CN106776710A - A kind of picture and text construction of knowledge base method based on vertical search engine - Google Patents
A kind of picture and text construction of knowledge base method based on vertical search engine Download PDFInfo
- Publication number
- CN106776710A CN106776710A CN201611014342.4A CN201611014342A CN106776710A CN 106776710 A CN106776710 A CN 106776710A CN 201611014342 A CN201611014342 A CN 201611014342A CN 106776710 A CN106776710 A CN 106776710A
- Authority
- CN
- China
- Prior art keywords
- image
- text
- webpage
- picture
- retrieval
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of picture and text construction of knowledge base method based on vertical search engine, and the method includes:First in the webpage of server background timing orientation crawl encyclopaedia class website, then the network encyclopaedic knowledge content of crawl is analyzed, graph text information in webpage is extracted respectively, then textural characteristics, color characteristic, shape facility of image etc. are extracted and carrys out pie graph as weighted feature, set up according to given keyword title on this basis and indexed, the word description and image and characteristics of image in webpage are associated storage;The retrieval mode based on text and image is finally provided the user, retrieval result includes image and word description.The present invention provides a kind of picture and text knowledge base for building and having picture and text query function, realizes that the quick of knowledge is shared.
Description
Technical field
The present invention relates to picture and text construction of knowledge base, especially a kind of picture and text construction of knowledge base side based on vertical search engine
Method.
Technical background
With the rapid growth of mobile Internet, by the demand rapid growth of intelligent terminal search inquiry encyclopaedic knowledge.
At present, encyclopaedic knowledge is distributed in the professional website of many correlations, such as wikipedia, Baidupedia, interactive encyclopaedia.Can be using logical
The knowledge of correlation is got with search engine.
The usual scale of construction of these knowledge bases is huge, and both pictures and texts are excellent, and same knowledge is often distributed in different web sites, it is difficult to once
Get to property the comprehensive information of such knowledge.Meanwhile, these encyclopaedic knowledge websites are provided solely for classified inquiry and text retrieval
Function.Therefore, being built using vertical search engine can provide the picture and text knowledge base of picture and text query function, realize picture and text knowledge
Quick share.
The content of the invention
For deficiency of the prior art, the present invention provides a kind of picture and text construction of knowledge base based on vertical search engine
Method.The webpage of the requirement timing orientation crawl encyclopaedia class website according to theme in the server, and according to subject key words come
Content in analysis webpage, realizes the association of Engineering drawing, on this basis to its Texture eigenvalue of image zooming-out therein, structure
Build the database according to keyword ranking and image characteristics index.Text and image retrieval can be provided in later retrieval.Should
Method improves the professional of knowledge base, solves the problems, such as that knowledge base can not be provided based on image retrieval.
According to design provided by the present invention, a kind of side of the picture and text construction of knowledge base based on vertical search engine
Method, specifically comprises the steps of:
Webpage of the step 1. in server background timing orientation crawl encyclopaedia class website.
The network encyclopaedic knowledge content of step 2. analysis crawl, Engineering drawing web page contents.
Step 3. extracts textural characteristics, color characteristic, shape facility of image etc., builds image weighted feature.
Step 4. is set up according to given keyword title and is indexed, and the word description and image and characteristics of image in webpage are entered
Row associated storage.
Step 5, provides text based retrieval and Image Feature Matching retrieval feedback during retrieval, feedback information includes image
And word description.
Further, in step 1, the webpage of the server background timing orientation crawl encyclopaedia class website, specially:First
The letter as knowledge base including the multiple encyclopaedic knowledge websites including wikipedia Chinese network, Baidupedia, interactive encyclopaedia is set
Source is ceased, then the taxonomic structure to these websites is analyzed, and according to set theme, selects the webpage of the subject categories
Crawl is timed, to obtain the URL of related subject webpage.
Further, in step 2, the network encyclopaedic knowledge content that described analysis is captured, Engineering drawing web page contents, specifically
It is that the text message in webpage is extracted into storage, the pictures/images in webpage is preserved.
Further, in step 3, described textural characteristics, color characteristic, shape facility of extraction image etc., build figure
Picture weighted feature, specially:Image is carried out into size normalization, the textural characteristics T, color characteristic C of image are then extracted respectively
With shape facility S, and weight w is assigned respectively to these three features1,w2,w3, and w1+w2+w3=1, the weighted feature F=of image
w1T+w2C+w3S。
Preferably, described texture feature extraction, the textural characteristics of image are extracted using Gabor filter;
Preferably, the color characteristic of described image, using the color histogram feature of RGB color;
Preferably, described shape facility, describes son to extract shape facility using Fourier.
Further, in step 4, described foundation according to given keyword title indexes, by the word description in webpage
Storage is associated with image and characteristics of image, specially:In database, using keyword as database major key, image
Feature is stored text description and image object as auxiliary keyword.
Preferentially, described image object is stored, using during image object individually to store specified classified catalogue,
Store path+the image name of storage image object in database.
Further, in steps of 5, text based retrieval and Image Feature Matching retrieval feedback are provided during the retrieval,
Feedback information includes image and word description, specially:Situation according to query object is divided into three types.The first, in number
Can be searched in database by being input into text to be retrieved according in storehouse, the result that feedback searching is arrived, and the text
Corresponding image;Secondth, image to be checked is input into, the characteristic value of query image is calculated using the method in step 3, then
Compared using Euclidean distance with the image feature value stored in database, according to difference feedback image and its correspondence from small to large
Text;The third, input be the image feature value for having calculated, then directly will inquire about characteristic value and database in
The image feature value of storage is compared using Euclidean distance, according to difference feedback image and its corresponding text from small to large.
Beneficial effects of the present invention:The present invention lacks unified effective management for the dispersion of existing encyclopaedic knowledge content,
The related content analyzed and extract in typical encyclopaedic knowledge is screened according to the knowledge dictionary specified using vertical search engine,
The graph text information of these contents is analyzed and characteristics of image is extracted and text description, further picture and text knowledge is built
Storehouse, the knowledge base of structure is provided and is based on text retrieval, image retrieval function, there is provided a kind of structure of new picture and text knowledge base
Build and application mode.
Brief description of the drawings
Fig. 1 schematic flow sheets of the invention.
Fig. 2 flow charts provided in an embodiment of the present invention.
Specific embodiment
In order that the purpose of the present invention, technical scheme are advantage become more apparent, it is right below in conjunction with drawings and Examples
The present invention is further described.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and without
It is of the invention in limiting.
Embodiment one, with reference to shown in Fig. 1, a kind of picture and text construction of knowledge base method based on vertical search engine, its feature
It is, including:
Webpage of the step 101. in server background timing orientation crawl encyclopaedia class website.
The network encyclopaedic knowledge content of step 102. analysis crawl, Engineering drawing web page contents.
Step 103. extracts textural characteristics, color characteristic, shape facility of image etc., builds image weighted feature.
Step 104. sets up index according to the knowledge library name for building, and the word description and image in webpage and image is special
Levy and be associated storage.
Text based retrieval and Image Feature Matching retrieval feedback are provided when step 105. is retrieved, feedback information includes figure
Picture and word description.
The webpage of the server background timing orientation crawl encyclopaedia class website, specially:Setting first includes Wiki hundred
Section's Chinese network, Baidupedia, interactive encyclopaedia interior multiple encyclopaedic knowledge websites as knowledge base information source, then to this
The taxonomic structure of a little websites is analyzed, and according to set theme, selects the webpage of the subject categories to be timed crawl, with
Obtain the URL of related subject webpage.
The network encyclopaedic knowledge content of described analysis crawl, Engineering drawing web page contents, specifically by the text in webpage
This information extraction is stored, and the pictures/images in webpage are preserved.
Described textural characteristics, color characteristic, shape facility of extraction image etc., build image weighted feature, specially:
Image is carried out into size normalization, the textural characteristics T, color characteristic C and shape facility S of image are then extracted respectively, and to this
Three kinds of features assign weight w respectively1,w2,w3, and w1+w2+w3=1, the weighted feature F=w of image1T+w2C+w3S。
Described being set up according to given keyword title is indexed, and the word description and image and characteristics of image in webpage are entered
Row associated storage, specially:In database, using keyword as database major key, characteristics of image, will used as auxiliary keyword
Text is described and image object is stored.
Text based retrieval and Image Feature Matching retrieval feedback be provided during the retrieval, feedback information comprising image and
Word description, specially:Situation according to query object is divided into three types.The first, in database can by input
Text to be retrieved is searched in database, the result that feedback searching is arrived, and the corresponding image of the text;Secondth,
Input image to be checked, the characteristic value of query image is calculated using the method in step 103, then with storage in database
Image feature value is compared using Euclidean distance, according to difference feedback image and its corresponding text from small to large;The third, it is defeated
What is entered is the image feature value for having calculated, then the image feature value stored in characteristic value and database that will directly inquire about
Compared using Euclidean distance, according to difference feedback image and its corresponding text from small to large.
Embodiment two:With reference to shown in Fig. 2, a kind of method of the picture and text construction of knowledge base based on vertical search engine, it is special
Levy and be, including:
In step 201, botanical name, animal name noun dictionary are built, according to the knowledge base to be built, provide knowledge
The classification in storehouse and its corresponding noun dictionary, for example, build flowers dictionary, then list corresponding flowers title.
In step 202., build beam search search website database, setting include wikipedia Chinese network,
Baidupedia, interactive encyclopaedia interior multiple encyclopaedic knowledge websites as knowledge base information source, then to these websites
Taxonomic structure is analyzed, and according to set theme, selects the webpage of the subject categories.
In step 203, directed information crawl will set the information scratching of webpage and preserves by vertical search engine.
In step 204, web page analysis, the webpage that will be grabbed first determines whether whether content therein has captured simultaneously
Stored, if it is not, webpage Chinese version information is extracted, by image zooming-out out.
In step 205, text analyzing and index are set up, and will analyze text subject information, and according to knowledge base noun word
The primary keyword of database is set up in storehouse.
In step 206, image characteristic analysis are extracted, using the textural characteristics, color characteristic, the shape facility that extract image
Deng, image weighted feature is built, image is carried out into size normalization, the textural characteristics T of image, color characteristic are then extracted respectively
C and shape facility S, and weight w is assigned respectively to these three features1,w2,w3, and w1+w2+w3=1, the weighted feature F=of image
w1T+w2C+w3S。
In step 207, picture and text associated storage, the characteristics of image that will be calculated in step 206 as auxiliary keyword with
Noun dictionary is associated in step 205, by the storage such as noun dictionary, text message and image information, characteristics of image to data
In storehouse.
In a step 208, user's inquiry, retrieval, provide the user the side inquired about based on text, image and characteristics of image
Formula, and corresponding text and image information are returned into user..
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
The present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those within the art that:It still may be used
Modified with to the technical scheme described in foregoing embodiments, or equivalent is carried out to which part technical characteristic;
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (7)
1. a kind of picture and text construction of knowledge base method based on vertical search engine, it is characterised in that including:First after server
The webpage of platform timing orientation crawl encyclopaedia class website, then analyzes the network encyclopaedic knowledge content of crawl, by the picture and text in webpage
Information is extracted respectively, and textural characteristics, color characteristic, the shape facility etc. for then extracting image carry out pie graph as weighted feature,
Set up according to given keyword title on the basis of this and indexed, the word description and image and characteristics of image in webpage are associated
Storage;Text based retrieval and Image Feature Matching retrieval feedback are provided when end user is retrieved, feedback information includes image
And word description.
2. described construction method is required according to right 1, described server background timing orientation captures the net of encyclopaedia class website
Page, specially:Set first including the multiple encyclopaedic knowledge websites including wikipedia Chinese network, Baidupedia, interactive encyclopaedia
Used as the information source of knowledge base, then the taxonomic structure to these websites is analyzed, and according to set theme, selection should
The webpage of subject categories is timed crawl, to obtain the URL of related subject webpage.
3. described construction method, the network encyclopaedic knowledge content that described analysis is captured, Engineering drawing net are required according to right 1
Page content, specifically extracts storage by the text message in webpage, and the pictures/images in webpage are preserved.
4. described construction method, the described textural characteristics, color characteristic, the shape facility that extract image are required according to right 1
Deng, structure image weighted feature, specially:Image is carried out into size normalization, the textural characteristics T of image is then extracted respectively,
Color characteristic C and shape facility S, and weight w is assigned respectively to these three features1,w2,w3, and w1+w2+w3=1, image plus
Power feature F=w1T+w2C+w3S。
5. described construction method is required according to right 1, described being set up according to given keyword title is indexed, by webpage
Word description and image and characteristics of image are associated storage, specially:In database, using keyword as database master
Key, characteristics of image is stored text description and image object as auxiliary keyword.
6. described construction method is required according to right 1, text based retrieval and Image Feature Matching are provided during the retrieval
Retrieval feedback, feedback information includes image and word description, specially:Situation according to query object is divided into three types.The
It is a kind of, can be searched in database by being input into text to be retrieved in database, the result that feedback searching is arrived, with
And the corresponding image of the text;Secondth, image to be checked is input into, query image is calculated using characteristics of image method of weighting
Characteristic value, is then compared with the image feature value stored in database using Euclidean distance, is fed back from small to large according to difference
Image and its corresponding text;The third, input be the image feature value for having calculated, then directly will inquire about feature
Value is compared with the image feature value stored in database using Euclidean distance, according to difference feedback image and its right from small to large
The text answered.
7. construction method according to claim 1, it is characterised in that the structure of picture and text knowledge base includes following steps:
Webpage of the step 101. in server background timing orientation crawl encyclopaedia class website;
The network encyclopaedic knowledge content of step 102. analysis crawl, Engineering drawing web page contents;
Step 103. extracts textural characteristics, color characteristic, shape facility of image etc., builds image weighted feature;
Step 104. is set up according to given keyword title and is indexed, and the word description and image and characteristics of image in webpage are carried out
Associated storage;
Step 105, provides text based retrieval and Image Feature Matching retrieval feedback during retrieval, feedback information comprising image and
Word description.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611014342.4A CN106776710A (en) | 2016-11-18 | 2016-11-18 | A kind of picture and text construction of knowledge base method based on vertical search engine |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611014342.4A CN106776710A (en) | 2016-11-18 | 2016-11-18 | A kind of picture and text construction of knowledge base method based on vertical search engine |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106776710A true CN106776710A (en) | 2017-05-31 |
Family
ID=58969284
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611014342.4A Pending CN106776710A (en) | 2016-11-18 | 2016-11-18 | A kind of picture and text construction of knowledge base method based on vertical search engine |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106776710A (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107247772A (en) * | 2017-06-06 | 2017-10-13 | 合肥创旗信息科技有限公司 | A kind of picture and text search engine based on internet |
CN107679110A (en) * | 2017-09-15 | 2018-02-09 | 广州唯品会研究院有限公司 | The method and device of knowledge mapping is improved with reference to text classification and picture attribute extraction |
CN107705066A (en) * | 2017-09-15 | 2018-02-16 | 广州唯品会研究院有限公司 | Information input method and electronic equipment during a kind of commodity storage |
CN107729900A (en) * | 2017-09-15 | 2018-02-23 | 广州唯品会研究院有限公司 | It is a kind of that the method and apparatus for completing typing information completion is extracted using picture attribute |
CN107861972A (en) * | 2017-09-15 | 2018-03-30 | 广州唯品会研究院有限公司 | The method and apparatus of the full result of display of commodity after a kind of user's typing merchandise news |
CN108537283A (en) * | 2018-04-13 | 2018-09-14 | 厦门美图之家科技有限公司 | A kind of image classification method and convolutional neural networks generation method |
CN109446196A (en) * | 2018-09-26 | 2019-03-08 | 努比亚技术有限公司 | Major key generates processing and determines that method, major key generates and application server |
CN109446363A (en) * | 2018-10-22 | 2019-03-08 | 镇江悦乐网络科技有限公司 | A kind of method for quickly retrieving based on three-dimensional CG animation |
WO2019129293A1 (en) * | 2017-12-29 | 2019-07-04 | 华为技术有限公司 | Feature data generation method and apparatus and feature matching method and apparatus |
CN110019867A (en) * | 2017-10-10 | 2019-07-16 | 阿里巴巴集团控股有限公司 | Image search method, system and index structuring method and medium |
CN110807121A (en) * | 2019-09-29 | 2020-02-18 | 广东墨痕教育科技有限公司 | Electronic education resource matching method based on image-text intelligent identification and computer readable storage medium |
CN111914201A (en) * | 2020-08-07 | 2020-11-10 | 腾讯科技(深圳)有限公司 | Network page processing method and device |
CN112149653A (en) * | 2020-09-16 | 2020-12-29 | 北京达佳互联信息技术有限公司 | Information processing method, information processing device, electronic equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101141456A (en) * | 2007-10-09 | 2008-03-12 | 南京财经大学 | Vertical search based network data excavation method |
CN101373479A (en) * | 2008-09-27 | 2009-02-25 | 华中科技大学 | Method and system for searching computer picture of mammary gland x-ray radiography |
CN102254043A (en) * | 2011-08-17 | 2011-11-23 | 电子科技大学 | Semantic mapping-based clothing image retrieving method |
WO2015023518A2 (en) * | 2013-08-12 | 2015-02-19 | Microsoft Corporation | Browsing images via mined hyperlinked text snippets |
CN104967749A (en) * | 2015-07-29 | 2015-10-07 | 努比亚技术有限公司 | Device and method for processing picture and text information |
-
2016
- 2016-11-18 CN CN201611014342.4A patent/CN106776710A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101141456A (en) * | 2007-10-09 | 2008-03-12 | 南京财经大学 | Vertical search based network data excavation method |
CN101373479A (en) * | 2008-09-27 | 2009-02-25 | 华中科技大学 | Method and system for searching computer picture of mammary gland x-ray radiography |
CN102254043A (en) * | 2011-08-17 | 2011-11-23 | 电子科技大学 | Semantic mapping-based clothing image retrieving method |
WO2015023518A2 (en) * | 2013-08-12 | 2015-02-19 | Microsoft Corporation | Browsing images via mined hyperlinked text snippets |
CN104967749A (en) * | 2015-07-29 | 2015-10-07 | 努比亚技术有限公司 | Device and method for processing picture and text information |
Non-Patent Citations (2)
Title |
---|
杨宏玉: "《基于内容的图片垂直搜索引擎设计与实现》", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
雷方元 等: "《基于多尺度形态变换的纹理形状图像检》", 《计算机应用》 * |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107247772A (en) * | 2017-06-06 | 2017-10-13 | 合肥创旗信息科技有限公司 | A kind of picture and text search engine based on internet |
CN107679110A (en) * | 2017-09-15 | 2018-02-09 | 广州唯品会研究院有限公司 | The method and device of knowledge mapping is improved with reference to text classification and picture attribute extraction |
CN107705066A (en) * | 2017-09-15 | 2018-02-16 | 广州唯品会研究院有限公司 | Information input method and electronic equipment during a kind of commodity storage |
CN107729900A (en) * | 2017-09-15 | 2018-02-23 | 广州唯品会研究院有限公司 | It is a kind of that the method and apparatus for completing typing information completion is extracted using picture attribute |
CN107861972A (en) * | 2017-09-15 | 2018-03-30 | 广州唯品会研究院有限公司 | The method and apparatus of the full result of display of commodity after a kind of user's typing merchandise news |
CN107705066B (en) * | 2017-09-15 | 2022-01-07 | 唯品会(海南)电子商务有限公司 | Information input method and electronic equipment during commodity warehousing |
CN107729900B (en) * | 2017-09-15 | 2021-04-27 | 广州唯品会研究院有限公司 | Method and equipment for completing input information completion by utilizing picture attribute extraction |
CN110019867A (en) * | 2017-10-10 | 2019-07-16 | 阿里巴巴集团控股有限公司 | Image search method, system and index structuring method and medium |
WO2019129293A1 (en) * | 2017-12-29 | 2019-07-04 | 华为技术有限公司 | Feature data generation method and apparatus and feature matching method and apparatus |
CN109993178A (en) * | 2017-12-29 | 2019-07-09 | 华为技术有限公司 | A kind of characteristic generates and feature matching method and device |
CN109993178B (en) * | 2017-12-29 | 2024-02-02 | 华为技术有限公司 | Feature data generation and feature matching method and device |
CN108537283A (en) * | 2018-04-13 | 2018-09-14 | 厦门美图之家科技有限公司 | A kind of image classification method and convolutional neural networks generation method |
CN109446196A (en) * | 2018-09-26 | 2019-03-08 | 努比亚技术有限公司 | Major key generates processing and determines that method, major key generates and application server |
CN109446363A (en) * | 2018-10-22 | 2019-03-08 | 镇江悦乐网络科技有限公司 | A kind of method for quickly retrieving based on three-dimensional CG animation |
CN110807121A (en) * | 2019-09-29 | 2020-02-18 | 广东墨痕教育科技有限公司 | Electronic education resource matching method based on image-text intelligent identification and computer readable storage medium |
CN111914201A (en) * | 2020-08-07 | 2020-11-10 | 腾讯科技(深圳)有限公司 | Network page processing method and device |
CN111914201B (en) * | 2020-08-07 | 2023-11-07 | 腾讯科技(深圳)有限公司 | Processing method and device of network page |
CN112149653A (en) * | 2020-09-16 | 2020-12-29 | 北京达佳互联信息技术有限公司 | Information processing method, information processing device, electronic equipment and storage medium |
CN112149653B (en) * | 2020-09-16 | 2024-03-29 | 北京达佳互联信息技术有限公司 | Information processing method, information processing device, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106776710A (en) | A kind of picture and text construction of knowledge base method based on vertical search engine | |
WO2022022002A1 (en) | Information display method, information search method and apparatus | |
US9588990B1 (en) | Performing image similarity operations using semantic classification | |
US9489401B1 (en) | Methods and systems for object recognition | |
US20160034514A1 (en) | Providing search results based on an identified user interest and relevance matching | |
Clinchant et al. | Semantic combination of textual and visual information in multimedia retrieval | |
CN104685501B (en) | Text vocabulary is identified in response to visual query | |
US20180189325A1 (en) | Clustering search results based on image composition | |
CN104991959B (en) | A kind of method and system of the same or similar image of information retrieval based on contents | |
CN107944035B (en) | Image recommendation method integrating visual features and user scores | |
US10482146B2 (en) | Systems and methods for automatic customization of content filtering | |
CN109937417A (en) | The system and method for context searchig for electronical record | |
US20120102014A1 (en) | Matching and Recommending Relevant Videos and Media to Individual Search Engine Results | |
CN109614504A (en) | A kind of management system and method for internet electronic book | |
Aramaki et al. | Text detection in manga by combining connected-component-based and region-based classifications | |
WO2011097041A2 (en) | Recommending user image to social network groups | |
US9244919B2 (en) | Organizing books by series | |
CN111563382A (en) | Text information acquisition method and device, storage medium and computer equipment | |
US10191921B1 (en) | System for expanding image search using attributes and associations | |
KR20120047622A (en) | System and method for managing digital contents | |
Papavassiliou et al. | The ilsp/arc submission to the wmt 2016 bilingual document alignment shared task | |
CN117171650A (en) | Document data processing method, system and medium based on web crawler technology | |
Cremaschi et al. | s-elBat: A Semantic Interpretation Approach for Messy taBle-s. | |
CN110765305A (en) | Medium information pushing system and visual feature-based image-text retrieval method thereof | |
EP3905060A1 (en) | Artificial intelligence for content discovery |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170531 |