CN113051457A - Image-text extraction method and terminal - Google Patents
Image-text extraction method and terminal Download PDFInfo
- Publication number
- CN113051457A CN113051457A CN201911369293.XA CN201911369293A CN113051457A CN 113051457 A CN113051457 A CN 113051457A CN 201911369293 A CN201911369293 A CN 201911369293A CN 113051457 A CN113051457 A CN 113051457A
- Authority
- CN
- China
- Prior art keywords
- text
- terminal
- image
- pictures
- extracting
- 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
- 238000000605 extraction Methods 0.000 title abstract description 13
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000012937 correction Methods 0.000 claims abstract description 5
- 238000004891 communication Methods 0.000 claims description 6
- 238000011084 recovery Methods 0.000 claims description 3
- 230000001502 supplementing effect Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 2
- 230000003993 interaction Effects 0.000 claims description 2
- 230000001915 proofreading effect Effects 0.000 claims description 2
- 239000000758 substrate Substances 0.000 claims description 2
- 238000012552 review Methods 0.000 abstract description 4
- 238000012217 deletion Methods 0.000 abstract description 2
- 230000037430 deletion Effects 0.000 abstract description 2
- 239000013589 supplement Substances 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9532—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/146—Aligning or centring of the image pick-up or image-field
- G06V30/1475—Inclination or skew detection or correction of characters or of image to be recognised
- G06V30/1478—Inclination or skew detection or correction of characters or of image to be recognised of characters or characters lines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Character Input (AREA)
Abstract
The application discloses an image-text extraction method and a terminal, wherein the image-text to be extracted is shot into a picture or a picture of the image-text to be extracted is read, the characters and an illustration to be extracted are drawn on the picture by using a selection drawing, the illustration and the illustration are drawn by using a selection frame, or redundant image-text is drawn by using an erasing drawing and an erasing frame in a joint mode, the selected image-text is segmented and cut into a plurality of pictures and sequentially arranged, manual editing or rechecking such as sorting, line division, supplement, deletion and the like is carried out on the cut pictures, OCR recognition processing is carried out on the cut pictures, image-text data corresponding to the cut pictures are extracted, and manual correction can also be carried out on the data. The method and the terminal provided by the invention can realize the recognition and the input of the text charts of the courseware and the homework of the students, particularly the text charts of the combination of the pictures and the texts, manuscripts, wrong lines, section selection, mixed editing of horizontal and vertical lines, disordered sentences and words and the like, improve the recognition range, the efficiency and the accuracy, and facilitate the intelligent review, the network query and the intelligent teaching of the homework.
Description
Technical Field
The invention relates to the technical field of computer information processing and image recognition, in particular to a method and a terminal for extracting pictures and texts.
Background
Students often encounter some difficult problems in learning and need to perform network query, the existing method is to input key words for searching, some rarely-used words, symbols, drawings, tables, non-standard handwriting jobs and the like are difficult to input, especially, text documents such as text-text combination, manuscripts, wrong lines, excerpts, horizontal and vertical mixed editing, sentence and word confusion and the like cannot be identified and extracted, innumerable similarity result recommendations are searched through the key words, correct answers are difficult to find from the recommendation results, meanwhile, with the development of artificial intelligence technology and the like, the intelligent paper reading and intelligent question reading technology is mature, but the quick and accurate entry of homework answers into a question reading system is difficult, the accuracy of the existing text extraction software cannot reach 100%, the chart entry is difficult, and the text extraction entry technology prevents the network query of courseware jobs, the remote teaching and the like, The application and popularization of technologies such as intelligent examination paper marking and intelligent teaching.
Disclosure of Invention
In order to solve the problems of low accuracy, difficult chart identification and the like of the conventional character identification method, the disclosure provides a technical scheme of an image-text extraction method and a terminal.
The method is characterized in that:
the first item of the disclosure is an image-text extraction method, which comprises the following steps:
the first step is as follows: shooting a picture for the picture and text to be extracted or reading a picture of the picture and text to be extracted;
the second step is that: using a selection scribing line to scribe characters and illustrations to be extracted in the picture, and using a selection frame to frame the illustrations and the illustrations, or using an erasing scribing line and an erasing frame to scribe redundant images and texts in a joint mode;
the third step: segmenting the selected image-text into a plurality of pictures which are arranged in sequence;
the fourth step: manual editing or rechecking such as sorting, dividing rows, supplementing and deleting is carried out on the cutting chart;
the fifth step: performing OCR recognition processing on the cut graph, and extracting graph-text data corresponding to the cut graph;
and a sixth step: rechecking or manually correcting the extracted image-text data;
furthermore, each selection scribing line corresponds to one cutting picture, the cutting of the selection pictures and texts is straight-edge cutting and curved-edge cutting which are carried out by taking a row of continuous pictures and texts covered by the scribed selection scribing line or above the selection scribing line as a unit, and the head and the tail of the arrangement of the cutting pictures are consistent with the head and the tail of the scribing line; the method comprises the following steps that an illustration is arranged on a selection scribing line, an illustration selection wire frame is used for selecting the illustration after the selection scribing line is scribed, the illustration is recovered into an original picture after OCR recognition processing, and the layout is recovered and inserted into the original position; the OCR recognition process includes: carrying out binarization, noise erasure, inclination correction, image distortion, image enhancement, layout analysis, character recognition, layout recovery and proofreading on a cut picture, wherein an attached figure is an independent picture, and the layout recovery is carried out after a chart and characters in the attached figure are identified by OCR; each group of extracted image-text information or the substrate is provided with color differences. The second step can use the selective scribing, the selective wire frame, the erasing scribing or the erasing wire frame independently, and can also use various joints, the fourth step checks without errors and does not do other operations, and the sixth step checks without errors and does not do other operations.
The second item of the disclosure is an image-text recognition terminal, comprising: a camera unit, a memory unit, a processor unit, a multimedia unit, a communication unit and a power supply.
Furthermore, the camera unit is used for shooting the image-text to be extracted, and the storage unit is used for storing programs and extracting image-text data; the processor unit is used for program operation to realize the image-text extraction method; the multimedia unit is used for outputting data and realizing interaction between the data and a user; the communication unit is used for data transmission, and data exchange among users, terminals, cloud terminals and the like is realized; the power supply is used for providing electric energy required by operation and work for the terminal.
Drawings
Fig. 1 is a flow chart of image-text extraction disclosed by the invention.
Fig. 2 is an explanatory diagram of the terminal disclosed in the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention are 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 embodiments.
Examples
Referring to fig. 1, a method for extracting graphics and text in an embodiment of the invention is disclosed.
Description of the drawings:
101-a first step of taking a picture for a picture and text to be identified or reading a picture of the picture and text to be extracted;
102-a second step, selecting the graphics context to be extracted or drawing out redundant graphics context, comprising: the characters and the illustrations 1021 to be extracted are marked out by using a selection line, the illustrations and the drawings 1022 are marked out by using a selection line frame, and the redundant pictures and texts 1023 are marked out by using an erasing line and an erasing line frame, wherein the characters and the illustrations can be used independently or in combination;
103-a third step of segmenting the selected pictures and texts into a plurality of pictures which are sequentially arranged;
104-the fourth step, carrying out manual editing or rechecking on the cutting chart, such as sorting, dividing rows, supplementing, deleting and the like;
105-a fifth step, performing OCR recognition processing to extract the image-text to be recognized;
106-sixth step, performing recheck or manual correction on the extracted data.
Fig. 1 is a diagram of an image-text extraction method of the present disclosure, for example, when a student sheet XX encounters a mathematic problem in learning, a network query needs to be performed through a mobile phone, some mathematic symbols in the problem and the inserted-drawing mobile phone cannot be input, and handwriting manuscripts are relatively disordered, more wrong lines and words are disordered, at this time, the student sheet XX takes a homework question as a picture 101, an erasing line is used to draw 1023 parts of multiple shots, then a selection line is used to draw a character and an inserted-drawing 1021 to be extracted line by line, then an inserted-drawing 1022 is framed by a selection line frame, an image-text extraction program cuts the selected image-text into multiple pictures and arranges 103 in sequence, the student sheet XX is a review cut-out picture and a typeset 104, if there is a problem, editing 104 such as manual sorting, line division, supplement, deletion and the like is performed, an OCR recognition process is performed by the image-text extraction program, the homework question 105 to be recognized is, if there is a deviation, a manual correction 106 is performed. The method effectively solves the input problems of rarely-used words, symbols, drawings, forms, non-standard handwriting operation and the like, and ensures that the image-text extraction accuracy reaches 100 percent.
Referring to fig. 2, a teletext extraction terminal according to an embodiment of the invention.
Description of the drawings:
200-an image-text extraction terminal;
201-memory unit, 202-processor unit, 203-camera unit, 204-multimedia unit, 205-communication unit, 206-power supply.
Fig. 2 shows an image-text extraction terminal of the present disclosure, for example, student li X needs an intelligent paper examination system to review its own paper examination paper, student li X opens a camera unit 203 of a mobile intelligent terminal 200 to take a picture of a handwritten paper examination paper, at this time, the picture is automatically stored in a storage unit 201, then a processor unit 202 calls an image-text extraction program, performs required image-text selection and drawing frame selection through a multimedia unit 204, the image-text extraction program performs OCR recognition processing on the examination paper to generate a digital examination paper, and then transmits the digital examination paper to a cloud intelligent examination paper reading system through a communication unit 205, and the intelligent examination paper reading system reviews the digital examination paper to give an examination paper reading result. Throughout the process, the power supply 206 provides power to the terminal.
The above-described embodiments of the present invention should not be construed as limiting the scope of the present invention. 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.
Claims (11)
1. A method and a terminal for extracting pictures and texts are characterized by comprising the method and the terminal for extracting the pictures and texts, wherein the method for extracting the pictures and texts comprises the following steps:
the first step is as follows: shooting a picture for the picture and text to be extracted or reading a picture of the picture and text to be extracted;
the second step is that: using a selection scribing line to scribe characters and illustrations to be extracted in the picture, and using a selection frame to frame the illustrations and the illustrations, or using an erasing scribing line and an erasing frame to scribe redundant images and texts in a joint mode;
the third step: segmenting the selected image-text into a plurality of pictures and arranging the pictures in sequence;
the fourth step: manual editing or rechecking such as sorting, dividing rows, supplementing and deleting is carried out on the cutting chart;
the fifth step: performing OCR recognition processing on the cut graph, and extracting graph-text data corresponding to the cut graph;
and a sixth step: and rechecking or manually correcting the extracted image-text data.
2. The method and terminal for extracting teletext according to claim 1, wherein each selection line corresponds to a cut picture.
3. The method and the terminal for extracting the image-text according to claim 1, wherein the illustration is on a selection line, the illustration selection line frame is a frame selection of the illustration after the selection line is marked out, the illustration is recovered to an original image after being subjected to OCR recognition processing, and the layout is recovered and inserted to an original position.
4. The method and the terminal for extracting the pictures and texts according to claim 1, wherein the cutting of the selected pictures and texts is straight-edge cutting and curved-edge cutting which are performed by taking a row of continuous pictures and texts covered by the drawn selected drawing line or above the drawn selected drawing line as a unit, and the head and the tail of the arrangement of the cut pictures are consistent with the head and the tail of the drawn drawing line.
5. The method and the terminal for extracting the image-text according to claim 1, wherein the figure is an independent picture, and the image-text is subjected to the page restoration after the chart and the characters in the figure are identified by the OCR.
6. The method and terminal for extracting graphics context according to claim 1, wherein the OCR recognition process includes: and carrying out binarization, noise erasure, inclination correction, image enhancement, layout analysis, character recognition, layout recovery and proofreading on the cut picture.
7. The method and terminal for extracting graphics and text according to claim 6, wherein each group of graphics and text information or substrate extracted after OCR recognition processing is provided with color difference.
8. The method and terminal for extracting graphics context according to claim 1, wherein the second step can use selection scribe, selection wire frame, erasing scribe or erasing wire frame alone or in combination.
9. The method and terminal for extracting teletext according to claim 1, wherein the fourth step of checking that no errors occur does not perform any other operation.
10. The method and terminal for extracting teletext according to claim 1, wherein the sixth step of checking if there is no error does not perform any other operation.
11. The method and terminal for extracting teletext according to claim 1, wherein the terminal comprises: the mobile terminal comprises a camera unit, a storage unit, a processor unit, a multimedia unit, a communication unit and a power supply, wherein the camera unit is used for shooting pictures and texts to be extracted, the storage unit is used for storing programs and extracting picture and text data, the processor unit is used for program operation, the multimedia unit is used for data output and interaction, the communication unit is used for data transmission, and the power supply is used for providing electric energy required by operation and work for the terminal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911369293.XA CN113051457A (en) | 2019-12-26 | 2019-12-26 | Image-text extraction method and terminal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911369293.XA CN113051457A (en) | 2019-12-26 | 2019-12-26 | Image-text extraction method and terminal |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113051457A true CN113051457A (en) | 2021-06-29 |
Family
ID=76505587
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911369293.XA Pending CN113051457A (en) | 2019-12-26 | 2019-12-26 | Image-text extraction method and terminal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113051457A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114220305A (en) * | 2021-12-08 | 2022-03-22 | 安徽新华传媒股份有限公司 | Teaching system based on artificial intelligence image recognition technology |
CN115509373A (en) * | 2022-10-11 | 2022-12-23 | 北京数科网维技术有限责任公司 | Method for improving rarely-used character input |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533317A (en) * | 2008-03-13 | 2009-09-16 | 三星电子株式会社 | Fast recording device with handwriting identifying function and method thereof |
CN102479326A (en) * | 2010-11-30 | 2012-05-30 | 方正国际软件(北京)有限公司 | Man-operated proofreading auxiliary method of picture-text identification and system thereof |
CN107451582A (en) * | 2017-07-13 | 2017-12-08 | 安徽声讯信息技术有限公司 | A kind of graphics context identifying system and its recognition methods |
CN110210413A (en) * | 2019-06-04 | 2019-09-06 | 哈尔滨工业大学 | A kind of multidisciplinary paper content detection based on deep learning and identifying system and method |
-
2019
- 2019-12-26 CN CN201911369293.XA patent/CN113051457A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533317A (en) * | 2008-03-13 | 2009-09-16 | 三星电子株式会社 | Fast recording device with handwriting identifying function and method thereof |
CN102479326A (en) * | 2010-11-30 | 2012-05-30 | 方正国际软件(北京)有限公司 | Man-operated proofreading auxiliary method of picture-text identification and system thereof |
CN107451582A (en) * | 2017-07-13 | 2017-12-08 | 安徽声讯信息技术有限公司 | A kind of graphics context identifying system and its recognition methods |
CN110210413A (en) * | 2019-06-04 | 2019-09-06 | 哈尔滨工业大学 | A kind of multidisciplinary paper content detection based on deep learning and identifying system and method |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114220305A (en) * | 2021-12-08 | 2022-03-22 | 安徽新华传媒股份有限公司 | Teaching system based on artificial intelligence image recognition technology |
CN114220305B (en) * | 2021-12-08 | 2024-04-02 | 安徽新华传媒股份有限公司 | Teaching system based on artificial intelligent image recognition technology |
CN115509373A (en) * | 2022-10-11 | 2022-12-23 | 北京数科网维技术有限责任公司 | Method for improving rarely-used character input |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8509563B2 (en) | Generation of documents from images | |
US20030004991A1 (en) | Correlating handwritten annotations to a document | |
US20170220858A1 (en) | Optical recognition of tables | |
CN112115111A (en) | OCR-based document version management method and system | |
CN114021543B (en) | Document comparison analysis method and system based on table structure analysis | |
CN112115301A (en) | Video annotation method and system based on classroom notes | |
CN113051457A (en) | Image-text extraction method and terminal | |
CN111276149A (en) | Voice recognition method, device, equipment and readable storage medium | |
CN110837793A (en) | Intelligent recognition handwriting mathematical formula reading and amending system | |
CN112149680A (en) | Wrong word detection and identification method and device, electronic equipment and storage medium | |
CN111563372A (en) | Typesetting document content self-duplication checking method based on teaching book publishing | |
CN114722842A (en) | Computer artificial intelligent foreign language translation method and translation system thereof | |
CN114781997A (en) | Intelligent examination system and implementation method for special construction scheme of critical engineering | |
KR101626500B1 (en) | System and method for ordering word based on o c r character recognition | |
Bagley et al. | Editing images of text | |
CN112749692A (en) | Intelligent reading and amending system | |
CN114579796B (en) | Machine reading understanding method and device | |
CN115565193A (en) | Questionnaire information input method and device, electronic equipment and storage medium | |
CN114328804A (en) | Method and system for searching key words containing character pictures | |
JPH05303619A (en) | Electronic scrap book | |
CN112364632A (en) | Book checking method and device | |
CN111523307A (en) | Online translation new word note generation system based on symbolic marks | |
CN112784780B (en) | Review method, review device, computer equipment and storage medium | |
KR20230174530A (en) | Cartoon cut auto processing using deep learning | |
CN115202542B (en) | Automatic link and skip method for circuit ports in electronic drawing based on OCR technology |
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: 20210629 |