CN106599887A - Test paper answering area identifying and test question marking method and system - Google Patents
Test paper answering area identifying and test question marking method and system Download PDFInfo
- Publication number
- CN106599887A CN106599887A CN201611026941.8A CN201611026941A CN106599887A CN 106599887 A CN106599887 A CN 106599887A CN 201611026941 A CN201611026941 A CN 201611026941A CN 106599887 A CN106599887 A CN 106599887A
- Authority
- CN
- China
- Prior art keywords
- answer
- region
- adaptive value
- information
- answer region
- 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.)
- Granted
Links
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
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a test paper answering area identifying and test question marking method and system, and belongs to the technical field of mobile Internet. The method comprises that m pieces of answering handwriting information are obtained; according to handwriting interval distances in each piece of answering handwriting information, answering areas are divided successively, and a test question amount set C and an answering area set R are formed; the amount n of test questions and the amount of effective answering areas are determined; information of each effective answering area of each piece of answering handwriting information is identified and counted; answers to test questions are determined; and each effective answering area of each piece of answering handwriting information is marked according to each answer to the test question, and a test question marking result of each piece of answering handwriting information is obtained. The method and system can be used to identify the answering areas automatically and accurately, determine the amount of test questions, mark objective questions of test paper, save time and improve the efficiency.
Description
Technical field
The present invention relates to mobile internet technical field, and in particular to a kind of paper answer region recognition and the examination question side of reading and appraising
Method, system.
Background technology
At present, partial students have begun to be tested using the mode of electronic test paper, and with no paper, electronization is education neck
Following trend in domain.For the objective item in paper, traditional technology of correcting automatically is recorded including machine-read card form and electronics examination question
Enter form, it is expensive, it is complicated using process.For machine-read card form, school needs extra card-reading apparatus, expensive,
And need to prepare different disposable cards, waste serious, it is also required to use pencil using machine-read card, pencil is also running stores.
Also, machine-read card is not directly perceived in use, and student often sees presbyopia, poor user experience.
For electronics test questions input form, electronics test questions input is loaded down with trivial details for the process of exam pool, also, teacher's often basis
The concrete condition of this class of student, selects essence topic to explain, and when teacher is selected a topic using exam pool, needs to expend very big energy.If allowing
The test questions input of teachers part is one electronics examination question, then teacher needs to submit examination question and answer to by the form of regulation, greatly
Increase the difficulty of preparing lessons of teacher.
Paper answer region how is automatically and accurately recognized, examination question quantity, and grade papers objective item is determined, during saving
Between, efficiency is improved, it is the problem of those skilled in the art's urgent need to resolve.
The content of the invention
For defect of the prior art, the present invention provides a kind of paper answer region recognition and examination question reads and appraises method, is
System, can automatically and accurately recognize paper answer region, determine examination question quantity, and grade papers objective item, time-consuming, improve
Efficiency.
In a first aspect, the present invention provides a kind of paper answer region recognition and examination question reads and appraises method, the method includes：Obtain
Objective item purpose answering information and model answer；
Answering information is contrasted with model answer, objective item purpose examination question is obtained and is read and appraised result.
The present invention provides another kind of paper answer region recognition and examination question reads and appraises method, and the method includes：
Answer person's handwriting information acquiring step：Obtain m part answer handwriting informations；
Examination question quantity determines step：According to the spacing distance of person's handwriting in every part of answer handwriting information, answer area is divided successively
Domain, forms examination question magnitude-set C and answer regional ensemble R, wherein,
C={ C1 C2 … Ci … Cm, CiFor the answer region quantity of i-th part of answer handwriting information, i=1,2 ..., m,
Ri,jFor j-th answer area of i-th part of answer handwriting information
Domain, j=1,2 ..., Ci；
Maximum answer region quantity in examination question magnitude-set C, determines examination question quantity n；
Effective answer area determination step：By answer region Ri,jEnter with each answer region of kth part answer handwriting information
Row similarity is tested, and determines k-th adaptive value arrayIn k-th adaptive value array
Maximum adaptive value Vkmax, wherein, Vk,jFor answer region Ri,jIt is suitable with j-th answer region of kth part answer handwriting information
Property value, k=1,2 ..., m, and k ≠ i；
According to the maximum adaptive value in each adaptive value array, it is determined that final adaptive valueFinal adaptive valueTo answer
Topic region Ri,jAdaptive value；
By final adaptive valueCompare with given threshold：
If final adaptive valueLess than given threshold, then answer region R is judgedi,jFor invalid answer region,
If final adaptive valueMore than or equal to given threshold, then answer region R is judgedi,jFor effective answer region；
Answer handwriting information reads and appraises step：By t-th of every part of answer handwriting information effective answer region and set information
Contrasted, and Statistical Comparison result：
If the number in effective answer region identical with set information is less than m/2, information is generated,
If the number in effective answer region identical with set information is more than or equal to m/2,：
T-th effective answer region of every part of answer handwriting information is identified, t-th effective answer region of acquisition
Identification information set Dt, wherein, Dt={ D1,t D2,t … Di,t … Dm,t, Di,tFor t-th of i-th part of answer handwriting information
The identification information in answer region, t=1,2 ..., i ... n；
To identification information set DtCounted, and determined t-th script, t-th script is every part of answer
The answer of t-th effective answer region correspondence examination question of handwriting information；
Each effective answer region of every part of answer handwriting information is read and appraised according to each script, and obtains every
The examination question of part answer handwriting information reads and appraises result.
Further, t-th effective answer region of every part of answer handwriting information is identified, obtain t-th it is effective
The identification information set D in answer regiont, specifically include：
T-th effective answer region of every part of answer handwriting information is identified：
If recognizing the identification information set D for successfully obtaining t-th effective answer regiont,
If recognition failures, the answer handwriting information and recognition failures number of times of recognition failures is recorded, and according to pre-seting
Character recognition order, is identified to t-th effective answer region of the answer handwriting information of recognition failures, until obtaining t
The identification information set D in individual effective answer regiont, or recognition failures number of times reaches maximum.
Further, the maximum adaptive value in each adaptive value array, it is determined that final adaptive valueSpecifically include：
According to the maximum adaptive value in each adaptive value array, by equation below, it is determined that final adaptive value
Wherein, VkmaxFor in k-th adaptive value array
Maximum adaptive value, k=1,2 ..., m, and k ≠ i.
Further, to identification information set DtCounted, and determined t-th script, specifically included：To identification
Information aggregate DtCounted, obtained the answer region quantity with same identification information；
By identification information of the answer region quantity more than or equal to m/2, it is defined as t-th script.
Embodiment of the method is read and appraised based on above-mentioned any paper answer region recognition and examination question, further, the method is also wrapped
Include：Edit instruction is received, script is edited.
Second aspect, the present invention provides a kind of paper answer region recognition and examination question reads and appraises system, and the system includes answer
Handwriting information obtains subsystem, examination question quantity and determines that subsystem, effective answer region determine that subsystem and answer handwriting information are commented
Subsystem is read, answer person's handwriting information acquisition subsystem is used to obtain m part answer handwriting informations；Examination question quantity determines that subsystem is used for
According to the spacing distance of person's handwriting in every part of answer handwriting information, answer region is divided successively, form examination question magnitude-set C and answer
Regional ensemble R, wherein, C={ C1 C2 … Ci … Cm, CiFor the answer region quantity of i-th part of answer handwriting information, i=1,
2 ..., m,Ri,jFor j-th answer area of i-th part of answer handwriting information
Domain, j=1,2 ..., Ci；Maximum answer region quantity in examination question magnitude-set C, determines examination question quantity n；Effectively answer
Region determines that subsystem is used for answer region Ri,jSimilarity survey is carried out with each answer region of kth part answer handwriting information
Examination, determines k-th adaptive value arrayFit with the maximum in k-th adaptive value array
Property value Vkmax, wherein, Vk,jFor answer region Ri,jWith the adaptive value in j-th answer region of kth part answer handwriting information, k=
1,2 ..., m, and k ≠ i；According to the maximum adaptive value in each adaptive value array, it is determined that final adaptive valueFinal adaptive valueFor answer region Ri,jAdaptive value；By final adaptive valueCompare with given threshold：If final adaptive valueLess than setting
Determine threshold value, then judge answer region Ri,jFor invalid answer region, if final adaptive valueMore than or equal to given threshold, then judge
Answer region Ri,jFor effective answer region；Answer handwriting information reads and appraises subsystem for by t-th of every part of answer handwriting information
Effectively answer region is contrasted with set information, and Statistical Comparison result：If effective answer region identical with set information
Number is less than m/2, then generate information, right if the number in effective answer region identical with set information is more than or equal to m/2
T-th effective answer region of every part of answer handwriting information is identified, and obtains the identification information collection in t-th effective answer region
Close Dt, wherein, Dt={ D1,t D2,t … Di,t … Dm,t, Di,tFor t-th answer region of i-th part of answer handwriting information
Identification information, t=1,2 ..., i ... n；To identification information set DtCounted, and determined t-th script, t-th examination
Topic answer is the answer of t-th effective answer region correspondence examination question of every part of answer handwriting information；According to each script pair
Each effective answer region of every part of answer handwriting information is read and appraised, and is obtained the examination question of every part of answer handwriting information and read and appraised knot
Really.
Further, answer handwriting information reads and appraises subsystem in t-th effective answer area to every part of answer handwriting information
Domain is identified, and obtains the identification information set D in t-th effective answer regiontWhen, specifically for：To every part of answer person's handwriting letter
T-th effective answer region of breath is identified：If recognizing the identification information collection for successfully obtaining t-th effective answer region
Close DtIf recognition failures record the answer handwriting information and recognition failures number of times of recognition failures, and according to the character for pre-seting
Recognition sequence, is identified to t-th effective answer region of the answer handwriting information of recognition failures, until obtain t-th having
The identification information set D in effect answer regiont, or recognition failures number of times reaches maximum.
Further, effective answer region determines subsystem in the maximum adaptive value in each adaptive value array, really
Fixed final adaptive valueWhen, specifically for：According to the maximum adaptive value in each adaptive value array, by equation below, it is determined that
Final adaptive value Wherein, VkmaxFor k-th adaptive
Maximum adaptive value in value array, k=1,2 ..., m, and k ≠ i.
Further, answer handwriting information reads and appraises subsystem to identification information set DtCounted, and determined t-th
During script, specifically for：To identification information set DtCounted, obtained the answer number of regions with same identification information
Amount；By identification information of the answer region quantity more than or equal to m/2, it is defined as t-th script.
As shown from the above technical solution, paper answer region recognition of the present invention and examination question read and appraise method, system, using big number
According to the mode of real-time statistics, for the examination question of different-format, answer handwriting information is obtained, even if during non-typing script, should
Method also can precisely determine examination question quantity and identification answer region, and the answer region to a large amount of answer handwriting informations carries out phase
Like degree test, according to the final adaptive value in different answer regions, validity identification is carried out to answer region, it is determined that effectively answer area
Domain, is prevented from effective information loss, and advantageously reduces data processing complexity, without the need for using card-reading apparatus or card, lead
The running stores such as pen.
Meanwhile, the method is identified for the content in effective answer region, is commented with determining script, and carrying out examination question
Read, information recognition accuracy is high, script reliability, operation, examination examination question is greatly improved and corrects efficiency, contribute to teacher's realization
Test and explanation with hall, reduction is prepared lessons difficulty, shorten that student inscribes, examination question corrects time of judgement.
Therefore, paper answer region recognition of the present invention and examination question read and appraise method, system, can automatically and accurately recognize paper
Answer region, determines examination question quantity, and grade papers objective item, saves the office hours of staff, improves efficiency.
Description of the drawings
In order to be illustrated more clearly that the specific embodiment of the invention or technical scheme of the prior art, below will be to concrete
The accompanying drawing to be used needed for embodiment or description of the prior art is briefly described.In all of the figs, similar element
Or part is typically identified by similar reference.In accompanying drawing, each element or part might not draw according to actual ratio.
Fig. 1 shows that a kind of paper answer region recognition provided by the present invention and examination question read and appraise method flow diagram；
Fig. 2 shows that another kind of paper answer region recognition provided by the present invention and examination question read and appraise method flow diagram；
Fig. 3 shows that a kind of paper answer region recognition provided by the present invention and examination question read and appraise the structural representation of system.
Specific embodiment
The embodiment of technical solution of the present invention is described in detail below in conjunction with accompanying drawing.Following examples are only used for
Technical scheme is clearly illustrated, therefore is intended only as example, and the protection of the present invention can not be limited with this
Scope.
It should be noted that unless otherwise stated, technical term used in this application or scientific terminology should be this
The ordinary meaning that bright one of ordinary skill in the art are understood.
In a first aspect, the embodiment of the present invention provides a kind of paper answer region recognition and examination question reads and appraises method, the method bag
Include：Obtain objective item purpose answering information and model answer；
Answering information is contrasted with model answer, objective item purpose examination question is obtained and is read and appraised result.
The embodiment of the present invention provides another kind of paper answer region recognition and examination question reads and appraises method, with reference to Fig. 1, the method bag
Include：
Answer person's handwriting information acquiring step S1：M part answer handwriting informations are obtained, here, answer handwriting information is for not
With the answer that the examination question of electronic format is provided, examination question form can be picture format that is various, such as capturing；
Examination question quantity determines step S2：According to the spacing distance of person's handwriting in every part of answer handwriting information, answer is divided successively
Region, forms examination question magnitude-set C and answer regional ensemble R, wherein,
C={ C1 C2 … Ci … Cm, CiFor the answer region quantity of i-th part of answer handwriting information, i=1,2 ..., m,
Ri,jFor j-th answer area of i-th part of answer handwriting information
Domain, j=1,2 ..., Ci；Here, the pixel count in each answer region is 40 × 20；
Maximum answer region quantity in examination question magnitude-set C, determines examination question quantity n；
Effectively answer area determination step S3：By answer region Ri,jWith each answer region of kth part answer handwriting information
Similarity test is carried out, k-th adaptive value array is determinedWith k-th adaptive value number
Maximum adaptive value V in groupkmax, wherein, Vk,jFor answer region Ri,jWith j-th answer region of kth part answer handwriting information
Adaptive value, k=1,2 ..., m, and k ≠ i；Here, the similarity in two answer regions is higher, corresponding adaptive value can be bigger；
According to the maximum adaptive value in each adaptive value array, it is determined that final adaptive valueFinal adaptive valueTo answer
Topic region Ri,jAdaptive value；
By final adaptive valueCompare with given threshold：
If final adaptive valueLess than given threshold, then answer region R is judgedi,jFor invalid answer region,
If final adaptive valueMore than or equal to given threshold, then answer region R is judgedi,jFor effective answer region；
Answer handwriting information reads and appraises step S4：
T-th of every part of answer handwriting information effective answer region is contrasted with set information, and Statistical Comparison is tied
Really：
If the number in effective answer region identical with set information is less than m/2, information is generated, such as generate " non-visitor
Sight is inscribed " suggestion voice, or " non-objective item " prompt text etc., set information is in " A ", " B ", " C ", " D ", " E " five kinds of characters
One kind, or any combination of five kinds of characters, and symbol " √ " or " X ".
If the number in effective answer region identical with set information is more than or equal to m/2,：
T-th effective answer region of every part of answer handwriting information is identified, t-th effective answer region of acquisition
Identification information set Dt, wherein, Dt={ D1,t D2,t … Di,t … Dm,t, Di,tFor t-th of i-th part of answer handwriting information
The identification information in answer region, t=1,2 ..., i ... n；
To identification information set DtCounted, and determined t-th script, t-th script is every part of answer
The answer of t-th effective answer region correspondence examination question of handwriting information；
Each effective answer region of every part of answer handwriting information is read and appraised according to each script, and obtains every
The examination question of part answer handwriting information reads and appraises result.
In actual application, student's answer can adopt answer client, obtain examination question, and upload answer person's handwriting letter
Cease to server, server is obtained after answer handwriting information, determine that step S2 determines examination question quantity using examination question quantity, adopting has
Effect answer area determination step S3 determines effective answer region, reads and appraises step S4 using answer handwriting information and determines script,
And carry out examination question and read and appraise, correct, and the script that will be determined, and examination question reads and appraises result and sends to teacher and check client
End, is reviewed, and its concrete implementation procedure can refer to Fig. 2.
As shown from the above technical solution, the present embodiment paper answer region recognition and examination question read and appraise method, using big data
The mode of real-time statistics, for the examination question of different-format, obtains answer handwriting information, even if during non-typing script, the party
Method also can precisely determine examination question quantity and identification answer region, and the answer region of a large amount of answer handwriting informations be carried out similar
Degree test, according to the final adaptive value in different answer regions, to answer region validity identification is carried out, it is determined that effectively answer area
Domain, is prevented from effective information loss, and advantageously reduces data processing complexity, without the need for using card-reading apparatus or card, lead
The running stores such as pen.
Meanwhile, the method is identified for the content in effective answer region, is commented with determining script, and carrying out examination question
Read, information recognition accuracy is high, script reliability, operation, examination examination question is greatly improved and corrects efficiency, contribute to teacher's realization
Test and explanation with hall, reduction is prepared lessons difficulty, shorten that student inscribes, examination question corrects time of judgement.
Therefore, the present embodiment paper answer region recognition and examination question read and appraise method, can automatically and accurately recognize that paper is answered
Topic region, determines examination question quantity, and grade papers objective item, saves the office hours of staff, improves efficiency.
The precision of method is read and appraised in order to further improve the present embodiment paper answer region recognition and examination question, specifically,
When the identification information in each effective answer region is obtained, the specific implementation of the method is as follows, to every part of answer person's handwriting letter
T-th effective answer region of breath is identified：If recognizing the identification information collection for successfully obtaining t-th effective answer region
Close DtIf recognition failures record the answer handwriting information and recognition failures number of times of recognition failures, and according to the character for pre-seting
Recognition sequence, is identified to t-th effective answer region of the answer handwriting information of recognition failures, until obtain t-th having
The identification information set D in effect answer regiont, or recognition failures number of times reaches maximum.
Here, the method can arrange priority, such as character " H " and " A " for hand-written inaccurate character, " 0 " and
" D ", when hand-written inaccurate, can increase the method intractability, and the method, will character using the character recognition order for pre-seting
The priority of " A " and " D " shifts to an earlier date, by the priority of character " H " and " 0 " rearward, abnormal to reduce information identification, improves information and knows
The other degree of accuracy.
Specifically, it is determined that each answer region final adaptive value when, the method to implement process as follows, according to
Maximum adaptive value in each adaptive value array, by equation below, it is determined that final adaptive value Wherein, VkmaxFit for the maximum in k-th adaptive value array
Property value, k=1,2 ..., m, and k ≠ i.
Here, the calculation rough draft of the non-effective answering information in for answer handwriting information, such as student, the method will be a large amount of
The answer region of answer handwriting information is contrasted, and asks for the mean value of maximum adaptive value in answer handwriting information, data ginseng
Consider greatly, subsequently to carry out answer region availability deciding Informational support is provided, be conducive to improving effective answer regional determination
The degree of accuracy.Because the calculation rough draft of each student can be very low with the similarity of other students, i.e., final adaptive value can be minimum, must
Less than given threshold, so as to be judged to non-effective answer region.
Specifically, it is determined that during script, it is as follows that the method implements process, to identification information set DtUnited
Meter, obtains the answer region quantity with same identification information；Answer region quantity is more than or equal to into the identification information of m/2, really
It is set to t-th script.Here, when exceeding half for identification information identical quantity, the method will the identification information
It is defined as the script of the examination question, to improve the order of accuarcy of script determination, reduces examination question erroneous judgement degree.
Specifically, the method can also carry out script modification, correction, after script is obtained, staff
The script of automatic identification can be checked whether correct, if there is wrong script, the method can also receive editor
Instruction, edits script, to change script, improves the degree of accuracy that examination question is corrected, read and appraised.
Meanwhile, the method can also correct examination question quantity, such as it is determined that effectively after answer region, according to every part of answering pen
The number in effective answer region of mark information, corrects examination question quantity n, here, effective answer region of each answer handwriting information
Number be identical, correct or adjust examination question quantity n with this, the degree of accuracy is high.
Second aspect, the embodiment of the present invention provides a kind of paper answer region recognition and examination question reads and appraises system, with reference to Fig. 3,
The system determines that subsystem 2, effective answer region determine subsystem 3 including answer person's handwriting information acquisition subsystem 1, examination question quantity
Subsystem 4 is read and appraised with answer handwriting information, answer person's handwriting information acquisition subsystem 1 is used to obtain m part answer handwriting informations；Examination question
Quantity determines subsystem 2 for according to the spacing distance of person's handwriting in every part of answer handwriting information, answer region being divided successively, is formed
Examination question magnitude-set C and answer regional ensemble R, wherein, C={ C1 C2 … Ci … Cm, CiFor i-th part of answer handwriting information
Answer region quantity, i=1,2 ..., m,Ri,jFor i-th part of answer person's handwriting
J-th answer region of information, j=1,2 ..., Ci；Maximum answer region quantity in examination question magnitude-set C, it is determined that examination
Topic quantity n；Effectively answer region determines that subsystem 3 is used for answer region Ri,jWith each answer of kth part answer handwriting information
Region carries out similarity test, determines k-th adaptive value arrayWith k-th adaptive value
Maximum adaptive value V in arraykmax, wherein, Vk,jFor answer region Ri,jWith j-th answer area of kth part answer handwriting information
The adaptive value in domain, k=1,2 ..., m, and k ≠ i；According to the maximum adaptive value in each adaptive value array, it is determined that final adaptive valueFinal adaptive valueFor answer region Ri,jAdaptive value；By final adaptive valueCompare with given threshold：If final suitable
Property valueLess than given threshold, then answer region R is judgedi,jFor invalid answer region, if final adaptive valueMore than or equal to setting
Determine threshold value, then judge answer region Ri,jFor effective answer region；Answer handwriting information reads and appraises subsystem 4 to be used for every part of answer
T-th effective answer region of handwriting information is contrasted with set information, and Statistical Comparison result：If identical with set information
Effectively the number in answer region is less than m/2, then information is generated, if the number in effective answer region identical with set information is big
In m/2 is equal to, then：T-th effective answer region of every part of answer handwriting information is identified, t-th effective answer is obtained
The identification information set D in regiont, wherein, Dt={ D1,t D2,t … Di,t … Dm,t, Di,tFor i-th part of answer handwriting information
The identification information in t-th answer region, t=1,2 ..., i ... n；To identification information set DtCounted, and determined t-th examination
Topic answer, t-th script is the answer of t-th effective answer region correspondence examination question of every part of answer handwriting information；According to
Each script is read and appraised to each effective answer region of every part of answer handwriting information, and obtains every part of answer person's handwriting letter
The examination question of breath reads and appraises result.
As shown from the above technical solution, the present embodiment paper answer region recognition and examination question read and appraise system, using big data
The mode of real-time statistics, for the examination question of different-format, obtains answer handwriting information, even if during non-typing script, this is
System also can precisely determine examination question quantity and identification answer region, and the answer region of a large amount of answer handwriting informations be carried out similar
Degree test, according to the final adaptive value in different answer regions, to answer region validity identification is carried out, it is determined that effectively answer area
Domain, is prevented from effective information loss, and advantageously reduces data processing complexity, without the need for using card-reading apparatus or card, lead
The running stores such as pen.
Meanwhile, the system is identified for the content in effective answer region, is commented with determining script, and carrying out examination question
Read, information recognition accuracy is high, script reliability, operation, examination examination question is greatly improved and corrects efficiency, contribute to teacher's realization
Test and explanation with hall, reduction is prepared lessons difficulty, shorten that student inscribes, examination question corrects time of judgement.
Therefore, the present embodiment paper answer region recognition and examination question read and appraise system, can automatically and accurately recognize that paper is answered
Topic region, determines examination question quantity, and grade papers objective item, saves the office hours of staff, improves efficiency.
Specifically, answer handwriting information reads and appraises subsystem 4 in the identification information set D for obtaining t-th effective answer regiont
When, it is identified specifically for t-th effective answer region to every part of answer handwriting information：
If recognizing the identification information set D for successfully obtaining t-th effective answer regiont,
If recognition failures, the answer handwriting information and recognition failures number of times of recognition failures is recorded, and according to pre-seting
Character recognition order, is identified to t-th effective answer region of the answer handwriting information of recognition failures, until obtaining t
The identification information set D in individual effective answer regiont, or recognition failures number of times reaches maximum.Here, the system can be directed to hand
Write inaccurate character and priority is set, such as character " H " and " A ", " 0 " and " D ", when hand-written inaccurate, can increase at the system
Reason difficulty, the answer handwriting information reads and appraises subsystem 4 using the character recognition order for pre-seting, will character " A " and " D " it is excellent
First level shifts to an earlier date, by the priority of character " H " and " 0 " rearward, abnormal to reduce information identification, improves information recognition accuracy.
Specifically, effective answer region determines that subsystem 3 determines final adaptive valueWhen, specifically for suitable according to each
Property value array in maximum adaptive value, by equation below, it is determined that final adaptive value Wherein, VkmaxFor the maximum adaptive value in k-th adaptive value array, k=1,2 ..., m, and k ≠ i.
Here, the calculation rough draft of the non-effective answering information in for answer handwriting information, such as student, effective answer region determines son
System 3 is contrasted in the answer region of a large amount of answer handwriting informations, and asks in answer handwriting information the flat of maximum adaptive value
Average, data reference amount is big, and subsequently to carry out answer region availability deciding Informational support is provided, and is conducive to improving effective answer
The degree of accuracy of regional determination.
Specifically, answer handwriting information reads and appraises subsystem 4 it is determined that during script, specifically for identification information set
DtCounted, obtained the answer region quantity with same identification information；By identification of the answer region quantity more than or equal to m/2
Information, is defined as t-th script.Here, when exceeding half for identification information identical quantity, the answer handwriting information
Read and appraise subsystem 4 will the identification information be defined as the script of the examination question, to improve the order of accuarcy of script determination,
Reduce examination question erroneous judgement degree.
Finally it should be noted that：Various embodiments above only to illustrate technical scheme, rather than a limitation；To the greatest extent
Pipe has been described in detail with reference to foregoing embodiments to the present invention, it will be understood by those within the art that：Its according to
So the technical scheme described in foregoing embodiments can be modified, either which part or all technical characteristic are entered
Row equivalent；And these modifications or replacement, do not make the essence disengaging various embodiments of the present invention technology of appropriate technical solution
The scope of scheme, it all should cover in the middle of the claim of the present invention and the scope of specification.
Claims (10)
1. a kind of paper answer region recognition and examination question read and appraise method, it is characterised in that include：
Obtain objective item purpose answering information and model answer；
The answering information is contrasted with the model answer, the objective item purpose examination question is obtained and is read and appraised result.
2. a kind of paper answer region recognition and examination question read and appraise method, it is characterised in that include：
Answer person's handwriting information acquiring step：Obtain m part answer handwriting informations；
Examination question quantity determines step：According to the spacing distance of person's handwriting in every part of answer handwriting information, answer region, shape are divided successively
Into examination question magnitude-set C and answer regional ensemble R, wherein,
C={ C1 C2 … Ci … Cm, CiFor the answer region quantity of i-th part of answer handwriting information, i=1,2 ..., m,
Ri,jFor j-th answer region of i-th part of answer handwriting information, j
=1,2 ..., Ci；
Maximum answer region quantity in the examination question magnitude-set C, determines examination question quantity n；
Effective answer area determination step：By the answer region Ri,jEnter with each answer region of kth part answer handwriting information
Row similarity is tested, and determines k-th adaptive value arrayWith k-th adaptive value
Maximum adaptive value V in arraykmax, wherein, Vk,jFor the answer region Ri,jWith the jth of kth part answer handwriting information
The adaptive value in individual answer region, k=1,2 ..., m, and k ≠ i；
According to the maximum adaptive value in each adaptive value array, it is determined that final adaptive valueThe final adaptive valueFor described
Answer region Ri,jAdaptive value；
By the final adaptive valueCompare with given threshold：
If the final adaptive valueLess than the given threshold, then the answer region R is judgedi,jFor invalid answer region,
If the final adaptive valueMore than or equal to the given threshold, then the answer region R is judgedi,jFor effective answer area
Domain；
Answer handwriting information reads and appraises step：T-th of every part of answer handwriting information effective answer region is carried out with set information
Contrast, and Statistical Comparison result：
If the number in effective answer region identical with the set information be less than m/2, generate information, if with the setting
The number in the identical effective answer region of information is more than or equal to m/2, then：
T-th effective answer region of every part of answer handwriting information is identified, t-th effective answer area is obtained
The identification information set D in domaint, wherein, Dt={ D1,t D2,t … Di,t … Dm,t, Di,tFor the of i-th part of answer handwriting information
The identification information in t answer region, t=1,2 ..., i ... n；
To the identification information set DtCounted, and determined t-th script, t-th script is described every
The answer of t-th effective answer region correspondence examination question of part answer handwriting information；
Each effective answer region of every part of answer handwriting information is read and appraised according to each script, and obtains described every
The examination question of part answer handwriting information reads and appraises result.
3. according to claim 2 paper answer region recognition and examination question read and appraise method, it is characterised in that described that per part is answered
T-th effective answer region of topic handwriting information is identified, and obtains the identification information set D in t-th effective answer regiont,
Specifically include：
T-th effective answer region of every part of answer handwriting information is identified：
If recognizing successfully, the identification information set D in described t-th effective answer region is obtainedt,
If recognition failures, the answer handwriting information and recognition failures number of times of recognition failures is recorded, and according to the character for pre-seting
Recognition sequence, is identified to t-th effective answer region of the answer handwriting information of the recognition failures, until obtaining described
The identification information set D in t-th effective answer regiont, or the recognition failures number of times reaches maximum.
4. according to claim 2 paper answer region recognition and examination question read and appraise method, it is characterised in that described according to each
Maximum adaptive value in adaptive value array, it is determined that final adaptive valueSpecifically include：It is described according in each adaptive value array
Maximum adaptive value, by equation below, determine the final adaptive value
Wherein, VkmaxFor in k-th adaptive value array most
Big adaptive value, k=1,2 ..., m, and k ≠ i.
5. according to claim 2 paper answer region recognition and examination question read and appraise method, it is characterised in that to the identification letter
Breath set DtCounted, and determined t-th script, specifically included：To the identification information set DtCounted, obtained
There must be the answer region quantity of same identification information；
By identification information of the answer region quantity more than or equal to m/2, it is defined as t-th script.
6. paper answer region recognition and examination question read and appraise method according to claim 2～5 any one, it is characterised in that
The method also includes：Edit instruction is received, the script is edited.
7. a kind of paper answer region recognition and examination question read and appraise system, it is characterised in that include：
Answer person's handwriting information acquisition subsystem：For obtaining m part answer handwriting informations；
Examination question quantity determines subsystem：For according to the spacing distance of person's handwriting in every part of answer handwriting information, answer being divided successively
Region, forms examination question magnitude-set C and answer regional ensemble R, wherein, C={ C1 C2 … Ci … Cm, CiFor i-th part of answer
The answer region quantity of handwriting information,Ri,jFor i-th
J-th answer region of part answer handwriting information, j=1,2 ..., Ci；According to the maximum answer in the examination question magnitude-set C
Region quantity, determines examination question quantity n；
Effectively answer region determines subsystem：For by the answer region Ri,jWith each answer of kth part answer handwriting information
Region carries out similarity test, determines k-th adaptive value arrayIt is suitable with described k-th
Maximum adaptive value V in property value arraykmax, wherein, Vk,jFor the answer region Ri,jWith kth part answer handwriting information
The adaptive value in j-th answer region, k=1,2 ..., m, and k ≠ i；According to the maximum adaptive value in each adaptive value array, really
Fixed final adaptive valueThe final adaptive valueFor the answer region Ri,jAdaptive value；By the final adaptive valueCompare with given threshold：If the final adaptive valueLess than the given threshold, then the answer region R is judgedi,jFor
Invalid answer region, if the final adaptive valueMore than or equal to the given threshold, then the answer region R is judgedi,jFor
Effective answer region；
Answer handwriting information reads and appraises subsystem：For t-th of every part of answer handwriting information effective answer region to be believed with setting
Breath is contrasted, and Statistical Comparison result：If the number in effective answer region identical with the set information is less than m/2, give birth to
Into information, if the number in effective answer region identical with the set information is more than or equal to m/2, to every part of answer
T-th effective answer region of handwriting information is identified, and obtains the identification information set in t-th effective answer region
Dt, wherein, Dt={ D1,t D2,t … Di,t … Dm,t, Di,tFor the knowledge in t-th answer region of i-th part of answer handwriting information
Other information, t=1,2 ..., i ... n；To the identification information set DtCounted, and determined t-th script, described
T script is the answer of t-th effective answer region correspondence examination question of every part of answer handwriting information；According to each examination
Topic answer is read and appraised to each effective answer region of every part of answer handwriting information, and obtains every part of answer handwriting information
Examination question read and appraise result.
8. according to claim 7 paper answer region recognition and examination question read and appraise system, it is characterised in that the answer person's handwriting
Information is read and appraised subsystem and is identified in t-th effective answer region to every part of answer handwriting information, is obtained t-th and is effectively answered
The identification information set D in topic regiontWhen, specifically for：T-th effective answer region of every part of answer handwriting information is entered
Row identification：If recognizing successfully, the identification information set D in described t-th effective answer region is obtainedtIf recognition failures are remembered
The answer handwriting information and recognition failures number of times of record recognition failures, and according to the character recognition order for pre-seting, to the identification
T-th effective answer region of the answer handwriting information of failure is identified, until obtaining t-th effective answer region
Identification information set Dt, or the recognition failures number of times reaches maximum.
9. according to claim 7 paper answer region recognition and examination question read and appraise system, it is characterised in that effective answer
Region determines subsystem in the maximum adaptive value in each adaptive value array, it is determined that final adaptive valueWhen, it is concrete to use
In：The maximum adaptive value in each adaptive value array, by equation below, determines the final adaptive value Wherein, VkmaxFit for the maximum in k-th adaptive value array
Property value, k=1,2 ..., m, and k ≠ i.
10. according to claim 7 paper answer region recognition and examination question read and appraise system, it is characterised in that the answering pen
Mark information reads and appraises subsystem to identification information set DtWhen being counted, and being determined t-th script, specifically for：It is right
The identification information set DtCounted, obtained the answer region quantity with same identification information；By the answer region
Identification information of the quantity more than or equal to m/2, is defined as t-th script.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611026941.8A CN106599887B (en) | 2016-11-16 | 2016-11-16 | Paper answer region recognition and examination question read and appraise method, system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611026941.8A CN106599887B (en) | 2016-11-16 | 2016-11-16 | Paper answer region recognition and examination question read and appraise method, system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106599887A true CN106599887A (en) | 2017-04-26 |
| CN106599887B CN106599887B (en) | 2019-10-18 |
Family
ID=58591783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611026941.8A Active CN106599887B (en) | 2016-11-16 | 2016-11-16 | Paper answer region recognition and examination question read and appraise method, system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106599887B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107346418A (en) * | 2017-06-13 | 2017-11-14 | 卢宏洲 | A kind of answer recognition methods of real-time synchronization and efficient identification |
| CN109147444A (en) * | 2018-09-27 | 2019-01-04 | 广东小天才科技有限公司 | Learning condition feedback method and intelligent desk lamp |
| CN109146404A (en) * | 2018-07-16 | 2019-01-04 | 付坤元 | A kind of machine examination data processing method and system based on block chain technology |
| CN109669619A (en) * | 2019-01-01 | 2019-04-23 | 王思源 | A kind of with no paper answer method based on touch type intelligent terminal |
| CN110399838A (en) * | 2019-07-25 | 2019-11-01 | 联想(北京)有限公司 | Processing method and electronic equipment |
| CN110544400A (en) * | 2019-08-06 | 2019-12-06 | 安徽学云教育科技有限公司 | Automatic marking method for blank filling questions based on image recognition |
| CN112308177A (en) * | 2020-10-21 | 2021-02-02 | 泰州锐比特智能科技有限公司 | Test paper other person writing-replacing identification platform |
| CN113076872A (en) * | 2021-04-01 | 2021-07-06 | 杭州云梯科技有限公司 | Intelligent test paper correcting method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239850A (en) * | 2014-07-16 | 2014-12-24 | 华南理工大学 | Answer sheet filling information automatic recognition method based on comprehensive features of images |
| CN104866847A (en) * | 2015-05-20 | 2015-08-26 | 李宇 | Rapid and robust answering card identifying method |
| US20150317907A1 (en) * | 2014-04-30 | 2015-11-05 | Way On Ho | Computer-aided integrated examination and assignment system and method |
| CN105590101A (en) * | 2015-12-28 | 2016-05-18 | 杭州淳敏软件技术有限公司 | Hand-written answer sheet automatic processing and marking method and system based on mobile phone photographing |
-
2016
- 2016-11-16 CN CN201611026941.8A patent/CN106599887B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150317907A1 (en) * | 2014-04-30 | 2015-11-05 | Way On Ho | Computer-aided integrated examination and assignment system and method |
| CN104239850A (en) * | 2014-07-16 | 2014-12-24 | 华南理工大学 | Answer sheet filling information automatic recognition method based on comprehensive features of images |
| CN104866847A (en) * | 2015-05-20 | 2015-08-26 | 李宇 | Rapid and robust answering card identifying method |
| CN105590101A (en) * | 2015-12-28 | 2016-05-18 | 杭州淳敏软件技术有限公司 | Hand-written answer sheet automatic processing and marking method and system based on mobile phone photographing |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107346418A (en) * | 2017-06-13 | 2017-11-14 | 卢宏洲 | A kind of answer recognition methods of real-time synchronization and efficient identification |
| CN109146404A (en) * | 2018-07-16 | 2019-01-04 | 付坤元 | A kind of machine examination data processing method and system based on block chain technology |
| CN109147444A (en) * | 2018-09-27 | 2019-01-04 | 广东小天才科技有限公司 | Learning condition feedback method and intelligent desk lamp |
| CN109669619A (en) * | 2019-01-01 | 2019-04-23 | 王思源 | A kind of with no paper answer method based on touch type intelligent terminal |
| CN110399838A (en) * | 2019-07-25 | 2019-11-01 | 联想(北京)有限公司 | Processing method and electronic equipment |
| CN110544400A (en) * | 2019-08-06 | 2019-12-06 | 安徽学云教育科技有限公司 | Automatic marking method for blank filling questions based on image recognition |
| CN112308177A (en) * | 2020-10-21 | 2021-02-02 | 泰州锐比特智能科技有限公司 | Test paper other person writing-replacing identification platform |
| CN113076872A (en) * | 2021-04-01 | 2021-07-06 | 杭州云梯科技有限公司 | Intelligent test paper correcting method |
| CN113076872B (en) * | 2021-04-01 | 2022-07-15 | 杭州云梯科技有限公司 | Intelligent test paper correcting method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106599887B (en) | 2019-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106599887B (en) | Paper answer region recognition and examination question read and appraise method, system | |
| CN108932508B (en) | Method and system for intelligently identifying and correcting subjects | |
| CN110110581B (en) | Test paper correcting method and system based on artificial intelligence | |
| CN110659584B (en) | Intelligent mark-remaining paper marking system based on image recognition | |
| CN110008933A (en) | A kind of universal intelligent marking system and method | |
| CN108537218B (en) | Answer sheet identification processing method and device | |
| CN110956138B (en) | Auxiliary learning method based on home education equipment and home education equipment | |
| CN103310082A (en) | Paper inspection method and device | |
| CN109242856A (en) | A kind of papery operation electronization reads and makes comments method and device | |
| CN107784264A (en) | Test paper analysis method and system based on image procossing | |
| CN106874508A (en) | Test paper generation and method to go over files based on gridding image procossing | |
| CN101866415A (en) | Answer sheet recognition device and method of computer marking system | |
| CN107038438A (en) | It is a kind of that method is read and appraised based on image recognition | |
| CN108388895A (en) | A kind of paper answering card automatic processing method based on machine learning | |
| CN102081742B (en) | Method for automatically evaluating writing ability | |
| CN110909035A (en) | Personalized review question set generation method and device, electronic equipment and storage medium | |
| CN109902683A (en) | Paper core point and Data Input method and system | |
| CN112381099A (en) | Question recording system based on digital education resources | |
| CN113159014A (en) | Objective question reading method, device, equipment and storage medium based on handwritten question numbers | |
| CN112785205A (en) | Intelligent teaching comprehensive analysis system based on education big data | |
| CN111008594B (en) | Error-correction question review method, related device and readable storage medium | |
| CN108428073A (en) | A kind of intelligent evaluation system for teachers ' teaching quality | |
| CN113793538B (en) | Method and system for collecting teaching assistance operation wrong questions | |
| JP2008020506A (en) | Image processor and image processing program | |
| CN108681713A (en) | A kind of system for teaching quality evaluation for teachers |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 705, Building 5, Information Road, Songshan Lake Park, Dongguan City, Guangdong Province Applicant after: Guangdong Mozi Education Technology Co., Ltd. Address before: Room 109, 1st floor, No. 5 Information Road, Songshan Lake High-tech Industrial Development Zone, Dongguan City, Guangdong Province Applicant before: Dongguan excellent Electronic Technology Co., Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |