
Step 405, in response to receive user terminal transmission search text, then search text is analyzed with
Generate the 3rd keyword set.
In the present embodiment, above-mentioned electronic equipment can be used and the analysis method phase being analyzed to above-mentioned text message
With analysis method search is analyzed with text, generating the 3rd keyword set.
Step 406, the 3rd keyword set is matched with text index.
In the present embodiment, above-mentioned electronic equipment can be by each in the 3rd keyword set and above-mentioned text index
Text index is matched.As an example, above-mentioned electronic equipment can will include the 3rd keyword set in above-mentioned text index
At least one of keyword text index as the text index matched with the 3rd keyword set.
In some optional implementations of the present embodiment, above-mentioned electronic equipment can calculate the 3rd keyword set with
The matching degree of each text index in above-mentioned text index, above-mentioned electronic equipment can by with the 3rd keyword set
The text index of the 3rd threshold value (such as 70%) is not less than as the text index matched with the 3rd keyword set with degree.As
Example, the 3rd keyword set includes keyword " star ", " moon ", " black " and " yellow ", in above-mentioned text index one
Individual text index includes " star " and " moon ", then the above-mentioned text index comprising " star " and " moon " and the 3rd keyword set
The matching degree of conjunction can be 2 and 4 ratio, i.e., 50%.Here, above-mentioned 3rd threshold value can be artificial setting, can basis
Realize that needs are modified.
Step 407, the picture indicated by text index that will be matched is shown as Search Results.
In the present embodiment, above-mentioned electronic equipment can in above-mentioned text index comprising in the 3rd keyword set extremely
Picture indicated by a few text index for keyword is shown as Search Results to user terminal.
In some optional implementations of the present embodiment, above-mentioned electronic equipment can by above-mentioned text index with
The matching degree of three keyword sets be not less than picture indicated by the text index of the 3rd threshold value according to matching degree by it is high on earth
Order is ranked up, and then the picture after sequence is shown to user terminal.
Figure 4, it is seen that compared with the corresponding embodiments of Fig. 2, the method for processing picture in the present embodiment
Flow 400 highlight step 405, step 406 and step 407.Thus, the scheme of the present embodiment description can realize high precision
The index of degree is set up and is stored；Carried out by by each text index in the 3rd keyword set and above-mentioned text index
Match somebody with somebody, picture search process can be optimized, improve the correlation of Search Results and search text, and improve picture searching effect
Rate.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, scheme for processing this application provides one kind
One embodiment of the device of piece, the device embodiment is corresponding with the embodiment of the method shown in Fig. 2, and the device can specifically be answered
For in various electronic equipments.
As shown in figure 5, the present embodiment can include for processing the device 500 of picture：Acquiring unit 501, text rope
Draw generation unit 502, vision index generation unit 503 and memory cell 504.Wherein, acquiring unit 501, are configured to obtain
Picture and the text message related to above-mentioned picture；Text index generation unit 502, is configured to be analyzed above-mentioned picture
To obtain the first keyword set, above-mentioned text message is analyzed to obtain the second keyword set, according to above-mentioned first
Keyword set and above-mentioned second keyword set generate the text index for being indicated above-mentioned picture；Vision index life
Into unit 503, it is configured to be chosen from above-mentioned first keyword set the key of the visual signature for characterizing above-mentioned picture
Word, is indexed with the vision generated for being indicated above-mentioned picture；Memory cell 504, is configured to above-mentioned text index
Stored with above-mentioned vision index.
For in the device 500 for processing picture：Acquiring unit 501, the index generation of text index generation unit 502, vision
Unit 503 and memory cell 504 can be referring to the step 201 in the corresponding embodiments of Fig. 2, step 202, step 203 and steps 204
The associated description of implementation, will not be repeated here.
In certain embodiments, said apparatus 500 can also include：First processing units (not shown), configuration is used
In the search picture in response to receiving user terminal transmission, then above-mentioned search picture is analyzed special to extract vision
Levy, above-mentioned search is indexed with the visual signature of picture with above-mentioned vision and is matched, the vision index that will be matched is indicated
Picture be shown as Search Results.
In certain embodiments, said apparatus 500 can also include：Second processing unit (not shown), configuration is used
In the search text in response to receiving user terminal transmission, then above-mentioned search text is analyzed to generate the 3rd pass
Keyword set, above-mentioned 3rd keyword set is matched with above-mentioned text index, indicated by the text index that will be matched
Picture be shown as Search Results.
In certain embodiments, above-mentioned first processing units can include：Computation subunit (not shown), configuration is used
The matching degree that each vision in the visual signature for calculating above-mentioned search picture is indexed with above-mentioned vision is indexed；Treatment
Unit (not shown), in being configured to index above-mentioned vision with the matching degree of the visual signature of above-mentioned search picture not
The picture indicated less than the vision index of first threshold as with above-mentioned search picture identical picture, will be with above-mentioned search
Above-mentioned first threshold is less than with the matching degree of the visual signature of picture and be not less than indicated by the vision index of Second Threshold
Picture as to the above-mentioned search similar picture of picture.
In certain embodiments, above-mentioned first processing units can include：Displaying subelement (not shown), configuration is used
It is shown as Search Results to above-mentioned search picture identical picture and/or similar picture in using above-mentioned.
In certain embodiments, above-mentioned visual signature can include it is following at least one：Color, texture, shape.
Below with reference to Fig. 6, it illustrates the computer system 600 for being suitable to the server for realizing the embodiment of the present application
Structural representation.
As shown in fig. 6, computer system 600 includes CPU (CPU) 601, it can be according to storage read-only
Program in memory (ROM) 602 or be loaded into program in random access storage device (RAM) 603 from storage part 608 and
Perform various appropriate actions and treatment.In RAM 603, the system that is also stored with 600 operates required various programs and data.
CPU 601, ROM 602 and RAM 603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to always
Line 604.
I/O interfaces 605 are connected to lower component：Including the importation 606 of keyboard, mouse etc.；Penetrated including such as negative electrode
The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage part 608 including hard disk etc.；
And the communications portion 609 of the NIC including LAN card, modem etc..Communications portion 609 via such as because
The network of spy's net performs communication process.Driver 610 is also according to needing to be connected to I/O interfaces 605.Detachable media 611, such as
Disk, CD, magneto-optic disk, semiconductor memory etc., as needed on driver 610, in order to read from it
Computer program be mounted into as needed storage part 608.
Especially, in accordance with an embodiment of the present disclosure, the process above with reference to flow chart description may be implemented as computer
Software program.For example, embodiment of the disclosure includes a kind of computer program product, it includes being tangibly embodied in machine readable
Computer program on medium, above computer program bag is containing the program code for the method shown in execution flow chart.At this
In the embodiment of sample, the computer program can be downloaded and installed by communications portion 609 from network, and/or from removable
Medium 611 is unloaded to be mounted.When the computer program is performed by CPU (CPU) 601, in execution the present processes
The above-mentioned functions of restriction.
Flow chart and block diagram in accompanying drawing, it is illustrated that according to the system of the various embodiments of the application, method and computer journey
The architectural framework in the cards of sequence product, function and operation.At this point, each square frame in flow chart or block diagram can generation
One part for module, program segment or code of table a, part for above-mentioned module, program segment or code includes one or more
Executable instruction for realizing the logic function of regulation.It should also be noted that in some realizations as replacement, institute in square frame
The function of mark can also occur with different from the order marked in accompanying drawing.For example, two square frame reality for succeedingly representing
On can perform substantially in parallel, they can also be performed in the opposite order sometimes, and this is depending on involved function.Also
It is noted that the combination of the square frame in each square frame and block diagram and/or flow chart in block diagram and/or flow chart, Ke Yiyong
Perform the function of regulation or the special hardware based system of operation to realize, or can be referred to computer with specialized hardware
The combination of order is realized.
Being described in involved unit in the embodiment of the present application can be realized by way of software, it is also possible to by hard
The mode of part is realized.Described unit can also be set within a processor, for example, can be described as：A kind of processor bag
Include acquiring unit, text index generation unit, vision index generation unit and memory cell.Wherein, the title of these units exists
The restriction to the unit in itself is not constituted in the case of certain, for example, acquiring unit be also described as " obtain picture and with
The unit of the related text message of picture ".
As on the other hand, present invention also provides a kind of nonvolatile computer storage media, the non-volatile calculating
Machine storage medium can be the nonvolatile computer storage media included in said apparatus in above-described embodiment；Can also be
Individualism, without the nonvolatile computer storage media allocated into terminal.Above-mentioned nonvolatile computer storage media is deposited
One or more program is contained, when said one or multiple programs are performed by an equipment so that the said equipment：Obtain
Picture and the text message related to picture；Above-mentioned picture is analyzed to obtain the first keyword set, to above-mentioned text
Information is analyzed to obtain the second keyword set, according to above-mentioned first keyword set and above-mentioned second keyword set symphysis
Into the text index for being indicated above-mentioned picture；Chosen for characterizing above-mentioned picture from above-mentioned first keyword set
Visual signature keyword, indexed for the vision that is indicated above-mentioned picture with generating；To above-mentioned text index and upper
Vision index is stated to be stored.
Above description is only the preferred embodiment and the explanation to institute's application technology principle of the application.People in the art
Member is it should be appreciated that involved invention scope in the application, however it is not limited to the technology of the particular combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where foregoing invention design is not departed from, is carried out by above-mentioned technical characteristic or its equivalent feature
Other technical schemes for being combined and being formed.Such as features described above has similar work(with (but not limited to) disclosed herein
The technical scheme that the technical characteristic of energy is replaced mutually and formed.
Claims (12)
1. a kind of method for processing picture, it is characterised in that methods described includes：
Obtain picture and the text message related to the picture；
The picture is analyzed to obtain the first keyword set, the text message is analyzed to obtain the second pass
Keyword set, generates for indicating the picture according to first keyword set and second keyword set
Text index；
The keyword of the visual signature for characterizing the picture is chosen from first keyword set, to generate for right
The vision index that the picture is indicated；
The text index and vision index are stored.
2. method according to claim 1, it is characterised in that methods described also includes：
Search picture in response to receiving user terminal transmission, then be analyzed to extract vision to the search picture
Feature, the search is indexed with the visual signature of picture with the vision and is matched, and the vision index that will be matched is signified
The picture for showing is shown as Search Results.
3. method according to claim 1, it is characterised in that methods described also includes：
Search text in response to receiving user terminal transmission, then be analyzed to generate the 3rd to the search text
Keyword set, the 3rd keyword set is matched with the text index, and the text index that will be matched is signified
The picture for showing is shown as Search Results.
4. method according to claim 2, it is characterised in that the visual signature by the search picture with it is described
Vision index is matched, including：
Calculate the matching degree that each vision in the visual signature of the search picture and vision index is indexed；
The vision rope of first threshold is not less than during the vision is indexed with the matching degree of the visual signature of the search picture
Draw indicated picture as with the search picture identical picture, by with the visual signature of the search picture
The first threshold being less than with degree and being not less than the picture indicated by the vision index of Second Threshold used as with the search
The similar picture of picture.
5. method according to claim 4, it is characterised in that the indicated picture of the vision index that will be matched is made
For Search Results are shown, including：
To be shown as Search Results to the search picture identical picture and/or similar picture.
6. method according to claim 1, it is characterised in that the visual signature include it is following at least one：Color, line
Reason, shape.
7. a kind of device for processing picture, it is characterised in that described device includes：
Acquiring unit, is configured to obtain picture and the text message related to the picture；
Text index generation unit, is configured to be analyzed the picture to obtain the first keyword set, to the text
This information is analyzed to obtain the second keyword set, according to first keyword set and second keyword set
Generate the text index for being indicated the picture；
Vision indexes generation unit, is configured to choose the vision for characterizing the picture from first keyword set
The keyword of feature, is indexed with the vision generated for being indicated the picture；
Memory cell, is configured to store the text index and vision index.
8. device according to claim 7, it is characterised in that described device also includes：
First processing units, are configured to the search picture in response to receiving user terminal transmission, then the search is used
Picture is analyzed to extract visual signature, the search is indexed with the visual signature of picture with the vision and is matched,
The picture indicated by vision index that will be matched is shown as Search Results.
9. device according to claim 7, it is characterised in that described device also includes：
Second processing unit, is configured to the search text in response to receiving user terminal transmission, then the search is used
Text is analyzed to generate the 3rd keyword set, and the 3rd keyword set is matched with the text index,
The picture indicated by text index that will be matched is shown as Search Results.
10. device according to claim 8, it is characterised in that the first processing units include：
Computation subunit, each for being configured to calculate in the visual signature of the search picture and vision index is regarded
Feel the matching degree of index；
Treatment subelement is not low with the matching degree of the visual signature of the search picture in being configured to index the vision
In the indicated picture of the vision index of first threshold as with search picture identical picture, will be used with the search
The matching degree of the visual signature of picture is less than the first threshold and is not less than the figure indicated by the vision index of Second Threshold
Piece as to the search similar picture of picture.
11. devices according to claim 10, it is characterised in that the first processing units include：
Displaying subelement, is configured to be tied as search to the search picture identical picture and/or similar picture
Fruit is shown.
12. devices according to claim 7, it is characterised in that the visual signature include it is following at least one：Color,
Texture, shape.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611042535.0A CN106708940B (en) | 2016-11-11 | 2016-11-11 | Method and device for processing pictures |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611042535.0A CN106708940B (en) | 2016-11-11 | 2016-11-11 | Method and device for processing pictures |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106708940A true CN106708940A (en) | 2017-05-24 |
| CN106708940B CN106708940B (en) | 2020-06-30 |
Family
ID=58933739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611042535.0A Active CN106708940B (en) | 2016-11-11 | 2016-11-11 | Method and device for processing pictures |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106708940B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451194A (en) * | 2017-06-30 | 2017-12-08 | 百度在线网络技术（北京）有限公司 | A kind of image searching method and device |
| CN107491477A (en) * | 2017-06-30 | 2017-12-19 | 百度在线网络技术（北京）有限公司 | A kind of emoticon searching method and device |
| CN107748754A (en) * | 2017-09-15 | 2018-03-02 | 广州唯品会研究院有限公司 | A kind of knowledge mapping improving method and device |
| CN108829784A (en) * | 2018-05-31 | 2018-11-16 | 百度在线网络技术（北京）有限公司 | Panorama recommended method, device, equipment and computer-readable medium |
| CN110309324A (en) * | 2018-03-09 | 2019-10-08 | 北京搜狗科技发展有限公司 | A kind of searching method and relevant apparatus |
| CN110381339A (en) * | 2019-08-07 | 2019-10-25 | 腾讯科技（深圳）有限公司 | Picture transmission method and device |
| WO2020108234A1 (en) * | 2018-11-30 | 2020-06-04 | Oppo广东移动通信有限公司 | Image index generation method, image search method and apparatus, and terminal, and medium |
| CN111797765A (en) * | 2020-07-03 | 2020-10-20 | 北京达佳互联信息技术有限公司 | Image processing method, image processing apparatus, server, and storage medium |
| CN112527210A (en) * | 2020-12-22 | 2021-03-19 | 南京中兴力维软件有限公司 | Storage method and device of full data and computer readable storage medium |
| CN113590862A (en) * | 2020-04-30 | 2021-11-02 | 北京搜狗科技发展有限公司 | Picture information processing method and device and electronic equipment |
| CN113779297A (en) * | 2021-09-01 | 2021-12-10 | 北京橙色云科技有限公司 | Information searching method and device based on picture and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101458695A (en) * | 2008-12-18 | 2009-06-17 | 西交利物浦大学 | Mixed picture index construct and enquiry method based on key word and content characteristic and use thereof |
| US20110167053A1 (en) * | 2006-06-28 | 2011-07-07 | Microsoft Corporation | Visual and multi-dimensional search |
| CN102567483A (en) * | 2011-12-20 | 2012-07-11 | 华中科技大学 | Multi-feature fusion human face image searching method and system |
| 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 |
-
2016
- 2016-11-11 CN CN201611042535.0A patent/CN106708940B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110167053A1 (en) * | 2006-06-28 | 2011-07-07 | Microsoft Corporation | Visual and multi-dimensional search |
| CN101458695A (en) * | 2008-12-18 | 2009-06-17 | 西交利物浦大学 | Mixed picture index construct and enquiry method based on key word and content characteristic and use thereof |
| CN102567483A (en) * | 2011-12-20 | 2012-07-11 | 华中科技大学 | Multi-feature fusion human face image searching method and system |
| 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 |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107491477B (en) * | 2017-06-30 | 2021-02-19 | 百度在线网络技术（北京）有限公司 | Emotion symbol searching method and device |
| CN107491477A (en) * | 2017-06-30 | 2017-12-19 | 百度在线网络技术（北京）有限公司 | A kind of emoticon searching method and device |
| CN107451194A (en) * | 2017-06-30 | 2017-12-08 | 百度在线网络技术（北京）有限公司 | A kind of image searching method and device |
| CN107748754A (en) * | 2017-09-15 | 2018-03-02 | 广州唯品会研究院有限公司 | A kind of knowledge mapping improving method and device |
| CN110309324A (en) * | 2018-03-09 | 2019-10-08 | 北京搜狗科技发展有限公司 | A kind of searching method and relevant apparatus |
| CN110309324B (en) * | 2018-03-09 | 2024-03-22 | 北京搜狗科技发展有限公司 | Searching method and related device |
| CN108829784A (en) * | 2018-05-31 | 2018-11-16 | 百度在线网络技术（北京）有限公司 | Panorama recommended method, device, equipment and computer-readable medium |
| WO2020108234A1 (en) * | 2018-11-30 | 2020-06-04 | Oppo广东移动通信有限公司 | Image index generation method, image search method and apparatus, and terminal, and medium |
| CN110381339A (en) * | 2019-08-07 | 2019-10-25 | 腾讯科技（深圳）有限公司 | Picture transmission method and device |
| CN113590862A (en) * | 2020-04-30 | 2021-11-02 | 北京搜狗科技发展有限公司 | Picture information processing method and device and electronic equipment |
| CN111797765A (en) * | 2020-07-03 | 2020-10-20 | 北京达佳互联信息技术有限公司 | Image processing method, image processing apparatus, server, and storage medium |
| CN111797765B (en) * | 2020-07-03 | 2024-04-16 | 北京达佳互联信息技术有限公司 | Image processing method, device, server and storage medium |
| CN112527210A (en) * | 2020-12-22 | 2021-03-19 | 南京中兴力维软件有限公司 | Storage method and device of full data and computer readable storage medium |
| CN113779297A (en) * | 2021-09-01 | 2021-12-10 | 北京橙色云科技有限公司 | Information searching method and device based on picture and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106708940B (en) | 2020-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106708940A (en) | Method and device used for processing pictures | |
| US20210232760A1 (en) | Optimization techniques for artificial intelligence | |
| CN108628971B (en) | Text classification method, text classifier and storage medium for unbalanced data set | |
| US20210081611A1 (en) | Methods and systems for language-agnostic machine learning in natural language processing using feature extraction | |
| CN106599054B (en) | Method and system for classifying and pushing questions | |
| US9323741B2 (en) | System and method for searching functions having symbols | |
| CN107220386A (en) | Information-pushing method and device | |
| CN108363790A (en) | For the method, apparatus, equipment and storage medium to being assessed | |
| CN112667794A (en) | Intelligent question-answer matching method and system based on twin network BERT model | |
| CN106651696B (en) | Approximate question pushing method and system | |
| CN106021362A (en) | Query picture characteristic representation generation method and device, and picture search method and device | |
| CN107748784B (en) | Method for realizing structured data search through natural language | |
| CN108090048B (en) | College evaluation system based on multivariate data analysis | |
| CN110750995B (en) | File management method based on custom map | |
| CN107305551A (en) | The method and apparatus of pushed information | |
| CN107491435A (en) | Method and device based on Computer Automatic Recognition user feeling | |
| CN110147425A (en) | A kind of keyword extracting method, device, computer equipment and storage medium | |
| CN108491388A (en) | Data set acquisition methods, sorting technique, device, equipment and storage medium | |
| CN103678460B (en) | For identifying the method and system for the non-text elements for being suitable to be communicated in multi-language environment | |
| CN114065750A (en) | Commodity information matching and publishing method and device, equipment, medium and product thereof | |
| CN107291774A (en) | Error sample recognition methods and device | |
| CN116109732A (en) | Image labeling method, device, processing equipment and storage medium | |
| CN113343012B (en) | News matching method, device, equipment and storage medium | |
| CN111160699A (en) | Expert recommendation method and system | |
| CN110347806A (en) | Original text discriminating method, device, equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |