CN110175616A - A kind of paper image answer extraction method based on color - Google Patents
A kind of paper image answer extraction method based on color Download PDFInfo
- Publication number
- CN110175616A CN110175616A CN201910406972.3A CN201910406972A CN110175616A CN 110175616 A CN110175616 A CN 110175616A CN 201910406972 A CN201910406972 A CN 201910406972A CN 110175616 A CN110175616 A CN 110175616A
- Authority
- CN
- China
- Prior art keywords
- color
- paper
- obtains
- image
- topic
- 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 claims abstract description 13
- 239000003086 colorant Substances 0.000 claims abstract description 5
- 230000000007 visual effect Effects 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 12
- 238000000034 method Methods 0.000 description 21
- 230000011218 segmentation Effects 0.000 description 10
- 239000000284 extract Substances 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013517 stratification Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of paper image answer extraction method based on color, including step 1, according to paper topic classification, select different colours answers questions in writing topic；Step 2 obtains the digital picture of paper, and digital picture is converted to HSV space from RGB color, by the component threshold value of each color, obtains the color region containing each color, setting gray value obtains bianry image；Bianry image is found out all connected domains and corresponding boundary rectangle by chain code following mode by step 3；Step 4 finds out the maximum value of boundary rectangle, it is text filed accordingly to obtain each color to the boundary rectangle of same color.The present invention can complete different types of paper printed page analysis, effectively extracted according to color component to answer different types of in paper region, meet the technical requirements of paper image complexity and higher precision.
Description
Technical field
The present invention relates to technical field of image processing more particularly to a kind of paper image answer based on color to automatically extract
Method.
Background technique
Papery paper is automatically analyzed, needs first to be scanned paper, obtains paper image, then pass through image
The method of processing analyzes paper.In examination paper analysis, it is necessary first to carry out printed page analysis to paper, that is, tell paper
In each component part, the score in examination question region, answer region and paper including paper fills in region, examinee information
Region etc..Effective extract to these regions is the subsequent basis for carrying out paper content analysis.
It include that Page Segmentation, text block identification and printed page understanding are several for printed page analysis in field of image processing
Process.Page Segmentation is that document is divided into relatively independent different zones according to certain logical relation.In turn, subsequent
Region recognition after segmentation can be picture, paragraph, table etc. by text block identification.
Under the conditions of current technology, classical Page Segmentation method can be divided into stratification and non-hierarchical method.It is non-hierarchical
Method obtains pre-segmentation result by being split to original image.Then it again on the basis of pre-segmentation result, carries out feature and mentions
It takes, to obtain more accurate effect.Such methods segmentation precision with higher, but algorithm complexity is higher, is not easy
In real-time processing.Such as blank background Page Segmentation method and Page Segmentation method based on texture.
Hierarchical method handles document layout according to certain level as its name suggests.Bottom-up approach is by using office
Portion's feature models file and picture, obtains the zonule of document, and then constantly merge to obtain the region of entire document.The party
Method has good effect for document detail feature, is suitble to the more complicated space of a whole page.However its computation complexity is high, wants to equipment
Ask higher.Top-down method then needs the priori knowledge by document, models to the overall distribution of document, thus
To each logic region of document.This method speed is fast, however bad for complicated space of a whole page effect.In addition, it is desirable to more elder generation
Knowledge is tested, this point is often difficult to obtain in practice.Such as the Page Segmentation method based on context analyzer, need known text
The global shape of block.
Under the conditions of current technology, no matter which kind of method is used, both for specific file structure, without any one
Kind method is capable of handling all document cases.Therefore, the accuracy rate of document analysis can not all accomplish that 100% is correct.In needle
To in the printed page analysis of paper, since different paper typesettings differs widely, meanwhile, different times and different user take pictures and
The complicated multiplicity of scanning, the printed page analysis difficulty of paper image increase, and classic algorithm can not be often satisfied with for a variety of papers
Effect.Therefore, the present invention provides the printed page analysis method of another thinking, effectively classic algorithm can be overcome for difference
The Problem of Failure of paper printed page analysis is suitable for a variety of different scenes.
Summary of the invention
The purpose of the present invention is to provide a kind of paper image answer extraction method based on color, is mentioned by color
Method is taken, answer different types of in paper region is effectively extracted.
To achieve the above object, technical scheme is as follows:
A kind of paper image answer extraction method based on color, includes the following steps:
Step 1, according to paper topic classification, select different colours answers questions in writing topic；
Step 2 obtains the digital picture of paper, digital picture is converted to HSV space from RGB color, by every
The component threshold value of kind color, obtains the color region containing each color, and setting gray value obtains bianry image；
Bianry image is found out all connected domains and corresponding boundary rectangle by chain code following mode by step 3；
Step 4 finds out the maximum value of boundary rectangle to all boundary rectangles, obtains the corresponding text area of each color
Domain.
In the step 1, topic is answered questions in writing using red in paper objective item part, and paper subjective item part is answered questions in writing using green
Topic, paper visuals use blue pen answer.
In the step 2, red area selects the tonal range value of H=[0,8] and [130,180], obtains red color area
The bianry image in domain；Green area selects the tonal range value of H=[40,80], obtains the bianry image of green area；Blue
The tonal range value of H=[100,124] is selected in region, obtains the bianry image of blue region.
Paper image answer extraction method based on color of the invention, can complete the different types of paper space of a whole page
Analysis, effectively extracts answer different types of in paper region according to color component, meets paper image complexity and more
High-precision technical requirements.Meanwhile paper topic types being classified based on color, be conducive to the construction of comparison process classifier.
Detailed description of the invention
Fig. 1 is the flow chart of the paper image answer extraction method in one embodiment of the invention based on color.
Specific embodiment
Technical solution of the present invention is described in further detail with reference to the accompanying drawings and examples.
A kind of paper image answer extraction method based on color of the invention, as shown in Figure 1, including following step
It is rapid:
Step 1, according to paper topic classification, select different colours answers questions in writing topic；
Step 2 obtains the digital picture of paper, digital picture is converted to HSV space from RGB color, by every
The component threshold value of kind color, obtains the color region containing each color, and setting gray value obtains bianry image；
The bianry image of regions of different colours is found out all connected domain and phase by chain code following mode by step 3
The boundary rectangle answered；
Step 4 finds out the maximum value of boundary rectangle, it is literary accordingly to obtain each color to the boundary rectangle of same color
One's respective area.
By taking specific answer paper answer automatically extracts as an example
During answer, topic, including multiple-choice question, gap-filling questions and judgement are answered questions in writing using red to objective item part in paper
Topic；Subjective item part topic, including question-and-answer problem and theme are answered questions in writing using green；The drafting of figure uses blue pen.
Paper to be paved, is taken pictures using camera and obtains the digital picture of papery paper, specific color extraction method is,
Digital picture is converted into HSV space from RGB color, by extracting the H component threshold value of red, green, blue, extracts phase
The color region answered.
For the red component of HSV space, using the value in H=[0,8] and [130,180] range, obtained red
The bianry image in region；Green area selects the tonal range value of H=[40,80], obtains the bianry image of green area；It is blue
The tonal range value of H=[100,124] is selected in color region, obtains the bianry image of blue region.For the two of each color
It is worth image and the boundary rectangle of all connected domain and its corresponding color is found out by chain code following mode.
Finally, the maximum value of boundary rectangle is found out to the boundary rectangle of same color, to obtain every piece of phase of same color
Even text filed, finally obtains the text filed of corresponding color classification.
Paper image answer extraction method based on color of the invention, can complete the different types of paper space of a whole page
Analysis, effectively extracts answer different types of in paper region according to color component, meets paper image complexity and more
High-precision technical requirements.Meanwhile paper topic types being classified based on color, be conducive to the classification of comparison process classification design
Device avoids a piece of paper volume one classifier of design, and process is complicated, at high cost, difficulty is big.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that the foregoing is merely a specific embodiment of the invention, the guarantor that is not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all be contained in this hair
Within bright protection scope.
Claims (3)
1. a kind of paper image answer extraction method based on color, which comprises the steps of:
Step 1, according to paper topic classification, select different colours answers questions in writing topic；
Step 2 obtains the digital picture of paper, and digital picture is converted to HSV space from RGB color, passes through every kind of face
The component threshold value of color, obtains the color region containing each color, and setting gray value obtains bianry image；
The bianry image of each color by chain code following mode, is found out all connected domain and corresponding external by step 3
Rectangle；
Step 4 finds out the maximum value of boundary rectangle to the boundary rectangle of same color, obtains the corresponding text area of each color
Domain.
2. the paper image answer extraction method according to claim 1 based on color, it is characterised in that: step 1
In, topic is answered questions in writing using red in paper objective item part, and topic is answered questions in writing using green in paper subjective item part, and paper visuals uses
Blue pen answer.
3. the paper image answer extraction method according to claim 2 based on color, it is characterised in that: step 2
In, red area selects the tonal range value of H=[0,8] and [130,180], obtains the bianry image of red area；Green
The tonal range value of H=[40,80] is selected in region, obtains the bianry image of green area；Blue region selection H=[100,
124] tonal range value, obtains the bianry image of blue region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910406972.3A CN110175616A (en) | 2019-05-15 | 2019-05-15 | A kind of paper image answer extraction method based on color |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910406972.3A CN110175616A (en) | 2019-05-15 | 2019-05-15 | A kind of paper image answer extraction method based on color |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110175616A true CN110175616A (en) | 2019-08-27 |
Family
ID=67691287
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910406972.3A Pending CN110175616A (en) | 2019-05-15 | 2019-05-15 | A kind of paper image answer extraction method based on color |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110175616A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663869A (en) * | 2012-04-23 | 2012-09-12 | 国家消防工程技术研究中心 | Indoor fire detection method based on video monitoring platform |
| CN104574960A (en) * | 2014-12-25 | 2015-04-29 | 宁波中国科学院信息技术应用研究院 | Traffic light recognition method |
| CN104794948A (en) * | 2015-04-20 | 2015-07-22 | 西安青柠电子信息技术有限公司 | Automatic scoring system and method for applying same |
| CN104899586A (en) * | 2014-03-03 | 2015-09-09 | 阿里巴巴集团控股有限公司 | Method for recognizing character contents included in image and device thereof |
| CN105095892A (en) * | 2014-05-16 | 2015-11-25 | 上海市上海中学 | Student document management system based on image processing |
| CN106408846A (en) * | 2016-11-29 | 2017-02-15 | 周川 | Image fire hazard detection method based on video monitoring platform |
| CN108171297A (en) * | 2018-01-24 | 2018-06-15 | 谢德刚 | A kind of answer card identification method and device |
-
2019
- 2019-05-15 CN CN201910406972.3A patent/CN110175616A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663869A (en) * | 2012-04-23 | 2012-09-12 | 国家消防工程技术研究中心 | Indoor fire detection method based on video monitoring platform |
| CN104899586A (en) * | 2014-03-03 | 2015-09-09 | 阿里巴巴集团控股有限公司 | Method for recognizing character contents included in image and device thereof |
| CN105095892A (en) * | 2014-05-16 | 2015-11-25 | 上海市上海中学 | Student document management system based on image processing |
| CN104574960A (en) * | 2014-12-25 | 2015-04-29 | 宁波中国科学院信息技术应用研究院 | Traffic light recognition method |
| CN104794948A (en) * | 2015-04-20 | 2015-07-22 | 西安青柠电子信息技术有限公司 | Automatic scoring system and method for applying same |
| CN106408846A (en) * | 2016-11-29 | 2017-02-15 | 周川 | Image fire hazard detection method based on video monitoring platform |
| CN108171297A (en) * | 2018-01-24 | 2018-06-15 | 谢德刚 | A kind of answer card identification method and device |
Non-Patent Citations (1)
| Title |
|---|
| 海底小星星: ""【opencv】在hsv颜色空间识别区域颜色"", 《HTTPS://BLOG.CSDN.NET/U013270326/ARTICLE/DETAILS/80569003》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764074B (en) | Subjective item intelligently reading method, system and storage medium based on deep learning | |
| CN111401372B (en) | Method for extracting and identifying image-text information of scanned document | |
| CN102592126B (en) | For the method for binaryzation scanning document image | |
| CN104408449B (en) | Intelligent mobile terminal scene literal processing method | |
| US20090148043A1 (en) | Method for extracting text from a compound digital image | |
| CN110020692A (en) | A kind of handwritten form separation and localization method based on block letter template | |
| Roy et al. | Wavelet-gradient-fusion for video text binarization | |
| McBride et al. | A comparison of skin detection algorithms for hand gesture recognition | |
| CN106228157A (en) | Coloured image word paragraph segmentation based on image recognition technology and recognition methods | |
| Brisinello et al. | Optical Character Recognition on images with colorful background | |
| Bouillon et al. | Grayification: a meaningful grayscale conversion to improve handwritten historical documents analysis | |
| CN114972847A (en) | Image processing method and device | |
| CN103530625A (en) | Optical character recognition method based on digital image processing | |
| CN115082776A (en) | Electric energy meter automatic detection system and method based on image recognition | |
| KR20140049525A (en) | System and method for displaying visual information based on haptic display for blind person | |
| CN106033534A (en) | Electronic paper marking method based on linear detection | |
| Chen et al. | A knowledge-based system for extracting text-lines from mixed and overlapping text/graphics compound document images | |
| CN110298236A (en) | A kind of braille automatic distinguishing method for image and system based on deep learning | |
| CN103617423A (en) | Image segmentation and recognition method based on color parameter | |
| CN110175616A (en) | A kind of paper image answer extraction method based on color | |
| Ouji et al. | Chromatic/achromatic separation in noisy document images | |
| CN115393865A (en) | Character retrieval method, character retrieval equipment and computer-readable storage medium | |
| Mai et al. | A study about the reconstruction of remote, low resolution mobile captured text images for OCR | |
| CN114332866A (en) | Document curve separation and coordinate information extraction method based on image processing | |
| Wu et al. | Recursive algorithms for image segmentation based on a discriminant criterion |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190827 |