CN107067399A - A kind of paper image segmentation processing method - Google Patents
A kind of paper image segmentation processing method Download PDFInfo
- Publication number
- CN107067399A CN107067399A CN201710077708.0A CN201710077708A CN107067399A CN 107067399 A CN107067399 A CN 107067399A CN 201710077708 A CN201710077708 A CN 201710077708A CN 107067399 A CN107067399 A CN 107067399A
- Authority
- CN
- China
- Prior art keywords
- topic
- paper
- marking block
- locative marking
- page
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10008—Still image; Photographic image from scanner, fax or copier
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30204—Marker
Abstract
The present invention proposes a kind of paper image segmentation processing method, comprises the following steps：Obtain pending paper image；Obtain Examination Paper Template data, including page locative marking block and topic locative marking block data；All page locative marking blocks and topic locative marking block in paper are identified according to template data；Page cutting process is carried out to paper image according to page locative marking block；Topic cutting process is carried out to paper image according to topic locative marking block, wherein, original paper is carried out by segmentation paging according to each two adjacent page locative marking block, according to the topic locative marking block on current question both sides top and the topic locative marking block on next topic both sides top as four point coordinates of current question, the topic is split.Paper image accurately can be cut paging and further preserve the cutting classification of each topic by the present invention, and the topic after cutting is easy to display to check, particularly on mobile device.
Description
Technical field
The present invention relates to image processing field, and more particularly to a kind of paper image segmentation processing method.
Background technology
In recent years, develop with the continuous upgrading of information technology, the use of mobile terminal device is increasingly popularized, more
It is convenient and swift, efficiently work, mode of learning is favored all the more by people.In traditional education sector, gradually carry out
The IT application in education sector upgrading of a new generation is explored.Existing basic education stage, the main investigation shape of Students ' Learning situation at home
Formula is still various types of examinations, greatly to college entrance examination, in examine, the as low as operation of basic unit teacher everyday home, modular examination, and respectively
The interim final examination of class, nationwide examination for graduation qualification, hold examination jointly, mould is examined.In this situation, teacher bears the very big work corrected students' papers with paper
Work amount pressure.
Thus all kinds of standardization auxiliary methods of examination have gradually been used in all kinds of large-scale examinations, such as optical character reader (OCR)
(Optical Mark Reader, OMR) carries out the mode of answering card automatic marking processing.Optical character reader (OCR) optical scanner
Method recognizes the mark for printing or writing by certain format, and is converted into the receptible electric signal of computer, so that real
Existing automatic marking.The advantage of optical character reader (OCR) is that, for height cooperation, satisfactory answering card, it reads accurate (i.e. to paint point
Identification have a high accuracy, the bit error rate is less than thousand a ten thousandths), reading rate it is fast, more than 1,000 can be handled each second
Information point (processing speed in terms of A4 breadths, per hour five thousand sheets).This processing mode largely alleviates teacher comment
The workload pressure of paper, but it there is also many shortcomings, such as a whole set of hardware device is expensive, answering card quality requirement is severe
Carve, be only capable of processing objective item and subjective item can not be handled and corrected, so, so far, except in large-scale important choice class examination
Can be by using outer in (college entrance examination, in examine), other more conventional examinations work of going over examination papers can not use the system.
Therefore, the online paper-marking system occurred therewith, particular for objective item, passes through electron scanning, image recognition technology
Automatic scoring is realized, efficiency of going over examination papers is substantially increased.Made at present using the net marking of computer networking technology and scanning technique
For a kind of new mode of marking examination papers, obtain widely applying.Its method is, first by examinee's answering card by high speed scanner, with
Image mode is scanned in system from cover to cover, is formed electronic edition test paper, is not made any to the original image of examinee during being somebody's turn to do
Identity change, make scanning into system electronic edition test paper with examinee it is actual answer the questions in a test paper it is completely the same and all kinds of by computer
Storage device is stored and managed.
Image recognition software recognizes the printed locative marking block in answering card periphery (the size mark of these tag blocks first
Accurate, blackness standard, easily identification), the coordinate system of image is set up accordingly.Image recognition software is according to having stored in internal memory
The printing position (relative to the position of tag block) of the respective option, positions position of the respective option of each problem on image and (sits
Mark).Centered on these positions, the pixel (scope generally than allowing answer person to apply is smaller) of certain area is chosen, is calculated
The summation of the gray value of all pixels, is compared, being considered as answer person higher than it coats with some threshold value in each region
(have selected this option).The set of choices identified and model answer is compared in image recognition software, determines answer person
Whether the answer of every problem in painting is correct.
, it is necessary to carry out cutting page operations to obtained paper image after online paper-marking system is scanned operation to paper,
The paper image of A3 sizes is cut into the paper paging image of A4 sizes, is easy to subsequent operation and storage；By original examination
Volume image progress is cut after the paper paging image completion obtained after page operations, is also needed to the examination on paper paging image sometimes
Topic carries out operation of keeping to the point, and each examination question image is cut out, is easy to classify and stores.
It is by original paper mould when needing the paper image progress to online paper-marking system to cut page and operation pertinent to the point
Locating piece coordinate on plate recognizes paper picture that scanning is obtained, due to the paper of different Paper Printings, its locating piece
Coordinate can be changed, and also resulted in the changes in coordinates of locating piece during follow up scan because paper is out of alignment, at this moment needed
Manually go to be adjusted, expend more time, operation is cumbersome, is not easy to the substantial amounts of paper of batch processing.
The content of the invention
The present invention proposes a kind of paper image segmentation processing method, paper image accurately can be split into paging and further
Each topic segmentation and classification is preserved, the topic after segmentation is easy to display to check, particularly on mobile device.
In order to achieve the above object, the present invention proposes a kind of paper image segmentation processing method, comprises the following steps：
Obtain pending paper image；
Obtain Examination Paper Template data, including page locative marking block data and topic locative marking block data；
All page locative marking blocks in pending paper image are identified according to the page locative marking block data；
All topic locative marking blocks in pending paper image are identified according to the topic locative marking block data；
Page cutting process is carried out to the pending paper image according to the page locative marking block；
Topic cutting process is carried out to the pending paper image according to the topic locative marking block,
Wherein, the page dividing processing of being carried out to paper image is by original according to each two adjacent page locative marking block
Beginning paper carries out segmentation paging, and the topic dividing processing that carried out to paper image is the topic according to current question both sides top
Locative marking block and the topic locative marking block on next topic both sides top as current question four point coordinates, by the topic
Mesh is split.
Further, the page locative marking block data include the girth and area number of single page locative marking block
According to.
Further, when in the girth and area data and template of page locative marking block in the pending paper
Normal data error is identified as actual pages locative marking block when within 5%.
Further, the topic locative marking block data include the girth and area number of single topic locative marking block
According to.
Further, when in the girth and area data and template of topic locative marking block in the pending paper
Normal data error is identified as actual topic locative marking block when within 5%.
Further, the minimum position and maximum of horizontal direction are calculated from all identified topic locative marking blocks
Position, filters out the topic locative marking block of the misrecognition in centre position.
Further, the Examination Paper Template data are protected to obtain system after the Quick Response Code identification in pending paper
The corresponding templates information deposited.
Paper image segmentation processing method proposed by the present invention, passes through the page telltale mark block number in Examination Paper Template data
According to topic locative marking block data, recognize the page locative marking block and topic locative marking block in actual paper, and go
Except the locative marking block of misrecognition, paper image is split by paging by the adjacent page locative marking block of each two, according to working as
The topic locative marking block on preceding topic both sides top and the topic locative marking block on next topic both sides top are used as current topic
The point coordinates of purpose four, the topic is split.By the way that original paper is carried out after segmentation paging and topic dividing processing, just
Preserved in classifying to each topic, subsequent query is checked also can be more convenient, also allow for moving while single topic carries out preservation
Checked in dynamic equipment.
Brief description of the drawings
Fig. 1 show the paper image segmentation processing method flow chart of present pre-ferred embodiments.
Embodiment
The embodiment of the present invention is provided below in conjunction with accompanying drawing, but the invention is not restricted to following embodiment.Root
According to following explanation and claims, advantages and features of the invention will become apparent from.It should be noted that, accompanying drawing is using very simple
The form of change and use non-accurately ratio, be only used for conveniently, lucidly aid in illustrating the embodiment of the present invention purpose.
Fig. 1 is refer to, Fig. 1 show the paper image segmentation processing method flow chart of present pre-ferred embodiments.This hair
It is bright to propose a kind of paper image segmentation processing method, comprise the following steps：
Step S100：Obtain pending paper image；
Step S200：Obtain Examination Paper Template data, including page locative marking block data and topic locative marking block data；
Step S300：All page positioning in pending paper image are identified according to the page locative marking block data
Tag block；
Step S400：All topic positioning in pending paper image are identified according to the topic locative marking block data
Tag block；
Step S500：Page cutting process is carried out to the pending paper image according to the page locative marking block；
Step S600：Topic cutting process is carried out to the pending paper image according to the topic locative marking block,
Wherein, the page dividing processing of being carried out to paper image is by original according to each two adjacent page locative marking block
Beginning paper carries out segmentation paging, and the topic dividing processing that carried out to paper image is the topic according to current question both sides top
Locative marking block and the topic locative marking block on next topic both sides top as current question four point coordinates, by the topic
Mesh is split.
The page locative marking block data include the girth and area data of single page locative marking block.Treated when described
Normal data error in the paper of processing in the girth and area data and template of the page locative marking block when within 5%,
It is identified as actual pages locative marking block.When black side of the appearance similar to locative marking block in the pending paper
Normal data error in block, its girth and area data and template will not be identified as actual pages more than 5%, then and determine
Position tag block.Avoid black bars pollution misrecognition present on the black bars or paper of full-filling answer during student's answer
For page locative marking block.
According to present pre-ferred embodiments, the topic locative marking block data include the week of single topic locative marking block
Long and area data.Further, when the girth and area data and mould of topic locative marking block in the pending paper
Normal data error in plate is identified as actual topic locative marking block when within 5%.When the pending paper
Normal data error in the middle black bars occurred similar to locative marking block, its girth and area data and template is 5%
More than, then it will not be identified as actual topic locative marking block.Avoid the black bars of full-filling answer during student's answer or
Black bars pollution is mistakenly identified as topic locative marking block present on person's paper.
According to present pre-ferred embodiments, horizontal direction is calculated most from all identified topic locative marking blocks
Small position and maximum position, filter out the topic locative marking block of the misrecognition in centre position.When by passing through in previous step
The girth and area data of topic locative marking block are removed after the topic locative marking block of misrecognition, can also there is part knowledge by mistake
When other topic locative marking block and undiscovered situation, such as student's answer on the black bars or paper of full-filling answer
The black bars pollution of presence is missed, and the normal data error in its girth and area data and template is when within 5%, although with
The possibility that upper situation occurs is very low, but still there is a possibility that in theory certain, and now it can be misidentified as topic
Mesh locative marking block, it is therefore desirable to according to all identified topic locative marking blocks calculate horizontal direction minimum position and
Maximum position, filters out the topic locative marking block of the misrecognition in centre position, and the topic locative marking block of misrecognition can be present
In the centre position of the topic locative marking block of the left and right sides, by the topic locative marking block for judging away misrecognition.Pass through
Above-mentioned two-step pretreatment, the topic locative marking block of misrecognition can be filtered completely, it is ensured that the accuracy of data.
According to present pre-ferred embodiments, the page dividing processing that carried out to paper image is according to each two adjacent page
Original paper is carried out segmentation paging by face locative marking block, and the topic dividing processing that carried out to paper image is according to current topic
The topic locative marking block on mesh both sides top and the topic locative marking block on next topic both sides top are used as current question
Four point coordinates, the topic is split.After the completion of all dividing processings, whole paper is divided into independent examination question, after
It is continuous classification preservation to be carried out according to the type of different topics, label can be added and be easy to classify or inquire about, while being also convenient for
Checked on mobile device.
The correspondence that the Examination Paper Template data preserve for acquisition system after the Quick Response Code identification in pending paper
Template Information.A large amount of Examination Paper Template data are preserved in the database of online paper-marking system, by recognizing on pending paper
Quick Response Code, can find the corresponding Examination Paper Template information of pending paper, including standard page locative marking block data and topic
Mesh locative marking block data.
In summary, paper image segmentation processing method proposed by the present invention, is determined by the page in Examination Paper Template data
Position tag block data and topic locative marking block data, recognize the page locative marking block and topic telltale mark in actual paper
Block, and the locative marking block of misrecognition is removed, the segmentation of paper image is divided by each two adjacent page locative marking block
Page, makees according to the topic locative marking block on current question both sides top and the topic locative marking block on next topic both sides top
For four point coordinates of current question, the topic is split.By the way that original paper is carried out into segmentation paging and topic segmentation
After processing, being easy to classify to each topic preserves, and subsequent query is checked also can be more convenient, while single topic is preserved
It is easy to be checked on the mobile apparatus.
Although the present invention is disclosed above with preferred embodiment, so it is not limited to the present invention.Skill belonging to of the invention
Has usually intellectual in art field, without departing from the spirit and scope of the present invention, when can be used for a variety of modifications and variations.Cause
This, the scope of protection of the present invention is defined by those of the claims.
Claims (7)
1. a kind of paper image segmentation processing method, it is characterised in that comprise the following steps：
Obtain pending paper image；
Obtain Examination Paper Template data, including page locative marking block data and topic locative marking block data；
All page locative marking blocks in pending paper image are identified according to the page locative marking block data；
All topic locative marking blocks in pending paper image are identified according to the topic locative marking block data；
Page cutting process is carried out to the pending paper image according to the page locative marking block；
Topic cutting process is carried out to the pending paper image according to the topic locative marking block,
Wherein, the page dividing processing of being carried out to paper image is by original examination according to each two adjacent page locative marking block
Volume carries out segmentation paging, and the topic dividing processing that carried out to paper image is to be positioned according to the topic on current question both sides top
Tag block and the topic locative marking block on next topic both sides top enter the topic as four point coordinates of current question
Row segmentation.
2. paper image segmentation processing method according to claim 1, it is characterised in that the page telltale mark block number
According to girth and area data including single page locative marking block.
3. paper image segmentation processing method according to claim 2, it is characterised in that when in the pending paper
Normal data error in the girth and area data and template of page locative marking block is identified as reality when within 5%
Border page locative marking block.
4. paper image segmentation processing method according to claim 1, it is characterised in that the topic telltale mark block number
According to girth and area data including single topic locative marking block.
5. paper image segmentation processing method according to claim 4, it is characterised in that when in the pending paper
Normal data error in the girth and area data and template of topic locative marking block is identified as reality when within 5%
Border topic locative marking block.
6. the paper cutting process method of online paper-marking system according to claim 1, it is characterised in that known from all
The minimum position and maximum position of horizontal direction are calculated in other topic locative marking block, the misrecognition in centre position is filtered out
Topic locative marking block.
7. the paper cutting process method of online paper-marking system according to claim 1, it is characterised in that the paper mould
The corresponding templates information that plate data preserve for acquisition system after the Quick Response Code identification in pending paper.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710077708.0A CN107067399A (en) | 2017-02-13 | 2017-02-13 | A kind of paper image segmentation processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710077708.0A CN107067399A (en) | 2017-02-13 | 2017-02-13 | A kind of paper image segmentation processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107067399A true CN107067399A (en) | 2017-08-18 |
Family
ID=59598575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710077708.0A Pending CN107067399A (en) | 2017-02-13 | 2017-02-13 | A kind of paper image segmentation processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107067399A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230308A (en) * | 2017-12-29 | 2018-06-29 | 武汉璞华大数据技术有限公司 | A kind of handbill proofreading method, device and equipment |
| CN109815948A (en) * | 2019-01-14 | 2019-05-28 | 辽宁大学 | A kind of paper partitioning algorithm under complex scene |
| CN110008933A (en) * | 2019-04-18 | 2019-07-12 | 江苏曲速教育科技有限公司 | A kind of universal intelligent marking system and method |
| CN110309754A (en) * | 2019-06-25 | 2019-10-08 | 浙江飙速教育科技有限公司 | A kind of method and system of exercise acquisition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104820835A (en) * | 2015-04-29 | 2015-08-05 | 岭南师范学院 | Automatic examination paper marking method for examination papers |
| CN105989347A (en) * | 2015-02-28 | 2016-10-05 | 科大讯飞股份有限公司 | Intelligent marking method and system of objective questions |
| CN106023698A (en) * | 2016-07-29 | 2016-10-12 | 李铧 | Automatic reading and amending method for homework and exercise books |
| CN106142889A (en) * | 2015-04-08 | 2016-11-23 | 季争鸣 | A kind of answering card goed over examination papers for automatic scanning and application process thereof |
-
2017
- 2017-02-13 CN CN201710077708.0A patent/CN107067399A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105989347A (en) * | 2015-02-28 | 2016-10-05 | 科大讯飞股份有限公司 | Intelligent marking method and system of objective questions |
| CN106142889A (en) * | 2015-04-08 | 2016-11-23 | 季争鸣 | A kind of answering card goed over examination papers for automatic scanning and application process thereof |
| CN104820835A (en) * | 2015-04-29 | 2015-08-05 | 岭南师范学院 | Automatic examination paper marking method for examination papers |
| CN106023698A (en) * | 2016-07-29 | 2016-10-12 | 李铧 | Automatic reading and amending method for homework and exercise books |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230308A (en) * | 2017-12-29 | 2018-06-29 | 武汉璞华大数据技术有限公司 | A kind of handbill proofreading method, device and equipment |
| CN108230308B (en) * | 2017-12-29 | 2021-02-09 | 武汉璞华大数据技术有限公司 | Method, device and equipment for checking advertisement bill |
| CN109815948A (en) * | 2019-01-14 | 2019-05-28 | 辽宁大学 | A kind of paper partitioning algorithm under complex scene |
| CN110008933A (en) * | 2019-04-18 | 2019-07-12 | 江苏曲速教育科技有限公司 | A kind of universal intelligent marking system and method |
| CN110309754A (en) * | 2019-06-25 | 2019-10-08 | 浙江飙速教育科技有限公司 | A kind of method and system of exercise acquisition |
| CN110309754B (en) * | 2019-06-25 | 2023-11-24 | 浙江飙速教育科技有限公司 | Problem acquisition method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110008933B (en) | Universal intelligent marking system and method | |
| CN110659584B (en) | Intelligent mark-remaining paper marking system based on image recognition | |
| CN104143094A (en) | Test paper automatic test paper marking processing method and system without answer sheet | |
| CN110597806A (en) | Wrong question set generation and answer statistics system and method based on reading and amending identification | |
| CN107067399A (en) | A kind of paper image segmentation processing method | |
| CN109242856A (en) | A kind of papery operation electronization reads and makes comments method and device | |
| CN107066939A (en) | A kind of paper cutting process method of online paper-marking system | |
| CN110689013A (en) | Automatic marking method and system based on feature recognition | |
| CN109902683A (en) | Paper core point and Data Input method and system | |
| CN110837793A (en) | Intelligent recognition handwriting mathematical formula reading and amending system | |
| CN107220610A (en) | A kind of subjective item fraction recognition methods applied to marking system | |
| CN106815814A (en) | A kind of image contamination processing method for being applied to marking system | |
| CN114267046A (en) | Method and device for correcting direction of document image | |
| CN112348019B (en) | Answer sheet correction method and device, electronic equipment and storage medium | |
| CN107045635A (en) | A kind of paper image paging sub title processing method of online paper-marking system | |
| CN115482535A (en) | Test paper automatic correction method, storage medium and equipment | |
| CN113903039A (en) | Color-based answer area acquisition method for answer sheet | |
| CN114332898A (en) | Automatic correcting method and device for connection test questions and storage medium | |
| CN114550176A (en) | Examination paper correcting method based on deep learning | |
| CN113378824A (en) | Method for inputting test paper information into automatic correcting system | |
| CN113205527A (en) | Intelligent test paper cutting method and system and storage medium | |
| CN112613500A (en) | Campus dynamic scoring system based on deep learning | |
| CN110555396A (en) | Examination paper reading method of intelligent exercise book, computer readable storage medium and electronic equipment | |
| CN106803238A (en) | A kind of answering card image noise reduction processing method | |
| CN111291591A (en) | Intelligent test paper reading and printing method and device |
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 |
Application publication date: 20170818 |
|
| WD01 | Invention patent application deemed withdrawn after publication |