CN113505787A - Title correction method and system, adopted electronic equipment and computer readable medium - Google Patents
Title correction method and system, adopted electronic equipment and computer readable medium Download PDFInfo
- Publication number
- CN113505787A CN113505787A CN202110707925.XA CN202110707925A CN113505787A CN 113505787 A CN113505787 A CN 113505787A CN 202110707925 A CN202110707925 A CN 202110707925A CN 113505787 A CN113505787 A CN 113505787A
- Authority
- CN
- China
- Prior art keywords
- question
- type
- user
- questions
- 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
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000012937 correction Methods 0.000 title abstract description 24
- 230000004044 response Effects 0.000 claims description 24
- 238000013473 artificial intelligence Methods 0.000 claims description 9
- 238000013145 classification model Methods 0.000 claims description 4
- 238000004513 sizing Methods 0.000 claims description 4
- 230000001502 supplementing effect Effects 0.000 claims description 4
- 238000012986 modification Methods 0.000 claims description 3
- 230000004048 modification Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000002715 modification method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Tourism & Hospitality (AREA)
- Computing Systems (AREA)
- Educational Administration (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Educational Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Character Discrimination (AREA)
Abstract
A title correction method and system, an electronic device and a computer readable medium are provided. The title correction method comprises the following steps: performing character recognition on the picture to be corrected, and distinguishing a question stem part and a user answering part based on the question structure type; searching a standard answer based on the question type and the question stem part; and correcting the user answering part based on the obtained standard answers. The invention can automatically realize the identification of complex question types, so that the machine can flexibly distinguish the mutual position relation and the internal logic of the printed characters and the handwritten characters in various question types, avoid the wrong arrangement of words and sentences and the disordered semantic logic caused by the incapability of identifying the question type structure, and improve the accuracy of correction.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence, particularly relates to the technical field of artificial intelligence auxiliary image recognition and text processing, and more particularly relates to a subject correction method and system, and electronic equipment and a computer readable medium adopting the same.
Background
With the technical progress, the current students can search answers on the whole network by shooting the difficult problems which cannot be solved through a mobile phone when encountering the difficult problems, or test questions handwritten by pencils or ball pens are uploaded by shooting and corrected by a machine, so that the search for knowledge is greatly facilitated, and the shortage of teachers and materials is made up.
In the hand-written test questions which can be corrected by the machine, the selection questions, the blank filling questions and the judgment questions occupy a larger share, but the blank filling questions also have various variant forms, such as Chinese character writing questions by reading pinyin with an upper and lower solution, idioms connecting with one another with a grid structure, bubble questions with changed appearances, and the like.
Disclosure of Invention
In view of the above, the present invention is directed to a topic modification method and system, and an electronic device and a computer readable medium using the same, which are intended to at least partially solve at least one of the above technical problems.
In order to achieve the above object, as a first aspect of the present invention, there is provided a title revising method, comprising the steps of:
performing character recognition on the picture to be corrected, and distinguishing a question stem part and a user answering part based on the question structure type;
searching a standard answer based on the question type and the question stem part;
and correcting the user answering part based on the obtained standard answers.
Optionally, the method further comprises: acquiring a question type; and acquiring a corresponding topic structure type according to the topic type.
The step of obtaining the title type is realized by identifying key words describing the title type in the title of the picture to be corrected;
optionally, the keywords describing the topic type in the topic include: selecting questions, selecting questions singly, selecting multiple questions, filling in blank, judging questions, applying questions, completing the shape and filling in blank, filling in, selecting, supplementing, perfecting, comparing, sizing and answering;
optionally, the question types include a blank filling question, a selection question, a short answer question, a line connection question, a Chinese character writing question according to pinyin and a judgment question.
Optionally, the topic structure type comprises at least one of:
for a filling-in-the-blank topic, the topic structure type includes: the above, user response part and the below;
for choice questions, the question structure types include: the above, user response part, the below and options;
for judgment questions, the question structure types include: a question stem part and a user answering part;
for short-response questions, the question structure types include: a question stem part, a question and a user answering part aiming at the question;
for the question types written according to the pinyin, the question structure types comprise: the pinyin in the first position area, the pinyin above the second position area, the user answering part and the pinyin below the second position area, wherein the position relationship between the first position area and the second position area comprises up-down arrangement or left-right arrangement;
for a topic, the topic structure types include: the first item located in the third position area, the second item located in the fourth position area and the start end coordinate and the end coordinate of the connecting line answered by the user, wherein the position relation of the third position area and the fourth position area comprises vertical arrangement or horizontal arrangement.
Optionally, the step of distinguishing the question stem part and the user answer part based on the question structure type includes:
identifying printed characters and handwritten characters in the picture to be corrected, and splicing the printed characters and the handwritten characters into a question stem part and a user answering part according to position information of the printed characters and the handwritten characters, symbols representing a user answering area and a question structure type;
optionally, the step of recognizing the printed text and the handwritten text is implemented by a pre-trained artificial intelligence classification model;
optionally, the symbols characterizing the user response area include underline, median line, parentheses, square, and circle.
Optionally, in the step of splicing the question stem part and the user answering part, the single handwritten characters are spliced into a plurality of user answering parts according to the position information, and then the single printed characters are spliced into the rest parts except the user answering part in the corresponding question structure type according to the position information, the special question structure and the position information of the handwritten characters.
Optionally, the special question type structure refers to some question type structures with special arrangement formats stored in the system in advance, and includes a pinyin phonetic notation question, a sudoku question, a idiom connection question, a bubble question and a connection question.
Optionally, the method further comprises determining the size of the user response area based on the position information of the recognized printed and handwritten characters and/or the position information of the symbols representing the user response area;
optionally, the method further includes the step of judging whether the recognized handwritten characters located in the user answering area are redundant characters or characters answered by the user by using the trained artificial intelligence model.
As a second aspect of the present invention, there is also provided a title correction system, including:
the identification distinguishing unit is used for carrying out character identification on the picture to be corrected and distinguishing the question stem part and the user answering part based on the question structure type;
the retrieval unit is used for retrieving in a knowledge base to obtain a standard answer based on the question type and the question stem part;
and the correcting unit is used for correcting the user answering part based on the standard answer.
As a third aspect of the present invention, there is also provided an electronic device including a processor and a memory, the memory storing a computer-executable program, the processor executing the title correction method as described above when the computer-executable program is executed by the processor.
As a fourth aspect of the present invention, there is also provided a computer-readable medium storing a computer-executable program which, when executed, implements the title correction method as described above.
Based on the above technical solution, the title correction method and system of the present invention have at least one of the following advantages compared with the prior art:
the invention can realize the recognition of different question types in the shot image and accurately distinguish the corresponding structures, thereby enabling a machine to flexibly distinguish the mutual position relation and the internal logic of the printed characters and the handwritten characters in various question types, avoiding the wrong arrangement of the characters and sentences and the disordered semantic logic caused by the incapability of recognizing the special question type structures and improving the accuracy of correction;
according to the method, through deep machine learning, the printing fonts and the handwriting fonts in the shot image can be accurately distinguished;
the method can particularly identify the logical position relation of the characters through the special question type structure, and splice the single characters into the context which accords with the semantics according to the special question type structure, thereby avoiding the influence of the special question type structure and improving the accuracy of correction;
the scheme of the invention has good expandability, and the application range of the method of the invention can be enlarged by adding a specific topic structure in the system in advance.
Drawings
FIG. 1 is a block flow diagram of a topic modification method of the present invention;
FIG. 2 is a schematic diagram of a topic modification system according to the present invention;
FIG. 3 is a schematic diagram of the structure of the knowledge base of the present invention;
FIG. 4 is a schematic diagram of the electronic device of the present invention;
FIG. 5 is a schematic diagram of a computer readable medium of the present invention;
fig. 6-8 are photographs of some embodiments of the invention.
Detailed Description
In describing particular embodiments, specific details of structures, properties, effects, or other features are set forth in order to provide a thorough understanding of the embodiments by one skilled in the art. However, it is not excluded that a person skilled in the art may implement the invention in a specific case without the above-described structures, performances, effects or other features.
The flow chart in the drawings is only an exemplary flow demonstration, and does not represent that all the contents, operations and steps in the flow chart are necessarily included in the scheme of the invention, nor does it represent that the execution is necessarily performed in the order shown in the drawings. For example, some operations/steps in the flowcharts may be divided, some operations/steps may be combined or partially combined, and the like, and the execution order shown in the flowcharts may be changed according to actual situations without departing from the gist of the present invention.
The block diagrams in the figures generally represent functional entities and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and/or processing unit devices and/or microcontroller devices.
The same reference numerals denote the same or similar elements, components, or parts throughout the drawings, and thus, a repetitive description thereof may be omitted hereinafter. It will be further understood that, although the terms first, second, third, etc. may be used herein to describe various elements, components, or sections, these elements, components, or sections should not be limited by these terms. That is, these phrases are used only to distinguish one from another. For example, a first device may also be referred to as a second device without departing from the spirit of the present invention. Furthermore, the term "and/or", "and/or" is intended to include all combinations of any one or more of the listed items.
The invention is provided aiming at the technical problem that the existing machine is difficult to flexibly recognize the specific arrangement and meaning of characters from the paper surface, in particular to the technical problem that the meanings represented by the position relations of the characters are different when various question type structures are different, and the invention can understand the specific arrangement and the meanings behind the characters on the paper surface through deconstructing the various question type structures and accurately and efficiently complete the correction task.
As shown in FIG. 1, the present invention provides a title correction method, comprising the following steps:
performing character recognition on the picture to be corrected, and distinguishing a question stem part and a user answering part based on the question structure type;
searching a standard answer based on the question type and the question stem part;
and correcting the user answering part based on the obtained standard answers.
The method further comprises the steps of obtaining the theme type and obtaining the corresponding theme structure type according to the theme type. The step of obtaining the title type is realized by identifying key words describing the title type in the title of the picture to be corrected; optionally, the keywords describing the topic type in the topic include, for example: selecting questions, selecting questions singly, selecting multiple questions, filling in the blank, judging questions, applying questions, completing the shape and filling in the blank, filling in, selecting, supplementing, perfecting, comparing, sizing, answering and the like;
the question types comprise blank filling questions, selection questions, short answer questions, line connection questions, Chinese character writing questions according to pinyin, judgment questions and the like. Each topic type has a corresponding topic structure type, and the recognized characters can be structured by combining the corresponding topic structure types during topic recognition. The topic structure type includes at least one of:
for a filling-in-the-blank topic, the topic structure type includes: the above, user response part and the below;
for choice questions, the question structure types include: the above, user response part, the below and options;
for judgment questions, the question structure types include: a question stem part and a user answering part;
for short-response questions, the question structure types include: a question stem part, a question and a user answering part aiming at the question;
for the question types written according to the pinyin, the question structure types comprise: the pinyin in the first position area, the pinyin above the second position area, the user answering part and the pinyin below the second position area, wherein the position relationship between the first position area and the second position area comprises up-down arrangement or left-right arrangement;
for a topic, the topic structure types include: the first item located in the third position area, the second item located in the fourth position area and the start end coordinate and the end coordinate of the connecting line of the user answering part, and the position relation of the third position area and the fourth position area comprises vertical arrangement or horizontal arrangement. The line connection problem needs to calculate the coordinates of two end points of all lines, and records are paired.
For reading and solving the questions, the question structure types can have full texts, titles with the small questions, options of the small questions and a plurality of small questions, for example;
the title structure type also comprises the position relation or mapping relation of each structure, such as a complete form filling blank title, the title structure type comprises an original text, the blank in the original text needs to be identified, the position of the blank in the original text needs to be recorded, and meanwhile, an option is provided, and the option and the position in the title stem need to be mapped.
The step of distinguishing the question stem part and the user answering part based on the question structure type is to identify the printed characters and the handwritten characters in the picture to be corrected, and splice the printed characters and the handwritten characters into the question stem part and the user answering part corresponding to the question structure type according to the position information of the printed characters and the handwritten characters, the symbols representing the user answering area and the question structure type.
The step of recognizing the printed text and the handwritten text is realized by a pre-trained artificial intelligence classification model, such as a convolutional neural network model (CNN). For example, the characters may be classified and then recognized, wherein the method for recognizing the printed characters and/or the handwritten characters is, for example, a known OCR recognition method, to obtain a black-and-white contrast pattern of the pattern in the recognition area, and then matching and searching whether the same or similar pattern exists in the character library, or converting the same into a vector pattern and then judging whether the corresponding vector character library font is matched. Other known character recognition methods may be used to recognize printed and handwritten text.
The symbol representing the user response area refers to a symbol capable of indicating the user response area, and includes an underline, a median line, a bracket, a square frame, a circle and the like.
Some questions, such as blank filling questions and common selection questions, can be structured according to whether the characters are positioned on the same line under normal conditions, and generally do not have problems when the structural type of the questions is not designed, but some special question types, such as pinyin phonetic notation questions, sudoku questions, idiom connection questions, bubble questions, connection questions and the like, can not accurately identify related information only according to the line structure (whether the characters are positioned on the same line). Therefore, in some embodiments, it is preferable to design only some special topic structures, and to determine according to the special topic structures when identifying the special topic structures, and to determine according to the row structures when not having the special topic structures.
Specifically, when a computer takes a question, the question is firstly disassembled to obtain a line structure and a question type of the question; for example, selecting questions, a whole line of printed characters can be spliced together, and the handwritten characters which are finally answered are spliced together; if the character is a special question type, judging whether the character is an answer or redundancy according to the position of the recognized handwritten character in combination with a special question type structure, and then splicing. The judgment is carried out according to the characteristic strategy if the special question type structure exists, and the judgment is carried out according to the line structure if the special question type structure does not exist.
In the step of splicing the question stem part and the user answering part, for example, the single handwritten words are spliced into a plurality of sequentially numbered user answering parts according to the position information, and then the single printed words are spliced according to the corresponding question structure types according to the position information, the special question structure and the position information of the handwritten words, so that the device knows which texts are question stems, which texts are answers and which texts are redundant.
Optionally, the special question type structure refers to some question type structures with special arrangement formats stored in the system in advance, and includes a pinyin phonetic notation question, a sudoku question, a idiom connection question, a bubble question, a connection question and the like. The existence of the special topic structure influences the step of arranging and splicing the single characters according to the character lines, so that the special topic structure needs to be specially treated. The types of special question type structures can be gradually supplemented and expanded according to the richness of the actual test questions.
The method further comprises the step of determining the size of the user answering area based on the position information of the recognized printed characters and handwritten characters and/or the position information of symbols representing the user answering area; specifically, the method determines the approximate area of the handwritten character according to the coordinates of the last printed character before the handwritten character and the coordinates of the first printed character after the handwritten character, and judges whether the obtained user answering area is correct by combining whether a symbol representing the user answering area exists in the approximate area, such as underlines or brackets for filling in blank questions, brackets for judging the questions and the like. In some cases, the user response area may be determined only according to the symbol representing the user response area, for example, when the symbol representing the user response area is a closed structure of a circle or a square.
For example, the eleventh topic in fig. 8, the right box shows five options, and if the topic structure is not designed, the machine cannot recognize the special topic structure, and the words arranged in an approximate line are spliced into a sentence according to the character line, so as to obtain, for example, "you, mike. "such a wrong pattern, thereby causing the machine to be unable to recognize and judge.
In the invention, the special question type structure which is designed in advance is combined during identification, and the right square box is identified according to five options, so that'd. "do not match" you, Mike? "splicing according to character lines.
Wherein, the title type is preferably a blank filling title or a judgment title. The method of the present invention is initially proposed based on a blank question and a judgment question, but the method of the present invention is not limited thereto, and can be applied to various question types including a choice question, a simple answer question, and the like.
The method searches standard answers in a knowledge base based on the question types and the spliced question stem parts. The knowledge base used here may be, for example, a knowledge base originally created by the company, and is constructed by the following method: storing the knowledge points of minimum granularity in a single entry forms the knowledge base.
The step of searching for the standard answer based on the question type and the question stem part can also be implemented by other ways, such as directly searching for a pre-established answer database, searching for other forms of knowledge bases, and the like.
In the step of correcting the user's answer portion based on the obtained standard answer, a correction mark is added to a position adjacent to the handwritten character recognized as "answer character", for example, the correction mark is usually added to the latter side of the handwritten character, for example, "√" or "x". The correction mark has a color different from that of the text part, such as red or green, so as to avoid confusion with black or blue handwritten text which is shot or scanned and input.
As shown in FIG. 2, the present invention also discloses a topic modification system, which comprises:
the identification distinguishing unit is used for carrying out character identification on the picture to be corrected and distinguishing the question stem part and the user answering part based on the question structure type;
the retrieval unit is used for retrieving in a knowledge base and/or an answer database to obtain a standard answer based on the question type and the question stem part;
and the correcting unit is used for correcting the user answering part based on the standard answer.
And the identification and distinguishing unit also executes the operation of acquiring the topic type and acquiring the corresponding topic structure type according to the topic type. The operation of obtaining the title type is realized by identifying key words describing the title type in the title of the picture to be corrected; optionally, the keywords describing the topic type in the topic include, for example: selecting questions, selecting questions singly, selecting questions more, filling in the blank, judging questions, applying questions, completing the shape and filling in the blank, filling in, selecting, supplementing, perfecting, comparing, sizing, answering and the like.
The question types comprise blank filling questions, selection questions, short answer questions, line connection questions, Chinese character writing questions according to pinyin, judgment questions and the like. Each topic type has a corresponding topic structure type, and the recognized characters can be structured by combining the corresponding topic structure types during topic recognition.
The title structure type also comprises the position relation or mapping relation of each structure, such as a complete form filling blank title, the title structure type comprises an original text, the blank in the original text needs to be identified, the position of the blank in the original text needs to be recorded, and meanwhile, an option is provided, and the option and the position in the title stem need to be mapped.
The step of distinguishing the question stem part and the user answering part based on the question structure type is to identify the printed characters and the handwritten characters in the picture to be corrected, and splice the printed characters and the handwritten characters into the question stem part and the user answering part corresponding to the question structure type according to the position information of the printed characters and the handwritten characters, the symbols representing the user answering area and the question structure type.
The step of recognizing the printed text and the handwritten text is realized by a pre-trained artificial intelligence classification model, such as a convolutional neural network model (CNN). For example, the characters may be classified and then recognized, wherein the method for recognizing the printed characters and/or the handwritten characters is, for example, a known OCR recognition method, to obtain a black-and-white contrast pattern of the pattern in the recognition area, and then matching and searching whether the same or similar pattern exists in the character library, or converting the same into a vector pattern and then judging whether the corresponding vector character library font is matched. Other known character recognition methods may be used to recognize printed and handwritten text.
The symbol representing the user response area refers to a symbol capable of indicating the user response area, and includes an underline, a median line, a bracket, a square frame, a circle and the like.
In the step of splicing the question stem part and the user answering part, for example, the single handwritten characters can be spliced into a plurality of sequentially numbered user answering parts according to the position information, and then the single printed characters are spliced into the rest parts except the user answering part in the corresponding question structure type according to the position information, the special question structure and the position information of the handwritten characters.
Optionally, the special question type structure refers to some question type structures with special arrangement formats stored in the system in advance, and includes a pinyin phonetic notation question, a sudoku question, a idiom connection question, a bubble question, a connection question and the like. The existence of the special topic structure influences the step of arranging and splicing the single characters according to the character lines, so that the special topic structure needs to be specially treated. The types of special question type structures can be gradually supplemented and expanded according to the richness of the actual test questions.
The method further comprises the step of determining the size of the user answering area based on the position information of the recognized printed characters and handwritten characters and/or the position information of symbols representing the user answering area; specifically, the method determines the approximate area of the handwritten character according to the coordinates of the last printed character before the handwritten character and the coordinates of the first printed character after the handwritten character, and judges whether the obtained user answering area is correct by combining whether a symbol representing the user answering area exists in the approximate area, such as underlines or brackets for filling in blank questions, brackets for judging the questions and the like.
Wherein, the title type is preferably a blank filling title or a judgment title. The method of the present invention is initially proposed based on a blank question and a judgment question, but the method of the present invention is not limited thereto, and can be applied to various question types including a choice question, a simple answer question, and the like.
The method searches standard answers in a knowledge base based on the question types and the spliced question stem parts. The knowledge base used here may be, for example, a knowledge base originally created by the company, and is constructed by the following method: storing the knowledge points of minimum granularity in a single entry forms the knowledge base.
The step of searching for the standard answer based on the question type and the question stem part can also be implemented by other ways, such as directly searching for a pre-established answer database, searching for other forms of knowledge bases, and the like.
In the step of correcting the user's answer portion based on the obtained standard answer, a correction mark is added to a position adjacent to the handwritten character recognized as "answer character", for example, the correction mark is usually added to the latter side of the handwritten character, for example, "√" or "x". The correction mark has a color different from that of the text part, such as red or green, so as to avoid confusion with black or blue handwritten text which is shot or scanned and input.
The invention also discloses an electronic device comprising a processor and a memory for storing a computer executable program, wherein the processor performs the method as described above when the computer executable program is executed by the processor.
The electronic device may be embodied in the form of a general purpose computing device, for example. The number of the processors may be one, or may be multiple and work together. The invention also does not exclude that distributed processing is performed, i.e. the processors may be distributed over different physical devices. The electronic device of the present invention is not limited to a single entity, and may be a sum of a plurality of entity devices.
In which a memory stores a computer-executable program, typically machine-readable code, which is executable by the processor to enable an electronic device to perform the method of the invention, or at least some of the steps of the method.
The memory may include volatile memory, such as Random Access Memory (RAM) and/or cache memory, and may also be non-volatile memory, such as read-only memory (ROM).
Optionally, in this embodiment, the electronic device further includes an I/O interface, which is used for data exchange between the electronic device and an external device. The I/O interface may be a local bus representing one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, and/or a memory storage device using any of a variety of bus architectures.
Elements or components not shown in the above examples may also be included in the electronic device of the present invention. For example, some electronic devices further include a display unit such as a display screen, and some electronic devices further include a human-computer interaction element such as a button, a keyboard, and the like. Electronic devices are considered to be covered by the present invention as long as the electronic devices are capable of executing a computer-readable program in a memory to implement the method of the present invention or at least a part of the steps of the method.
The present invention also discloses a computer readable medium having a computer executable program stored thereon, wherein the computer executable program, when executed, implements a method as described above.
A computer readable storage medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable storage medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Python, Java, C + + or the like and conventional procedural programming languages, such as the C language, assembly language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
In order that the objects, technical solutions and advantages of the present invention will become more apparent, the present invention will be further described in detail with reference to the accompanying drawings in conjunction with the following specific embodiments. It should be noted that the following examples are only for illustrating the present invention and are not intended to limit the present invention.
Example 1
Fig. 3 is a schematic diagram of a data structure of the knowledge base of the present invention, and as shown in fig. 3, the knowledge base of this embodiment is constructed by the following method: storing the knowledge points of the smallest granularity in a single entry in the form of an assertion forms the knowledge base.
The mathematical knowledge points "triangle three interior angles sum is 180 °", "four eight three two" are exemplified in the figure, and as shown in fig. 3, it can be seen that each knowledge point is stored in the form of one text entry.
The title correction system of embodiment 1 can be installed on a mobile phone, and specifically includes:
the identification distinguishing unit is used for identifying characters of the picture to be corrected, distinguishing whether the characters are printed or handwritten characters through an artificial intelligence model, and identifying specific single characters through an OCR technology;
before or during the recognition, the picture to be corrected can be divided into different questions according to the question number, line spacing and/or the like of a single word of the whole page, for example, the questions with different question numbers framed by different square frames in fig. 8; judging the type of the question based on the identified keywords in the question, such as selecting the question, judging the question or filling in blank questions; then, according to the position information, the special question type structure and the position information of the hand-written characters, the single printed characters are identified and spliced according to the corresponding question structure types, so that the question stem part, answer characters answered by the user and redundant characters of other non-answer characters can be distinguished, such as the middle process of user calculation, labeling and the like.
For example, the gap-filling question can be further subdivided into a user answer part with an upper text and a lower text, while there is no context like a judgment question, and only a question stem part and an answer part answered by the user. The recognition distinguishing unit may further judge the position of the user's answering area based on the position information of the printed text and the handwritten text.
And the retrieval unit is used for retrieving in a knowledge base such as the knowledge base shown in the figure 3 of the invention to obtain a standard answer based on the question type and the question stem part. For a knowledge investigation type question, a retrieval process is to divide the question stem part (for example, a blank filling question comprises an upper part and a lower part) to obtain different real words, expand the real words and simultaneously input the expanded real words into a knowledge base for retrieval, when all the real words (key words) for retrieval fall into a certain knowledge item, a corresponding result is retrieved and compared with the knowledge item, if the real words for retrieval are blank filling questions or selection questions, the remaining key words which are not used for retrieval in the knowledge item are answers to be answered, if the real words for retrieval are judgment questions, whether the real words for retrieval are consistent with the description of the knowledge item or not is compared, if the real words for retrieval are consistent with the description of the knowledge item, the answers are yes, and if the real words for retrieval are inconsistent, the answers are no.
And a correction unit for correcting the answer portion of the user, determining the correction point position based on the question type structure or the handwritten answer character position, and correcting, for example, based on the question information (question type symbol of the square frame type) in the lower right corner of the square frame in fig. 7, and forming a green correction mark 'check mark' or 'x' at a position of about 1mm below the handwritten character based on the handwritten character in fig. 8.
Through the description of the above embodiment, those skilled in the art can easily understand that the specific logical relationship and meaning of the characters in different areas in the image to be recognized can be accurately recognized by subdividing the question structure type, and the accuracy of the machine for correcting the handwritten answer can be improved.
While the foregoing embodiments have described the objects, aspects and advantages of the present invention in further detail, it should be understood that the present invention is not inherently related to any particular computer, virtual machine or electronic device, and various general-purpose machines may be used to implement the present invention. The invention is not to be considered as limited to the specific embodiments thereof, but is to be understood as being modified in all respects, all changes and equivalents that come within the spirit and scope of the invention.
Claims (10)
1. A title approval method is characterized by comprising the following steps:
performing character recognition on the picture to be corrected, and distinguishing a question stem part and a user answering part based on the question structure type;
searching a standard answer based on the question type and the question stem part;
and correcting the user answering part based on the obtained standard answers.
2. The method of claim 1, further comprising: acquiring a question type; and acquiring a corresponding topic structure type according to the topic type.
3. The method according to claim 2, wherein the step of obtaining the title type is realized by identifying a keyword describing the title type in the title of the picture to be modified;
optionally, the keywords describing the topic type in the topic include: selecting questions, selecting questions singly, selecting multiple questions, filling in blank, judging questions, applying questions, completing the shape and filling in blank, filling in, selecting, supplementing, perfecting, comparing, sizing and answering;
optionally, the question types include a blank filling question, a selection question, a short answer question, a line connection question, a Chinese character writing question according to pinyin and a judgment question.
4. The method of any one of claims 1-3, wherein the topic structure type comprises at least one of:
for a filling-in-the-blank topic, the topic structure type includes: the above, user response part and the below;
for choice questions, the question structure types include: the above, user response part, the below and options;
for judgment questions, the question structure types include: a question stem part and a user answering part;
for short-response questions, the question structure types include: a question stem part, a question and a user answering part aiming at the question;
for the question types written according to the pinyin, the question structure types comprise: the pinyin in the first position area, the pinyin above the second position area, the user answering part and the pinyin below the second position area, wherein the position relationship between the first position area and the second position area comprises up-down arrangement or left-right arrangement;
for a topic, the topic structure types include: the first item located in the third position area, the second item located in the fourth position area and the start end coordinate and the end coordinate of the connecting line answered by the user, wherein the position relation of the third position area and the fourth position area comprises vertical arrangement or horizontal arrangement.
5. The method of claim 1, wherein the step of distinguishing the question stem part from the user response part based on the question structure type comprises:
identifying printed characters and handwritten characters in the picture to be corrected, and splicing the printed characters and the handwritten characters into a question stem part and a user answering part according to position information of the printed characters and the handwritten characters, symbols representing a user answering area and a question structure type;
optionally, the step of recognizing the printed text and the handwritten text is implemented by a pre-trained artificial intelligence classification model;
optionally, the symbols characterizing the user response area include underline, median line, parentheses, square, and circle.
6. The method according to claim 5, wherein in the step of splicing the question stem part and the user response part, the single handwritten text is spliced into a plurality of user response parts according to the position information, and then the single printed text is spliced into the rest parts except the user response part in the corresponding question structure type according to the position information, the special question type structure and the position information of the handwritten text;
optionally, the special question type structure refers to some question type structures with special arrangement formats stored in the system in advance, and includes a pinyin phonetic notation question, a sudoku question, a idiom connection question, a bubble question and a connection question.
7. The method according to claim 5 or 6,
the method further comprises determining the size of the user answering area based on the identified position information of the printed and handwritten characters and/or the position information of the symbols representing the user answering area;
optionally, the method further includes the step of judging whether the recognized handwritten characters located in the user answering area are redundant characters or characters answered by the user by using the trained artificial intelligence model.
8. A topic modification system, comprising:
the identification distinguishing unit is used for carrying out character identification on the picture to be corrected and distinguishing the question stem part and the user answering part based on the question structure type;
the retrieval unit is used for retrieving in a knowledge base and/or an answer database to obtain a standard answer based on the question type and the question stem part;
and the correcting unit is used for correcting the user answering part based on the standard answer.
9. An electronic device comprising a processor and a memory, the memory for storing a computer-executable program, characterized in that:
when the computer-executable program is executed by the processor, the processor performs the title wholesale method of any one of claims 1-7.
10. A computer-readable medium storing a computer-executable program, wherein the computer-executable program, when executed, implements the title revising method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110707925.XA CN113505787A (en) | 2021-06-24 | 2021-06-24 | Title correction method and system, adopted electronic equipment and computer readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110707925.XA CN113505787A (en) | 2021-06-24 | 2021-06-24 | Title correction method and system, adopted electronic equipment and computer readable medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113505787A true CN113505787A (en) | 2021-10-15 |
Family
ID=78010626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110707925.XA Pending CN113505787A (en) | 2021-06-24 | 2021-06-24 | Title correction method and system, adopted electronic equipment and computer readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113505787A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113900602A (en) * | 2021-12-09 | 2022-01-07 | 北京辰光融信技术有限公司 | Intelligent printing method and system for automatically eliminating target object filling information |
| CN114332900A (en) * | 2021-12-30 | 2022-04-12 | 科大讯飞股份有限公司 | Job correction method, device, equipment and storage medium |
| CN114550181A (en) * | 2022-02-10 | 2022-05-27 | 珠海读书郎软件科技有限公司 | Method, device and medium for identifying question |
| CN118609152A (en) * | 2024-08-08 | 2024-09-06 | 广州市小马知学技术有限公司 | Intelligent correction method and system for student homework |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108052687A (en) * | 2018-01-29 | 2018-05-18 | 赵宇航 | A kind of educational information search system based on internet |
| CN111597908A (en) * | 2020-04-22 | 2020-08-28 | 深圳中兴网信科技有限公司 | Test paper correcting method and test paper correcting device |
| CN112115736A (en) * | 2019-06-19 | 2020-12-22 | 广东小天才科技有限公司 | Job correction method and system based on image recognition and intelligent terminal |
-
2021
- 2021-06-24 CN CN202110707925.XA patent/CN113505787A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108052687A (en) * | 2018-01-29 | 2018-05-18 | 赵宇航 | A kind of educational information search system based on internet |
| CN112115736A (en) * | 2019-06-19 | 2020-12-22 | 广东小天才科技有限公司 | Job correction method and system based on image recognition and intelligent terminal |
| CN111597908A (en) * | 2020-04-22 | 2020-08-28 | 深圳中兴网信科技有限公司 | Test paper correcting method and test paper correcting device |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113900602A (en) * | 2021-12-09 | 2022-01-07 | 北京辰光融信技术有限公司 | Intelligent printing method and system for automatically eliminating target object filling information |
| CN114332900A (en) * | 2021-12-30 | 2022-04-12 | 科大讯飞股份有限公司 | Job correction method, device, equipment and storage medium |
| CN114550181A (en) * | 2022-02-10 | 2022-05-27 | 珠海读书郎软件科技有限公司 | Method, device and medium for identifying question |
| CN114550181B (en) * | 2022-02-10 | 2023-01-10 | 珠海读书郎软件科技有限公司 | Method, device and medium for identifying question |
| CN118609152A (en) * | 2024-08-08 | 2024-09-06 | 广州市小马知学技术有限公司 | Intelligent correction method and system for student homework |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113505787A (en) | Title correction method and system, adopted electronic equipment and computer readable medium | |
| CN107656922B (en) | Translation method, translation device, translation terminal and storage medium | |
| CN111753767A (en) | Method and device for automatically correcting operation, electronic equipment and storage medium | |
| CN111507330B (en) | Problem recognition method and device, electronic equipment and storage medium | |
| CN111626297A (en) | Character writing quality evaluation method and device, electronic equipment and recording medium | |
| CN111753120B (en) | Question searching method and device, electronic equipment and storage medium | |
| CN112434496B (en) | Method and terminal for identifying form data of bulletin document | |
| US12051256B2 (en) | Entry detection and recognition for custom forms | |
| US11562593B2 (en) | Constructing a computer-implemented semantic document | |
| CN113505786A (en) | Test question photographing and judging method and device and electronic equipment | |
| CN112149680A (en) | Wrong word detection and identification method and device, electronic equipment and storage medium | |
| CN106650720A (en) | Method, device and system for network marking based on character recognition technology | |
| CN116822634A (en) | Document visual language reasoning method based on layout perception prompt | |
| CN113569112A (en) | Tutoring strategy providing method, system, device and medium based on question | |
| CN113505195A (en) | Knowledge base, construction method and retrieval method thereof, and question setting method and system based on knowledge base | |
| US20230036812A1 (en) | Text Line Detection | |
| CN112084788A (en) | Automatic marking method and system for implicit emotional tendency of image captions | |
| CN116704508A (en) | Information processing method and device | |
| CN116225956A (en) | Automated testing method, apparatus, computer device and storage medium | |
| CN113392836A (en) | Redundant character recognition method, question correction method and system | |
| US20240078377A1 (en) | Completing typeset characters using handwritten strokes | |
| Taele et al. | A Geometric-based Sketch Recognition Approach for Handwritten Mandarin Phonetic Symbols I. | |
| WO2023170735A1 (en) | Learning method, estimation method, learning estimation system, and program | |
| JP5162622B2 (en) | Electronic ink processing | |
| CN117435708A (en) | Answering method, answering device, electronic equipment and storage medium |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230606 Address after: 6001, 6th Floor, No.1 Kaifeng Road, Shangdi Information Industry Base, Haidian District, Beijing, 100085 Applicant after: Beijing Baige Feichi Technology Co.,Ltd. Address before: 100085 4002, 4th floor, No.1 Kaifa Road, Shangdi Information Industry base, Haidian District, Beijing Applicant before: ZUOYEBANG EDUCATION TECHNOLOGY (BEIJING) CO.,LTD. |