CN108549708A - A kind of picture and text matching process and system - Google Patents
A kind of picture and text matching process and system Download PDFInfo
- Publication number
- CN108549708A CN108549708A CN201810354752.6A CN201810354752A CN108549708A CN 108549708 A CN108549708 A CN 108549708A CN 201810354752 A CN201810354752 A CN 201810354752A CN 108549708 A CN108549708 A CN 108549708A
- Authority
- CN
- China
- Prior art keywords
- picture
- keyword
- threshold value
- text
- word
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a kind of picture and text matching process and systems, judge whether text information matches with picture, export matching result；If it does, then display picture and word, if it does not match, monitoring flow consumption figures, judges whether the flow consumption figures is more than flow threshold, if it exceeds flow threshold, then do not show picture, only show text information.When the present invention can reduce browsing news or word to greatest extent, there is unnecessary pictorial information, helps user to save flow, simultaneously, also it helps user to put out unnecessary word because of the pictorial information unrelated to word, reduces the unnecessary flow waste that title party brings.
Description
Technical field
The present invention relates to computer realm, more particularly to a kind of picture and text matching process and system.
Background technology
In the prior art, with the development of science and technology, the mankind are increasing for the demand of data processing business, mobile phone is utilized
As information browse, the main channel of external information is connected as people, more and more people are not necessarily to pass through TV or newspaper
Solve News By Region, however, since Internet news brings quick real-time news, in addition it is various from media in order to obtain flow, lead to
Inflow-rate of water turbine weighs the value of its news, more and more news media when making news, by with the incoherent picture of word
Either title party or the picture by that can attract user attract user to click its news, can be obtained more with reaching
The purpose of flow, then this is unnecessary for the user of news browsing, and picture causes additional flow loss, and picture
Completely uncorrelated to word, this causes puzzlement to user to a certain extent.Therefore, in the prior art for the browsing of news
There is certain defect for a user.
Apply for content
To solve above-mentioned technical problem：The application proposes a kind of picture and text matching process, includes the following steps：
Judge whether text information matches with picture, exports matching result；
If it does, then display picture and word；If it does not match, monitoring flow consumption figures, judges that the flow disappears
Whether consumption value is more than flow threshold, if it exceeds flow threshold, then do not show picture, only show text information.
The picture and text matching process, it is described to judge whether text information matches with picture and specifically include：
Whether include " picture and text be not inconsistent " printed words, if including " picture and text are not inconsistent ", it is determined that picture and word if searching in word
Information mismatches；
Alternatively, searching pictorial information, picture keyword is exported；
The picture keyword and text information are compared, judges in text information whether to include the picture keyword；
If it is, output matching result, if it is not, then output mismatches.
The picture and text matching process, the lookup pictorial information, output picture keyword specifically include：
Picture is connected to network, picture searching is carried out, determines the keyword of picture, by list mode, output is a plurality of and schemes
The relevant keyword of piece.
The keyword of the picture and text matching process, the determining picture specifically includes：Detect the relevant title of picture or
Text information, compares multiple titles, and extraction frequency of occurrence is more than the word for searching threshold value, by more than the word for searching threshold value according to
The frequency is ranked sequentially, and rank results and keyword threshold comparison export the word more than the keyword threshold value.
The picture and text matching process, the lookup threshold value include first searching threshold value and second and searching threshold value, and described the
One, which searches threshold value, is more than the second lookup threshold value, when the frequency of word is more than the first lookup threshold value, determines that institute's predicate is to determine
Keyword, when the frequency of word between it is described first search threshold value and it is described second search threshold value between when, determine institute's predicate be can
Keyword is selected, when the frequency of word, which is less than described second, searches threshold value, determines that institute's predicate is fuzzy keyword；The keyword
Threshold value includes the first keyword threshold value and the second keyword threshold value, when the determining keyword is less than the first keyword threshold value
When, the difference of the first keyword threshold value and the determining keyword is selected from the optional keyword, according to described
Difference obtains the picture keyword of output with the determining keyword summation, when the determining keyword is more than first key
When word threshold value, the current picture keyword for determining keyword as output is exported, when the determining keyword is crucial more than second
When word threshold value, picture keyword of the keyword of the respective numbers of the second keyword threshold value as output is exported.
The picture and text matching process, it is described if it does, then display picture and word include：By picture into line label,
It is associated respectively with word for the picture of different labels, the picture that picture can not be associated with word is not shown then, figure
The picture that piece is associated with word then finds the corresponding label associated symbol of corresponding word setting picture.
The picture and text matching process only shows that text information further includes later if it does not match, not showing picture：
Again typesetting text information determines the author of text information, and the author is recorded in record sheet, by the record and early warning
Value is compared, if the record number of the author is more than early warning value, carries out early warning, and respective record early warning
Number pipes off corresponding author if early warning number surmounts preset value, is carried out at the same time list in blacklist
Report；If the author is first record, the author is inserted to a fixed position of record sheet.
The picture and text matching process, the output matching result include setting match grade, and the match grade is base
In word number of words and picture flow hierarchical relationship function, the hierarchical relationship function includes x word number of words block and y figure
Piece flow block, the x are proportionate with the y.
A kind of picture and text matching system, including：
Search module is used for search pictures keyword and text information；
Matching judgment module exports matching result for judging whether text information matches with picture；
Output module, for according to matching result output as a result, if it does, then display picture and word, if not
Match, then monitor flow consumption figures, judges whether the flow consumption figures is more than flow threshold, if it exceeds flow threshold, then not
It shows picture, only shows text information
The picture and text matching system, further include threshold setting module, networking module, sorting module, text composition module,
Author information logging modle and author information warning module, the threshold setting module search threshold value and keyword threshold for setting
Value, the threshold setting module are additionally operable to flow threshold when setting picture and text mismatch, and the lookup threshold value includes the first lookup
Threshold value and second searches threshold value, and the keyword threshold value includes the first keyword threshold value and the second keyword threshold value；The networking
Module carries out keyword search for connecting network, to picture；The sorting module be used for for search picture keyword into
Row sequence；The text composition module is used to be directed to text information progress, and typesetting and picture and text are associated operation again；It is described
For author information logging modle for recording the unmatched author information of picture and text, the author information warning module is used for will
The unmatched author's record of picture and text carries out early warning more than the author of early warning value；
The output matching result includes setting match grade, and the match grade is based on word number of words and picture flow
Hierarchical relationship function, the hierarchical relationship function include x word number of words block and y picture stream gauge block, the x and y
It is proportionate.
The application can automatic identification picture whether matched with word, to greatest extent save user browse news when stream
Amount reduces the puzzlement that unnecessary pictorial information is brought to user, promotes user and reads news or browse the experience of text information
Effect.
Description of the drawings
Fig. 1 is the picture and text matching process schematic diagram of the present invention.
Fig. 2 is the picture and text matching system schematic diagram of the present invention.
Specific implementation mode
The application is described in further detail below in conjunction with the accompanying drawings, it is necessary to it is indicated herein to be, implement in detail below
Mode is served only for that the application is further detailed, and should not be understood as the limitation to the application protection domain, the field
Technical staff can make some nonessential modifications and adaptations according to above-mentioned application content to the application.
Embodiment one：
As shown in Figure 1, proposing a kind of picture and text matching process for the present invention, include the following steps：
Judge whether text information matches with picture, exports matching result；
If it does, then display picture and word；If it does not match, monitoring flow consumption figures, judges that the flow disappears
Whether consumption value is more than flow threshold, if it exceeds flow threshold, then do not show picture, only show text information.
The picture and text matching process, it is described to judge whether text information matches with picture and specifically include：
Whether include " picture and text be not inconsistent " printed words, if including " picture and text are not inconsistent ", it is determined that picture and word if searching in word
Information mismatches；
Alternatively, searching pictorial information, picture keyword is exported；
The picture keyword and text information are compared, judges in text information whether to include the picture keyword；
If it is, output matching result, if it is not, then output mismatches.
The picture and text matching process, the lookup pictorial information, output picture keyword specifically include：
Picture is connected to network, picture searching is carried out, determines the keyword of picture, by list mode, output is a plurality of and schemes
The relevant keyword of piece.
The keyword of the picture and text matching process, the determining picture specifically includes：Detect the relevant title of picture or
Text information, compares multiple titles, and extraction frequency of occurrence is more than the word for searching threshold value, by more than the word for searching threshold value according to
The frequency is ranked sequentially, and rank results and keyword threshold comparison export the word more than the keyword threshold value.
The picture and text matching process, the lookup threshold value include first searching threshold value and second and searching threshold value, and described the
One, which searches threshold value, is more than the second lookup threshold value, when the frequency of word is more than the first lookup threshold value, determines that institute's predicate is to determine
Keyword, when the frequency of word between it is described first search threshold value and it is described second search threshold value between when, determine institute's predicate be can
Keyword is selected, when the frequency of word, which is less than described second, searches threshold value, determines that institute's predicate is fuzzy keyword；The keyword
Threshold value includes the first keyword threshold value and the second keyword threshold value, when the determining keyword is less than the first keyword threshold value
When, the difference of the first keyword threshold value and the determining keyword is selected from the optional keyword, according to described
Difference obtains the picture keyword of output with the determining keyword summation, when the determining keyword is more than first key
When word threshold value, the current picture keyword for determining keyword as output is exported, when the determining keyword is crucial more than second
When word threshold value, picture keyword of the keyword of the respective numbers of the second keyword threshold value as output is exported.
The picture and text matching process, it is described if it does, then display picture and word include：By picture into line label,
It is associated respectively with word for the picture of different labels, the picture that picture can not be associated with word is not shown then, figure
The picture that piece is associated with word then finds the corresponding label associated symbol of corresponding word setting picture.
The picture and text matching process only shows that text information further includes later if it does not match, not showing picture：
Again typesetting text information determines the author of text information, and the author is recorded in record sheet, by the record and early warning
Value is compared, if the record number of the author is more than early warning value, carries out early warning, and respective record early warning
Number pipes off corresponding author if early warning number surmounts preset value, is carried out at the same time list in blacklist
Report；If the author is first record, the author is inserted to a fixed position of record sheet.
The picture and text matching process, the output matching result include setting match grade, and the match grade is base
In word number of words and picture flow hierarchical relationship function, the hierarchical relationship function includes x word number of words block and y figure
Piece flow block, the x are proportionate with the y.
As value is carried out according to the following table in x, y：X1, x2, x3 are respectively exemplary different word number of words, the corresponding difference x of y
When being worth selected picture and text mismatch, the threshold value of flow consumption.The specific choice of the y values can be according to the terminal of user monthly
Flow package sets itself, following table are only schematic.
Embodiment two：
It proposes that a kind of picture and text mismatch flow monitoring method for the present invention, includes the following steps：
Judge whether text information matches with picture, exports matching result；
If it does, then display picture and word；If it does not match, monitoring flow consumption figures, judges whether to be more than stream
Threshold value is measured, picture is not shown then more than flow threshold, only shows text information.
It is described monitoring flow consumption figures specific method include：
Step 1：Picture and text are mismatched in a monitoring cycle and carry out flow monitoring, form Flux Loss waveform；Such as choosing
Take either half a day or fix some period as monitoring cycle within one day；
Step 2：The Flux Loss waveform is analyzed, in the monitoring cycle, to obtain the amplitude xi of the wave, wherein
I is 1 natural number for arriving n, and n is the number of wave in one monitoring cycle；
Step 3：Recursive calculation is carried out by following formula (1), finds out the mean value computation value Yn of the amplitude xi of the wave：Yn
=(1-k) Yn-1+kxn (1)
In formula (1), xn is the amplitude of the wave measured at the n-th moment；When Yn is n-th recurrence, the mean value computation of amplitude xi
Value；K is computational constant；
Step 4：Recursive calculation is carried out again by formula (2), finds out the calculated value of the mean-square value Zn of amplitude xi；
Zn=(1-k) Zn-1+kxn2 (2)
When Zn is n-th recurrence, the calculated value of the mean-square value of amplitude xi；
Step 5：The standard value standard (xi) of the calculated value of xi is sought by formula (3)；
Standard (x)=Yn2-Zn (3)；
Step 6：Again by formula (4), binary judgement is carried out to xi；
S=xi-Yn*standard (xi) (4)
Step 7：Think that the flow for occurring to be lost when picture and text mismatch is excessive when S is more than zero, that is, is more than threshold value.
The picture and text matching process, it is described to specifically include：It is described to judge whether text information is specific to matching with picture
Including：
Whether include " picture and text be not inconsistent " printed words, if including " picture and text are not inconsistent ", it is determined that picture and word if searching in word
Information mismatches；
The picture and text matching process, it is described to specifically include：It is described to judge whether text information is specific to matching with picture
Including：
Pictorial information is searched, picture keyword is exported；
The picture keyword and text information are compared, judges in text information whether to include the picture keyword；
If it is, output matching result, if it is not, then output mismatches.
The picture and text matching process, the lookup pictorial information, output picture keyword specifically include：
Picture is connected to network, picture searching is carried out, determines the keyword of picture, by list mode, output is a plurality of and schemes
The relevant keyword of piece.
The keyword of the picture and text matching process, the determining picture specifically includes：Detect the relevant title of picture or
Text information, compares multiple titles, and extraction frequency of occurrence is more than the word for searching threshold value, by more than the word for searching threshold value according to
The frequency is ranked sequentially, and rank results and keyword threshold comparison export the word more than the keyword threshold value.
The picture and text matching process, the lookup threshold value include first searching threshold value and second and searching threshold value, and described the
One, which searches threshold value, is more than the second lookup threshold value, when the frequency of word is more than the first lookup threshold value, determines that institute's predicate is to determine
Keyword, when the frequency of word between it is described first search threshold value and it is described second search threshold value between when, determine institute's predicate be can
Keyword is selected, when the frequency of word, which is less than described second, searches threshold value, determines that institute's predicate is fuzzy keyword；The keyword
Threshold value includes the first keyword threshold value and the second keyword threshold value, when the determining keyword is less than the first keyword threshold value
When, the difference of the first keyword threshold value and the determining keyword is selected from the optional keyword, according to described
Difference obtains the picture keyword of output with the determining keyword summation, when the determining keyword is more than first key
When word threshold value, the current picture keyword for determining keyword as output is exported, when the determining keyword is crucial more than second
When word threshold value, picture keyword of the keyword of the respective numbers of the second keyword threshold value as output is exported.
The picture and text matching process, it is described if it does, then display picture and word include：By picture into line label,
Picture distribution for different labels is associated with word, and the picture that picture can not be associated with word is not shown then, figure
The picture that piece is associated with word then finds the corresponding label associated symbol of corresponding word setting picture.
The picture and text matching process only shows that text information further includes later if it does not match, not showing picture：
Again typesetting text information determines the author of text information, and the author is recorded in record sheet, by the record and early warning
Value is compared, if the record number of the author is more than early warning value, carries out early warning, and respective record early warning
Number pipes off corresponding author if early warning number surmounts preset value, is carried out at the same time list in blacklist
Report；If the author is first record, the author is inserted to a fixed position of record sheet.
Embodiment three：
As shown in Fig. 2, for a kind of picture and text matching system of the present invention, a kind of picture and text matching system, including：
Search module is used for search pictures keyword and text information；
Matching judgment module exports matching result for judging whether text information matches with picture；
Output module, for according to matching result output as a result, if it does, then display picture and word, if not
Match, then monitor flow consumption figures, judges whether the flow consumption figures is more than flow threshold, if it exceeds flow threshold, then not
It shows picture, only shows text information
The picture and text matching system, further include threshold setting module, networking module, sorting module, text composition module,
Author information logging modle and author information warning module, the threshold setting module search threshold value and keyword threshold for setting
Value, the threshold setting module are additionally operable to flow threshold when setting picture and text mismatch, and the lookup threshold value includes the first lookup
Threshold value and second searches threshold value, and the keyword threshold value includes the first keyword threshold value and the second keyword threshold value；The networking
Module carries out keyword search for connecting network, to picture；The sorting module be used for for search picture keyword into
Row sequence；The text composition module is used to be directed to text information progress, and typesetting and picture and text are associated operation again；It is described
For author information logging modle for recording the unmatched author information of picture and text, the author information warning module is used for will
The unmatched author's record of picture and text carries out early warning more than the author of early warning value；
The output matching result includes setting match grade, and the match grade is based on word number of words and picture flow
Hierarchical relationship function, the hierarchical relationship function include x word number of words block and y picture stream gauge block, the x and y
It is proportionate.
As value is carried out according to the following table in x, y：X1, x2, x3 are respectively exemplary different word number of words, the corresponding difference x of y
When being worth selected picture and text mismatch, the threshold value of flow consumption.The specific choice of the y values can be according to the terminal of user monthly
Flow package sets itself, following table are only schematic.
The application can automatic identification picture whether matched with word, to greatest extent save user browse news when stream
Amount reduces the puzzlement that unnecessary pictorial information is brought to user, promotes user and reads news or browse the experience of text information
Effect, meanwhile, the waste of inessential flow, the figure that user sets according to self-demand can be also carried out according to user's given threshold
The flow that text can waste when mismatching, realizes the picture and text matched design of differentiation.
Claims (10)
1. a kind of picture and text matching process, which is characterized in that include the following steps：
Judge whether text information matches with picture, exports matching result；
If it does, then display picture and word；If it does not match, monitoring flow consumption figures, judges the flow consumption figures
Whether it is more than flow threshold, if it exceeds flow threshold, then do not show picture, only show text information.
2. picture and text matching process as described in claim 1, which is characterized in that it is described judge text information whether with picture phase
With specifically including：
Whether include " picture and text be not inconsistent " printed words, if including " picture and text are not inconsistent ", it is determined that picture and text information if searching in word
It mismatches；
Alternatively, searching pictorial information, picture keyword is exported；
The picture keyword and text information are compared, judges in text information whether to include the picture keyword；
If it is, output matching result, if it is not, then output mismatches.
3. picture and text matching process as claimed in claim 2, which is characterized in that the lookup pictorial information, output picture are crucial
Word specifically includes：
It connects picture and arrives network, carry out picture searching, determine the keyword of picture, by list mode, export a plurality of with picture phase
The keyword of pass.
4. picture and text matching process as claimed in claim 3, which is characterized in that the keyword of the determining picture specifically includes：
The relevant title of picture or text information are detected, compares multiple titles, extraction frequency of occurrence is more than the word for searching threshold value, will be more than
The word for searching threshold value is ranked sequentially according to the frequency, will be more than the keyword by rank results and keyword threshold comparison
The word of threshold value exports.
5. picture and text matching process as claimed in claim 4, which is characterized in that the lookup threshold value include first search threshold value and
Second searches threshold value, and described first, which searches threshold value, is more than the second lookup threshold value, when the frequency of word is more than the first lookup threshold value
When, determine institute's predicate be determine keyword, when word the frequency between it is described first search threshold value and it is described second search threshold value it
Between when, determine institute's predicate be may be selected keyword, when the frequency of word be less than it is described second search threshold value when, determine institute's predicate be mould
Paste keyword；The keyword threshold value includes the first keyword threshold value and the second keyword threshold value, when the determining keyword is small
When the first keyword threshold value, the first keyword threshold value and the determining pass are selected from the optional keyword
The difference of keyword obtains the picture keyword of output according to the difference, when the determining pass with the determining keyword summation
When keyword is more than the first keyword threshold value, the current picture keyword for determining keyword as output is exported, when described true
When determining keyword more than the second keyword threshold value, figure of the keyword of the respective numbers of the second keyword threshold value as output is exported
Piece keyword.
6. picture and text matching process as described in claim 1, which is characterized in that described if it does, then display picture and word
Including：By picture into line label, it is associated respectively with word for the picture of different labels, picture can not be closed with word
The picture of connection does not show that the picture that picture is associated with word then finds the corresponding label of corresponding word setting picture and closes then
Join symbol.
7. picture and text matching process as described in claim 1, which is characterized in that if it does not match, not showing picture, only show
Further include after text information：Again typesetting text information determines the author of text information, and record sheet is recorded in the author
In, the record is compared with early warning value, if the record number of the author is more than early warning value, early warning is carried out and carries
Show, and respective record early warning number, if early warning number surmounts preset value, corresponding author is included in black name
It is single, it is carried out at the same time list in blacklist and reports；If the author is first record, the author is inserted the one of record sheet
A fixed position.
8. picture and text matching process as described in claim 1, which is characterized in that the output matching result includes setting matching etc.
Grade, the match grade are based on word number of words and picture flow hierarchical relationship function, and the hierarchical relationship function includes x text
Word number of words block and y picture stream gauge block, the x are proportionate with the y.
9. a kind of picture and text matching system, which is characterized in that including：
Search module is used for search pictures keyword and text information；
Matching judgment module exports matching result for judging whether text information matches with picture；
Output module, for according to matching result output as a result, if it does, then display picture and word, if it does not match,
Flow consumption figures is monitored, judges whether the flow consumption figures is more than flow threshold, if it exceeds flow threshold, then do not show figure
Piece only shows text information.
10. picture and text matching system as claimed in claim 9, which is characterized in that further include threshold setting module, networking module,
Sorting module, text composition module, author information logging modle and author information warning module, the threshold setting module are used for
Threshold value and keyword threshold value are searched in setting, and the threshold setting module is additionally operable to flow threshold when setting picture and text mismatch, institute
It includes first searching threshold value and second and searching threshold value to state and search threshold value, and the keyword threshold value includes the first keyword threshold value and the
Two keyword threshold values；The networking module carries out keyword search for connecting network, to picture；The sorting module is used for needle
The picture keyword of search is ranked up；The text composition module is used to carry out typesetting and figure again for text information
Text is associated operation；The author information logging modle is for recording the unmatched author information of picture and text, the work
Person's information early warning module is used to the unmatched author's record of picture and text carrying out early warning more than the author of early warning value；
The output matching result includes setting match grade, and the match grade is based on word number of words and picture flow grade
Relation function, the hierarchical relationship function include x word number of words block and y picture stream gauge block, and the x and y is in just
It is related.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810354752.6A CN108549708B (en) | 2018-04-19 | 2018-04-19 | Image-text matching method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810354752.6A CN108549708B (en) | 2018-04-19 | 2018-04-19 | Image-text matching method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108549708A true CN108549708A (en) | 2018-09-18 |
| CN108549708B CN108549708B (en) | 2021-08-31 |
Family
ID=63515661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810354752.6A Active CN108549708B (en) | 2018-04-19 | 2018-04-19 | Image-text matching method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108549708B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109543058A (en) * | 2018-11-23 | 2019-03-29 | 连尚（新昌）网络科技有限公司 | For the method for detection image, electronic equipment and computer-readable medium |
| CN111666969A (en) * | 2020-04-22 | 2020-09-15 | 北京百度网讯科技有限公司 | Method and device for calculating image-text similarity, electronic equipment and readable storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040236565A1 (en) * | 2003-04-22 | 2004-11-25 | Saying Wen | System and method for playing vocabulary explanations using multimedia data |
| CN104331490A (en) * | 2014-11-14 | 2015-02-04 | 北京国双科技有限公司 | Network data processing method and device |
| CN106412213A (en) * | 2015-07-27 | 2017-02-15 | 中兴通讯股份有限公司 | Contact person information processing method and device and mobile terminal |
| CN106991150A (en) * | 2017-03-28 | 2017-07-28 | 维沃移动通信有限公司 | The display methods and mobile terminal of a kind of web data |
| CN107491432A (en) * | 2017-06-20 | 2017-12-19 | 北京百度网讯科技有限公司 | Low quality article recognition methods and device, equipment and medium based on artificial intelligence |
| CN107679063A (en) * | 2017-07-31 | 2018-02-09 | 平安科技（深圳）有限公司 | Image display method, server and computer-readable recording medium |
-
2018
- 2018-04-19 CN CN201810354752.6A patent/CN108549708B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040236565A1 (en) * | 2003-04-22 | 2004-11-25 | Saying Wen | System and method for playing vocabulary explanations using multimedia data |
| CN104331490A (en) * | 2014-11-14 | 2015-02-04 | 北京国双科技有限公司 | Network data processing method and device |
| CN106412213A (en) * | 2015-07-27 | 2017-02-15 | 中兴通讯股份有限公司 | Contact person information processing method and device and mobile terminal |
| CN106991150A (en) * | 2017-03-28 | 2017-07-28 | 维沃移动通信有限公司 | The display methods and mobile terminal of a kind of web data |
| CN107491432A (en) * | 2017-06-20 | 2017-12-19 | 北京百度网讯科技有限公司 | Low quality article recognition methods and device, equipment and medium based on artificial intelligence |
| CN107679063A (en) * | 2017-07-31 | 2018-02-09 | 平安科技（深圳）有限公司 | Image display method, server and computer-readable recording medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109543058A (en) * | 2018-11-23 | 2019-03-29 | 连尚（新昌）网络科技有限公司 | For the method for detection image, electronic equipment and computer-readable medium |
| CN111666969A (en) * | 2020-04-22 | 2020-09-15 | 北京百度网讯科技有限公司 | Method and device for calculating image-text similarity, electronic equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108549708B (en) | 2021-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103942337B (en) | It is a kind of based on image recognition and the video searching system that matches | |
| CN108874777B (en) | Text anti-spam method and device | |
| CN104516986B (en) | A kind of sentence recognition methods and device | |
| TWI653542B (en) | Method, system and device for discovering and tracking hot topics based on network media data flow | |
| CN105320778B (en) | A method of suitable for e-commerce Chinese website Commercial goods labels | |
| CN103902538B (en) | Information recommending apparatus and method based on decision tree | |
| CN111754302A (en) | Video live broadcast interface commodity display intelligent management system based on big data | |
| CN108763321A (en) | A kind of related entities recommendation method based on extensive related entities network | |
| CN108549708A (en) | A kind of picture and text matching process and system | |
| CN110597987A (en) | Search recommendation method and device | |
| CN102355517A (en) | Information classification apparatus, information classification method and terminal | |
| CN105989125B (en) | The searching method and system of tag recognition are carried out to no result word | |
| CN107315823A (en) | Data processing method and device based on ecommerce | |
| CN114186567A (en) | Sensitive word detection method and device, equipment, medium and product thereof | |
| CN110232133A (en) | A kind of image of clothing search method and system classified based on Fusion Features and style | |
| Rozen et al. | Answering product-questions by utilizing questions from other contextually similar products | |
| CN111783456A (en) | Public opinion analysis method using semantic analysis technology | |
| CN107944030A (en) | A kind of library automatic classification device | |
| CN110941757A (en) | Big data based policy information query pushing system and method | |
| CN108563765A (en) | A kind of intelligence picture and text matching process and system | |
| CN110688572A (en) | Method for identifying search intention in cold starting state | |
| CN106383917A (en) | Data processing method based on user logs | |
| CN116309963B (en) | Batch labeling method and device for images, electronic equipment and storage medium | |
| CN112733006B (en) | User portrait generation method, device and equipment and storage medium | |
| CN112784600B (en) | Information ordering method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210803 Address after: 252000 workshops 1, 2, 3 and 4 in the courtyard of Shandong Zhonglian rongchuang Food Co., Ltd., South Ring auxiliary road, Jiuzhou sub district office, high tech Zone, Liaocheng City, Shandong Province Applicant after: Liaocheng Zhongxin Intelligent Technology Co.,Ltd. Address before: 6288, 6th floor, building 2, No. 2, No. 117, Zhangcha 1st Road, Chancheng District, Foshan City, Guangdong Province Applicant before: FOSHAN LONGSHENG GUANGQI TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |