CN112241463A - Search method based on fusion of text semantics and picture information - Google Patents
Search method based on fusion of text semantics and picture information Download PDFInfo
- Publication number
- CN112241463A CN112241463A CN202011082579.2A CN202011082579A CN112241463A CN 112241463 A CN112241463 A CN 112241463A CN 202011082579 A CN202011082579 A CN 202011082579A CN 112241463 A CN112241463 A CN 112241463A
- Authority
- CN
- China
- Prior art keywords
- picture
- pictures
- text
- fusion
- method based
- 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
Images
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
- G06F16/50—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)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a search method based on fusion text semantics and picture information, which comprises the steps of acquiring search words and pictures input by a user, performing semantic analysis on the acquired search words and extracting key words, converting the key words into the pictures through a deep neural network, scanning the acquired pictures, performing semantic analysis and content identification on the pictures, acquiring visual characteristic information of the pictures, acquiring characteristic information of the pictures, checking the consistency of the picture semantics in the visual characteristic information of the pictures and the text context semantics, acquiring the pictures matched with the contents of the pictures from a database according to the characteristic information of the pictures, and presenting the selected pictures through a browser. The searching method based on the fusion of the text semantics and the picture information solves the problems of low accuracy and low response speed of the picture searching method in the traditional technology.
Description
Technical Field
The invention relates to the technical field of picture searching, in particular to a searching method based on fusion of text semantics and picture information.
Background
The picture information is related information of the picture, including picture name, picture source, description information of the picture and the like, and the picture information can reflect related contents of the picture through characters, so that a browser can conveniently view the picture. When a user views a picture stored in the terminal through the picture viewing software, the picture viewing software generally searches for picture information corresponding to the stored picture in a storage area, and then displays the picture information and the picture to the user.
The traditional picture searching method is low in accuracy, searching is carried out in mass data only through pictures, so that the searching accuracy is low, the response speed is low, the data volume in a picture library of the whole network is large, the number of pictures to be matched is large, and the response speed is slow as the number of matched pictures is large.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a searching method based on fusion of text semantics and picture information, and solves the problems of low accuracy and low response speed of a picture searching method in the prior art.
In order to achieve the purpose, the invention is realized by the following technical scheme: a search method based on fusion text semantics and picture information comprises the following steps;
step one; acquiring search words and pictures input by a user;
step two; performing semantic analysis on the acquired search words, extracting key words, and converting the key words into pictures through a deep neural network;
step three; scanning the acquired picture to acquire characteristic information of the picture;
step four; acquiring a picture matched with the content of the picture from a database according to the characteristic information of the picture;
step five; and presenting the selected picture through the browser.
Preferably: and in the second step, scanning the picture comprises performing semantic analysis and content identification on the picture to obtain visual characteristic information of the picture.
Preferably: and the step two of scanning the picture further comprises the step of checking the consistency of the picture semantic meaning in the visual characteristic information of the picture and the text context semantic meaning.
Preferably: and in the third step, the picture matched with the content of the picture is obtained from the database according to the picture, specifically, the obtained picture is changed or expanded through a deep learning technology to generate a related picture, and the picture matched with the related picture is obtained from the database according to the generated picture.
Preferably: and in the third step, the picture matched with the content of the picture is obtained from the database according to the picture, the picture is converted into a text by utilizing deep learning, and the picture label is matched with the text to obtain the picture matched with the text from the database.
Preferably: the pictures matched in the third step comprise a webpage picture, a photographed picture and a picture stored on the terminal.
Preferably: the pictures are presented through the browser, specifically, a preset number of pictures are selected according to the sequence of the matching degree from high to low, and the pictures are presented through the browser.
Advantageous effects
The invention provides a search method based on fusion of text semantics and picture information. The method has the following beneficial effects:
(1) according to the search method based on the fusion text semantics and the image information, the fusion text and the image are searched, a deep learning technology is utilized to generate a countermeasure network to change the input image and expand a system to automatically generate the related image to search, and therefore the search accuracy is high.
(2) According to the search method based on the fusion of the text semantics and the picture information, the keywords are converted into the pictures by using the deep neural network, then the pictures are searched, the picture search by using the pictures is realized, the pictures can be converted into the texts by using the deep learning, and then the texts are matched with the picture labels, so that the problem of slow response speed in the prior art is solved.
Drawings
Fig. 1 is a flowchart of a search method based on fusion of text semantics and picture information.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides a technical solution: a search method based on fusion text semantics and picture information comprises the following steps; the method comprises the steps of obtaining search words and pictures input by a user, carrying out semantic analysis on the obtained search words, extracting key words, converting the key words into the pictures through a deep neural network, scanning the obtained pictures to obtain characteristic information of the pictures, obtaining the pictures matched with the contents of the pictures from a database according to the characteristic information of the pictures, and presenting the selected pictures through a browser.
In the search method based on fusion of text semantics and picture information in the embodiment of the application, the following is preferable: the method comprises the steps of obtaining a search word and a picture input by a user, carrying out semantic analysis on the obtained search word and extracting a keyword, converting the keyword into the picture through a deep neural network, scanning the obtained picture, wherein the scanning of the picture comprises the steps of carrying out semantic analysis and content recognition on the picture, obtaining visual characteristic information of the picture, checking the consistency of the picture semantic in the visual characteristic information of the picture and the text context semantic, changing or expanding the obtained picture through a deep learning technology to generate a related picture, obtaining the picture matched with the picture from a database according to the generated picture, converting the picture into a text in cooperation with the deep learning, matching picture labels with the text, selecting a preset number of pictures from high to low according to the matching degree, and presenting the pictures through a browser.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. A searching method based on fusion of text semantics and picture information comprises the following steps:
step one; acquiring search words and pictures input by a user;
step two; performing semantic analysis on the acquired search words, extracting key words, and converting the key words into pictures through a deep neural network;
step three; scanning the acquired picture to acquire characteristic information of the picture;
step four; acquiring a picture matched with the content of the picture from a database according to the characteristic information of the picture;
step five; and presenting the selected picture through the browser.
2. The search method based on the fusion of text semantics and picture information according to claim 1, wherein: and in the second step, scanning the picture comprises performing semantic analysis and content identification on the picture to obtain visual characteristic information of the picture.
3. The search method based on the fusion of text semantics and picture information according to claim 1, wherein: and the step two of scanning the picture further comprises the step of checking the consistency of the picture semantic meaning in the visual characteristic information of the picture and the text context semantic meaning.
4. The search method based on the fusion of text semantics and picture information according to claim 1, wherein: and in the third step, the picture matched with the content of the picture is obtained from the database according to the picture, the obtained picture is changed or expanded through a deep learning technology to generate a related picture, and the picture matched with the related picture is obtained from the database according to the generated picture.
5. The search method based on the fusion of text semantics and picture information according to claim 1, wherein: and in the third step, the picture matched with the content of the picture is obtained from the database according to the picture, the picture is converted into a text by utilizing deep learning, and the picture label is matched with the text to obtain the picture matched with the text from the database.
6. The search method based on the fusion of text semantics and picture information according to claim 1, wherein: the pictures matched in the third step comprise a webpage picture, a photographed picture and a picture stored on the terminal.
7. The search method based on the fusion of text semantics and picture information according to claim 1, wherein: the pictures are presented through the browser, specifically, a preset number of pictures are selected according to the sequence of the matching degree from high to low, and the pictures are presented through the browser.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011082579.2A CN112241463A (en) | 2020-10-15 | 2020-10-15 | Search method based on fusion of text semantics and picture information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011082579.2A CN112241463A (en) | 2020-10-15 | 2020-10-15 | Search method based on fusion of text semantics and picture information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112241463A true CN112241463A (en) | 2021-01-19 |
Family
ID=74168683
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011082579.2A Pending CN112241463A (en) | 2020-10-15 | 2020-10-15 | Search method based on fusion of text semantics and picture information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112241463A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779363A (en) * | 2021-09-14 | 2021-12-10 | 广州汇通国信科技有限公司 | Search optimization method and device based on deep neural network |
-
2020
- 2020-10-15 CN CN202011082579.2A patent/CN112241463A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779363A (en) * | 2021-09-14 | 2021-12-10 | 广州汇通国信科技有限公司 | Search optimization method and device based on deep neural network |
| CN113779363B (en) * | 2021-09-14 | 2023-12-22 | 广州汇通国信科技有限公司 | Search optimization method and device based on deep neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110502621B (en) | Question answering method, question answering device, computer equipment and storage medium | |
| US9600533B2 (en) | Matching and recommending relevant videos and media to individual search engine results | |
| US12026194B1 (en) | Query modification based on non-textual resource context | |
| US8161059B2 (en) | Method and apparatus for collecting entity aliases | |
| CN111309877A (en) | Intelligent question-answering method and system based on knowledge graph | |
| US8577882B2 (en) | Method and system for searching multilingual documents | |
| CN111291210A (en) | Image material library generation method, image material recommendation method and related device | |
| CN107861753B (en) | APP generation index, retrieval method and system and readable storage medium | |
| CN108121715B (en) | Character labeling method and character labeling device | |
| CN111061828B (en) | Digital library knowledge retrieval method and device | |
| WO2020056977A1 (en) | Knowledge point pushing method and device, and computer readable storage medium | |
| CN109446313B (en) | Sequencing system and method based on natural language analysis | |
| CN112052332A (en) | Retrieval method, retrieval device, electronic equipment and readable storage medium | |
| CN116662495A (en) | Question-answering processing method, and method and device for training question-answering processing model | |
| US9165186B1 (en) | Providing additional information for text in an image | |
| CN112989811B (en) | History book reading auxiliary system based on BiLSTM-CRF and control method thereof | |
| CN114117242A (en) | Data query method and device, computer equipment and storage medium | |
| CN112241463A (en) | Search method based on fusion of text semantics and picture information | |
| CN117420998A (en) | Client UI interaction component generation method, device, terminal and medium | |
| CN111881695A (en) | Audit knowledge retrieval method and device | |
| CN103377199B (en) | Information processor and information processing method | |
| WO2015143911A1 (en) | Method and device for pushing webpages containing time-relevant information | |
| CN115203445A (en) | Multimedia resource searching method, device, equipment and medium | |
| CN112651244B (en) | TopK entity extraction method and system based on paper abstract QA | |
| CN115186240A (en) | Social network user alignment method, device and medium based on relevance information |
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 |