WO2020135472A1 - 一种错题本生成方法及装置 - Google Patents

一种错题本生成方法及装置 Download PDF

Info

Publication number
WO2020135472A1
WO2020135472A1 PCT/CN2019/128160 CN2019128160W WO2020135472A1 WO 2020135472 A1 WO2020135472 A1 WO 2020135472A1 CN 2019128160 W CN2019128160 W CN 2019128160W WO 2020135472 A1 WO2020135472 A1 WO 2020135472A1
Authority
WO
WIPO (PCT)
Prior art keywords
wrong
question
wrong question
area
generating
Prior art date
Application number
PCT/CN2019/128160
Other languages
English (en)
French (fr)
Inventor
何涛
毛礼辉
罗欢
陈明权
Original Assignee
杭州大拿科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 杭州大拿科技股份有限公司 filed Critical 杭州大拿科技股份有限公司
Priority to US17/418,238 priority Critical patent/US11410407B2/en
Publication of WO2020135472A1 publication Critical patent/WO2020135472A1/zh

Links

Images

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
    • G06V10/225—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 based on a marking or identifier characterising the area
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/14—Image acquisition
    • G06V30/1444—Selective acquisition, locating or processing of specific regions, e.g. highlighted text, fiducial marks or predetermined fields
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G06F16/3329—Natural language query formulation or dialogue systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching

Definitions

  • the present invention relates to the technical field of teaching and information processing, and in particular to a method, device, electronic device, and computer-readable storage medium for generating a wrong book.
  • Wrong book helps students quickly grasp weak knowledge points, optimize learning path and review. At present, students usually copy the wrong questions into their notebooks by hand to create their own wrong question books. This method is less efficient for generating wrong questions, and students may be reluctant to spend time and energy to write wrong questions, which will affect students' learning and achievements.
  • An object of the present invention is to provide a method, device, electronic device, and computer-readable storage medium for generating a wrong book to solve the problem of low efficiency in generating a wrong book in the prior art.
  • the present invention provides a method for generating a wrong question book.
  • the method includes:
  • the area of the wrong question is stored in the wrong question database to generate a wrong question book.
  • the identifying, based on the pre-trained wrong question recognition model, the question whose correction result is incorrect in the revised test paper includes:
  • the pre-trained wrong question recognition model identify the correction marks in the area of each topic, and determine the topics with the correction marks as the preset marks to be the ones whose correction results are wrong.
  • the correction mark in the area for identifying each topic includes:
  • the method further includes:
  • the storing of the wrong question area in the wrong question database includes:
  • the method further includes:
  • characters in the answer area of the wrong question are recognized, and the recognized wrong answer is stored in the wrong question database and associated with the wrong question.
  • the method further includes:
  • At least one of the test paper ID, the cause of the error, and the assessment knowledge point corresponding to the wrong question is stored in the wrong question database and associated with the wrong question.
  • the method further includes: identifying the type of the wrong question; and storing the wrong question area in the wrong question database includes:
  • the area of the wrong question is stored in the corresponding question type group in the wrong question database.
  • the method further includes:
  • a question bank is searched for a question corresponding to the wrong question, a standard answer for the wrong question is obtained, and the obtained standard answer is stored in the wrong question database and associated with the wrong question.
  • the method further includes:
  • the method further includes:
  • the storing of the wrong question area in the wrong question database to generate the wrong question book includes:
  • the area of the wrong question is stored in the wrong question database and associated with at least one of the identified student number and name and the class, to establish the wrong book of the student.
  • the present invention also provides a device for generating a wrong question book.
  • the device includes:
  • the obtaining module is used to obtain the image of the revised test paper
  • a first recognition module used to recognize the areas of each question in the revised test paper according to the pre-trained first area recognition model
  • a second recognition module which is used to identify the question whose correction result is wrong in the revised test paper according to the pre-trained wrong question recognition model, as a wrong question;
  • the generating module is used for storing the wrong question area in the wrong question database and generating the wrong question book.
  • the second identification module is specifically used for:
  • the pre-trained wrong question recognition model identify the correction marks in the area of each topic, and determine the topics with the correction marks as the preset marks to be the ones whose correction results are wrong.
  • the second identification module identifies the correction marks in each topic area, specifically:
  • the device further includes:
  • a third recognition module used to recognize the answer area and/or the correction area of the wrong question according to the pre-trained second area recognition model
  • the generating module is specifically used for:
  • the device further includes:
  • the first association module is used to identify the characters in the answer area of the wrong question according to the pre-trained character recognition model, store the recognized wrong answer in the wrong question database and associate it with the wrong question.
  • the device further includes:
  • the second association module is configured to store at least one of the test paper ID, the cause of the error, and the assessment knowledge point corresponding to the wrong question in the wrong question database and associate it with the wrong question.
  • the generating module is specifically used for:
  • the device further includes:
  • the third association module is used for searching the questions corresponding to the wrong question in the question bank, obtaining the standard answer of the wrong question, storing the obtained standard answer in the wrong question database and associating with the wrong question.
  • the device further includes:
  • the pushing module is used to select questions with the same or similar assessment knowledge points from the wrong question database for pushing according to the assessment knowledge points corresponding to the wrong questions.
  • the device further includes:
  • a fourth identification module which is used to identify at least one of the student ID and name of the middle school student in the revised test paper and the class;
  • the generating module is specifically used for:
  • the area of the wrong question is stored in the wrong question database and associated with at least one of the identified student number and name and the class, to establish the wrong book of the student.
  • the present invention also provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete each other through the communication bus Communication;
  • the memory is used to store computer programs
  • the processor executes the computer program stored in the memory, it is used to implement the method for generating a wrong question book as described in any one of the above.
  • the present invention also provides a computer-readable storage medium in which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned problem is realized. This generation method.
  • the present invention recognizes the area of each question in the revised test paper according to the pre-trained first region recognition model, and then recognizes the model based on the pre-trained wrong question recognition model. Identify the question whose correction result is wrong in the revised test paper as a wrong question, and then store the wrong question area in the wrong question database, thereby generating a wrong question book.
  • the solution of the present invention can automatically generate a wrong question book based on the wrong question in the student test paper, without the need for students to spend time and effort to manually copy the wrong question, which improves the efficiency of generating the wrong question book and reduces the burden on students.
  • FIG. 1 is a schematic flowchart of a method for generating a wrong book according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a device for generating a wrong question according to an embodiment of the present invention
  • FIG. 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.
  • embodiments of the present invention provide a method, apparatus, electronic device, and computer-readable storage medium for generating a wrong book.
  • the method for generating a wrong book in the embodiment of the present invention can be applied to the apparatus for generating a wrong book in the embodiment of the present invention, and the apparatus for generating a wrong book can be configured on an electronic device.
  • the electronic device may be a personal computer, a mobile terminal, etc.
  • the mobile terminal may be a hardware device with various operating systems, such as a mobile phone, a tablet computer, or the like.
  • FIG. 1 is a schematic flowchart of a method for generating a wrong book according to an embodiment of the present invention. Please refer to FIG. 1, a method for generating a wrong question book may include the following steps:
  • Step S101 Obtain an image of the revised test paper.
  • the revised test paper may be a test paper manually revised by a teacher, or a test paper automatically revised by a computer device.
  • the images of the approved test papers can be obtained by taking pictures with mobile phones and other terminal devices, or by scanning with printers.
  • step S102 according to the pre-trained first region recognition model, the regions of each question in the revised test paper are identified.
  • the first region recognition model may be a model based on a neural network, for example, it may be obtained by training a sample in a test sample training set based on a deep convolutional neural network (Convolutional Neural Networks, CNN).
  • CNN convolutional Neural Networks
  • each topic can be cut into a single image, or not actually cut, and each topic area is divided into a single area image for processing during processing, and sorted according to the position information of the topic.
  • step S103 according to the pre-trained wrong question recognition model, the question whose correction result in the revised test paper is wrong is identified as the wrong question.
  • the correction marks in each topic area can be identified according to a pre-trained wrong question recognition model, and the topic whose correction mark is the preset mark can be determined as the topic whose correction result is wrong. It is understandable that the correct marking of the correct and wrong questions in the test paper is different. For example, the correct marking of the correct answer can be " ⁇ ", and the correct marking of the wrong answer can be " ⁇ ". That is, the mark for correction includes " ⁇ " and " ⁇ ", where the preset mark is " ⁇ ”. Therefore, by identifying the correction marks in the question area, you can identify the questions in the test paper where the correction result is wrong, that is, identify the wrong question.
  • the pre-trained error recognition model can be a neural network-based model, and the neural network model can be trained by extracting a large number of features of the modified logo to establish the error recognition model.
  • the above-mentioned identification marking in each topic area may specifically include: performing binarization processing on the image corresponding to each topic area, and identifying the correction mark in each topic area according to the pixel value range.
  • the answer to the judgment question may be represented by “ ⁇ ” and “ ⁇ ”.
  • the correction marks of the question are also “ ⁇ ” and “ ⁇ ”, the recognition result of the wrong question recognition model will be generated. influences.
  • teachers usually use red and other specific color pens to correct the test paper (different from the color of the pen used by the students to answer).
  • the image corresponding to the subject area is binarized and judged according to the range of pixel values
  • the area where the handwriting is corrected in red or other colors, that is, the correction mark is recognized, can avoid the wrong question recognition model from misidentifying the student's answer as the correction mark.
  • step S104 the area of the wrong question is stored in the wrong question database to generate a wrong question book.
  • the problem area of the wrong problem identified in step S102 may be directly stored in the wrong problem database.
  • the handwritten answer part and/or the correction identification part in the problem area of the wrong question can also be removed, and only the stem area of the wrong question is reserved and stored in the wrong question database, specifically: in step S104 Before, according to the pre-trained second region recognition model, identify the answer area and/or correction area of the wrong question, and also identify the stem or picture area; at this time, step S104 stores the wrong question area to The wrong question database may specifically include: covering the answer area and/or the correction area of the wrong question, and storing the wrong question area after the covering process in the wrong question database.
  • the covering process may include:
  • the answer area and/or correction area shall be covered by background color or white, mosaic or blurring.
  • the second region recognition model may be a neural network-based model. For a specific training method of the second region recognition model, refer to the training method of the neural network model in the prior art, and details are not described herein.
  • characters in the answer area of the wrong question may be recognized, and the recognized wrong answer may be stored in the wrong question database and associated with the wrong question.
  • the character recognition model can be trained based on the characters of the handwritten font.
  • the character recognition model can be established based on the hollow convolution and attention model. Specifically, the hollow convolution is used to extract the feature of the answer area, and then pass The attention model decodes the extracted features into characters.
  • test paper ID corresponding to the wrong question, the cause of the error, and/or the assessment knowledge point may also be stored in the wrong question database and associated with the wrong question for easy viewing.
  • the causes of errors can include calculation formula errors, calculation process errors, calculation results errors, unit errors, etc. Specifically, by identifying the characters in the answer area of the wrong question, the answer process listed in the answer area is compared with the standard answer to determine the error The reason, or you can manually enter the cause of the error.
  • Assessment knowledge points can include subdivision knowledge points such as oral calculation, off-calculation, unit conversion, four operations, unary quadratic equation, etc.
  • the overall classification of knowledge points can be pre-classified according to the syllabus, and by identifying the stem characters of the wrong question The content determines the corresponding assessment knowledge point, or by querying the questions in the question bank corresponding to a wrong question, the assessment knowledge point category marked in advance for the corresponding question can be determined as the assessment knowledge point category of the wrong question.
  • Question types can include oral calculation questions, fill-in-the-blank questions, judgment questions, application questions, etc. Specifically, you can determine the type of the question pre-marked by the corresponding question as the type of the wrong question by querying the question corresponding to a wrong question in the question bank Or, you can use the pre-trained classification and recognition model to identify the type of the wrong question.
  • the question bank may be searched for a question corresponding to the wrong question to obtain a standard answer to the wrong question, and the obtained standard answer may be stored in the wrong question database and associated with the wrong question.
  • the feature vectors of each question in the test paper where the wrong question is located can be extracted, the corresponding test paper can be searched in the question bank according to the feature vector, and the question corresponding to the wrong question can be determined from the corresponding test paper, or directly based on the wrong answer
  • the feature vectors of the questions are found in the question bank.
  • the same or similar questions with the assessment knowledge points can be selected and pushed from the wrong question database, so that the students can practice the knowledge points assessed.
  • the same assessment knowledge points refer to the same assessment knowledge point, for example, belong to the unit conversion of this assessment knowledge point; similar assessment knowledge points refer to different assessment knowledge points, but the two assessment knowledge points are very similar, this point can be It is determined in advance according to the syllabus or teacher experience. For example, it can be considered that oral calculation and off-calculation are similar to the assessment knowledge point. Therefore, if the assessment knowledge point corresponding to the wrong question belongs to the oral calculation, the assessment knowledge point can also be pushed when pushing. The topic of formula calculation.
  • the assessment knowledge points corresponding to each wrong question can be marked, so that when a certain wrong question is stored, the knowledge corresponding to the wrong question can be based on Click to select similar wrong questions from the wrong question database to push.
  • the revised test paper may also include student information such as student ID and/or name, class, etc. Based on the student information, a wrong book for each student may be established so that each student can conduct targeted learning. Specifically, the method can also identify the student ID and/or name and class of the middle school student who has already revised the test paper; Step S104 stores the area of the wrong question in the wrong question database to generate a wrong question book, which may include: The area of the wrong question is stored in the wrong question database and associated with the identified student number and/or name, class, and a wrong book of the student is established.
  • the present invention recognizes the area of each question in the revised test paper according to the pre-trained first region recognition model, and then recognizes the model based on the pre-trained wrong question recognition model. Identify the question whose correction result is wrong in the revised test paper as a wrong question, and then store the wrong question area in the wrong question database, thereby generating a wrong question book.
  • the solution of the present invention can automatically generate a wrong question book based on the wrong question in the student test paper, without the need for students to spend time and effort to manually copy the wrong question, which improves the efficiency of generating the wrong question book and reduces the burden on students.
  • the present invention provides a device for generating a wrong question book.
  • the device may include:
  • the obtaining module 201 is used to obtain the image of the revised test paper
  • the first recognition module 202 is used to recognize the areas of each question in the revised test paper according to the pre-trained first area recognition model
  • the second recognition module 203 is used to identify the question whose correction result in the revised test paper is wrong according to the pre-trained wrong question recognition model as a wrong question;
  • the generating module 204 is configured to store the wrong question area in the wrong question database and generate a wrong question book.
  • the second identification module 203 is specifically used to:
  • the pre-trained wrong question recognition model identify the correction marks in the area of each topic, and determine the topics with the correction marks as the preset marks to be the ones whose correction results are wrong.
  • the second identification module 203 identifies the correction marks in each topic area, specifically:
  • Binary processing is performed on the images corresponding to each topic area, and the correction marks in each topic area are identified according to the pixel value range.
  • the device further includes:
  • a third recognition module used to recognize the answer area and/or the correction area of the wrong question according to the pre-trained second area recognition model
  • the generating module 204 is specifically used to:
  • the device further includes:
  • the first correlation module is used to recognize the characters in the answer area of the wrong question according to a pre-trained character recognition model, store the recognized wrong answer in the wrong question database and associate it with the wrong question.
  • the device further includes:
  • the second association module is configured to store the test paper ID, the cause of the error, and/or the assessment knowledge point corresponding to the wrong question in the wrong question database and associate it with the wrong question.
  • the generating module 204 is specifically used to:
  • the device further includes:
  • the third association module is used for searching the questions corresponding to the wrong question in the question bank, obtaining the standard answer of the wrong question, storing the obtained standard answer in the wrong question database and associating with the wrong question .
  • the device further includes:
  • the pushing module is used to select questions with the same or similar assessment knowledge points from the wrong question database for pushing according to the assessment knowledge points corresponding to the wrong questions.
  • the device further includes:
  • a fourth identification module which is used to identify the student ID and/or name and class of the middle school student who has already approved the examination paper;
  • the generating module 204 is specifically used to:
  • the wrong question area is stored in the wrong question database and associated with the identified student number and/or name, class, and the wrong question book of the student is established.
  • the present invention also provides an electronic device, as shown in FIG. 3, which includes a processor 301, a communication interface 302, a memory 303, and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 complete each other through the communication bus 304 Communication between,
  • Memory 303 used to store computer programs
  • the area of the wrong question is stored in the wrong question database to generate a wrong question book.
  • the communication bus mentioned in the above electronic equipment may be a peripheral component interconnection standard (Peripheral Component Interconnect, PCI) bus or an extended industry standard structure (Extended Industry Standard Architecture, EISA) bus, etc.
  • PCI peripheral component interconnection standard
  • EISA Extended Industry Standard Architecture
  • the communication bus can be divided into an address bus, a data bus, and a control bus. For ease of representation, only a thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.
  • the communication interface is used for communication between the electronic device and other devices.
  • the memory may include random access memory (Random Access Memory, RAM), or non-volatile memory (Non-Volatile Memory, NVM), for example, at least one disk memory.
  • RAM Random Access Memory
  • NVM Non-Volatile Memory
  • the memory may also be at least one storage device located away from the foregoing processor.
  • the aforementioned processor may be a general-purpose processor, including a central processor (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; it may also be a digital signal processor (Digital Signal Processing, DSP), dedicated integration Circuit (Application Specific Integrated Circuit, ASIC), field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.
  • a central processor Central Processing Unit, CPU
  • NP Network Processor
  • DSP Digital Signal Processing
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • the present invention also provides a computer-readable storage medium in which a computer program is stored.
  • a computer program is stored.
  • the method steps of the method for generating a wrong book described above are implemented.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Human Computer Interaction (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Educational Administration (AREA)
  • Educational Technology (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Image Analysis (AREA)

Abstract

一种错题本生成方法及装置，所述方法包括：获得已批改试卷的图像（S101）；根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域（S102）；根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题（S103）；将所述错题的区域存储到错题数据库中，生成错题本（S104）。应用上述方案可以解决现有技术中生成错题本效率较低的问题。

Description

一种错题本生成方法及装置 技术领域
本发明涉及教学及信息处理技术领域，尤其涉及一种错题本生成方法、装置、电子设备和计算机可读存储介质。
背景技术
错题本有助于学生快速掌握薄弱知识点、优化学习路径以及复习。目前，学生通常是通过手写的方式将错题抄到笔记本中来建立各自的错题本。这种方式生成错题本效率较低，且学生可能不愿意花费时间精力来手写错题，进而影响学生的学习和成绩。
发明内容
本发明的目的在于提供一种错题本生成方法、装置、电子设备和计算机可读存储介质，以解决现有技术中生成错题本效率较低的问题。
为达到上述目的，本发明提供了一种错题本生成方法，所述方法包括：
获得已批改试卷的图像；
根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域；
根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题；以及
将所述错题的区域存储到错题数据库中，生成错题本。
可选的，所述根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，包括：
根据预先训练的错题识别模型，识别各个题目的区域内的批改标识，将批改标识为预设标识的题目确定为批改结果为错误的题目。
可选的，所述识别各个题目的区域内的批改标识，包括：
对各个题目的区域对应的图像进行二值化处理，根据所述二值化处理得到的像素值范围来识别各个题目的区域内的批改标识。
可选的，所述方法还包括：
根据预先训练的第二区域识别模型，识别所述错题的答案区域和/或批改区域；
所述将所述错题的区域存储到错题数据库中，包括：
将所述错题的答案区域和/或批改区域进行遮盖处理，并将遮盖处理后的所述错题的区域存储到错题数据库中。
可选的，所述方法还包括：
根据预先训练的字符识别模型，识别所述错题的答案区域的字符，将所识别的错误答案存储到所述错题数据库中并与所述错题关联。
可选的，所述方法还包括：
将所述错题对应的试卷ID、错误原因和考核知识点中的至少一个存储到所述错题数据库中并与所述错题关联。
可选的，所述方法还包括：识别所述错题的题目类型；所述将所述错题的区域存储到错题数据库中，包括：
将所述错题的区域存储到错题数据库中相应的题目类型分组中。
可选的，所述方法还包括：
在题库中搜索与所述错题对应的题目，获得所述错题的标准答案，将所获得的标准答案存储到所述错题数据库中并与所述错题关联。
可选的，所述方法还包括：
根据所述错题对应的考核知识点，从所述错题数据库中选择考核知识点相同或相似的题目进行推送。
可选的，所述方法还包括：
识别所述已批改试卷中学生的学号和姓名中的至少一个以及班级；
所述将所述错题的区域存储到错题数据库中，生成错题本，包括：
将所述错题的区域存储到错题数据库中并与所识别的学号和姓名中的至少一个以及所述班级关联，建立所述学生的错题本。
为达到上述目的，本发明还提供了一种错题本生成装置，所述装置包括：
获得模块，用于获得已批改试卷的图像；
第一识别模块，用于根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域；
第二识别模块，用于根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题；
生成模块，用于将所述错题的区域存储到错题数据库中，生成错题本。
可选的，所述第二识别模块，具体用于：
根据预先训练的错题识别模型，识别各个题目的区域内的批改标识，将批改标识为预设标识的题目确定为批改结果为错误的题目。
可选的，所述第二识别模块识别各个题目的区域内的批改标识，具体为：
对各个题目的区域对应的图像进行二值化处理，根据所述二值化处理得到的像素值范围来识别各个题目的区域内的批改标识。
可选的，所述装置还包括：
第三识别模块，用于根据预先训练的第二区域识别模型，识别所述错题的答案区域和/或批改区域；
所述生成模块，具体用于：
将所述错题的答案区域和/或批改区域进行遮盖处理，并将遮盖处理后的所述错题的区域存储到错题数据库中。
可选的，所述装置还包括：
第一关联模块，用于根据预先训练的字符识别模型，识别所述错题的答案区域的字符，将所识别的错误答案存储到所述错题数据库中并与所述错题关联。
可选的，所述装置还包括：
第二关联模块，用于将所述错题对应的试卷ID、错误原因和考核知识点中的至少一个存储到所述错题数据库中并与所述错题关联。
可选的，所述生成模块，具体用于：
识别所述错题的题目类型，将所述错题的区域存储到错题数据库中相应的题目类型分组中。
可选的，所述装置还包括：
第三关联模块，用于在题库中搜索与所述错题对应的题目，获得所述错题的标准答案，将所获得标准答案存储到所述错题数据库中并与所述错题关联。
可选的，所述装置还包括：
推送模块，用于根据所述错题对应的考核知识点，从所述错题数据库中选择考核知识点相同或相似的题目进行推送。
可选的，所述装置还包括：
第四识别模块，用于识别所述已批改试卷中学生的学号和姓名中的至少一个以及班级；
所述生成模块，具体用于：
将所述错题的区域存储到错题数据库中并与所识别的学号和姓名中的至少一个以及所述班级关联，建立所述学生的错题本。
为达到上述目的，本发明还提供了一种电子设备，包括处理器、通信接口、存储器和通信总线，其中，所述处理器、所述通信接口和所述存储器通过所述通信总线完成相互间的通信；
所述存储器用于存放计算机程序；
所述处理器在执行所述存储器上所存放的所述计算机程序时，用于实现如上任一项所述的错题本生成方法。
为达到上述目的，本发明还提供了一种计算机可读存储介质，所述计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现如上任一项所述的错题本生成方法。
与现有技术相比，本发明在获得已批改试卷的图像之后，根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域，再根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题，然后将所述错题的区域存储到错题数据库中，从而生成错题本。应用本发明的方案可以自动根据学生试卷中的错题生成错题本，不需要学生花费时间精力手工抄写错题，提高了生成错题本的效率，减轻了学生的负担。
附图说明
图1是本发明一实施例提供的错题本生成方法的流程示意图；
图2是本发明一实施例提供的错题本生成装置的结构示意图；
图3是本发明一实施例提供的电子设备的结构示意图。
具体实施方式
以下结合附图和具体实施例对本发明提出的一种错题本生成方法、装置、电子设备及计算机可读存储介质作进一步详细说明。根据权利要求书和下面说明，本发明的优点和特征将更清楚。需说明的是，附图均采用非常简化的形式且均使用非精准的比例，仅用以方便、明晰地辅助说明本发明实施例的目的。
为解决现有技术的问题，本发明实施例提供了一种错题本生成方法、装置、电子设备及计算机可读存储介质。
需要说明的是，本发明实施例的错题本生成方法可应用于本发明实施例的错题本生成装置，该错题本生成装置可被配置于电子设备上。其中，该电子设备可以是个人计算机、移动终端等，该移动终端可以是手机、平板电脑等具有各种操作系统的硬件设备。
图1是本发明一实施例提供的一种错题本生成方法的流程示意图。请参考图1，一种错题本生成方法可以包括如下步骤：
步骤S101，获得已批改试卷的图像。
其中，已批改试卷可以为经过老师人工批改过的试卷，也可以为由计算机设备自动批改过的试卷。已批改试卷的图像可以通过手机等终端设备拍照的方式获取，也可以通过打印机扫描的方式获取。
步骤S102，根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域。
其中，所述第一区域识别模型可以为基于神经网络的模型，例如可以是基于深度卷积神经网络(Convolutional Neural Networks，CNN)对试卷样本训练集中的样本进行训练得到的。利用训练好的第一区域识别模型从已批改 试卷的图像中提取二维特征向量，在二维特征向量的每个网格生成不同形状的锚点，使用标注框(Groundtruth Boxes)将识别出的各个题目的区域进行标注，还可以将标注框与生成的锚点作回归(regression)处理，以使标注框更贴近题目的实际位置。识别完题目区域后可以将每道题目进行切割为单个影像，或者不实际切割，而在处理时将每个题目区域区分开为单个区域影像进行处理，根据题目位置信息进行排序。
步骤S103，根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题。
在一种实现方式中，可以根据预先训练的错题识别模型，识别各个题目的区域内的批改标识，将批改标识为预设标识的题目确定为批改结果为错误的题目。可以理解的是，已批改过试卷中答案正确和错误的题目的批改标识是不同的，例如答案正确的题目的批改标识可以为“√”，答案错误的题目的批改标识可以为“×”，即批改标识包括“√”和“×”，其中预设标识为“×”。因此，通过识别出题目区域内的批改标识，即可识别出试卷中批改结果为错误的题目，即识别出错题。具体的，预先训练的错题识别模型可以为基于神经网络的模型，通过提取大量的批改标识的特征对神经网络的模型进行训练可以建立该错题识别模型，该错题识别模型具体的训练方法可以参见现有技术中神经网络模型的训练方法，在此不做赘述。
进一步的，上述识别各个题目的区域内的批改标识，具体可以包括：对各个题目的区域对应的图像进行二值化处理，根据像素值范围来识别各个题目的区域内的批改标识。可以理解的是，对于判断题的答案可能会以“√”和“×”来表示，这时若题目的批改标识也是“√”和“×”，则会对错题识别模型的识别结果产生影响。而目前老师批改试卷时通常使用红色等特定颜色的笔进行批改(不同于学生作答所使用的笔的颜色)，因此，对题目区域对应的图像进行二值化处理，并根据像素值范围来判断红色或其他颜色批改笔迹的区域，即识别出批改标识，可以避免错题识别模型将学生答案误识别为批改标识。
步骤S104，将所述错题的区域存储到错题数据库中，生成错题本。
在一种实现方式中，可以将步骤S102识别出的错题的题目区域直接存储到错题数据库中。在另一种实现方式中，还可以去除错题的题目区域内的手写答案部分和/或批改标识部分，仅保留错题的题干区域存储到错题数据库中，具体为：在执行步骤S104之前，根据预先训练的第二区域识别模型，识别所述错题的答案区域和/或批改区域，同时也可以识别题干或者图片区域；此时，步骤S104将所述错题的区域存储到错题数据库中，具体可以包括：将所述错题的答案区域和/或批改区域进行遮盖处理，并将遮盖处理后的所述错题的区域存储到错题数据库中，遮盖处理可以包括：将答案区域和/或批改区域采用背景色覆盖或白色覆盖、打马赛克或模糊处理等处理方式。其中，第二区域识别模型可以为基于神经网络的模型，该第二区域识别模型具体的训练方法可以参见现有技术中神经网络模型的训练方法，在此不做赘述。
进一步的，还可以根据预先训练的字符识别模型，识别所述错题的答案区域的字符，将所识别的错误答案存储到所述错题数据库中并与所述错题关联。其中，所述字符识别模型可以基于手写字体的字符训练而成，该字符识别模型可以是基于空洞卷积和注意力模型建立的，具体的，采用空洞卷积对答案区域进行特征提取，再通过注意力模型将提取到的特征解码成字符。
进一步的，还可以将所述错题对应的试卷ID、错误原因和/或考核知识点存储到所述错题数据库中并与所述错题关联，以便于查看。错误原因可以包括计算公式错误、计算过程错误、计算结果错误、单位错误等，具体可以通过识别错题的答案区域的字符，将答案区域列出的解答过程与标准答案进行比对从而确定出错误原因，或者也可以由人工输入错误原因。考核知识点可以包括口算、脱式计算、单位换算，四则运算，一元二次方程等各细分知识点，具体可以预先根据教学大纲进行知识点的整体分类，并通过识别错题的题干字符内容确定对应的考核知识点，也可以通过查询题库中与某错题相应的题目，将该相应的题目所预先标注的考核知识点类别确定为该错题的考核知识点类别。
另外，还可以按题目类型对错题进行分类存储，具体的，在将所述错题的区域存储到错题数据库中时，识别所述错题的题目类型，将所述错题的区 域存储到错题数据库中相应的题目类型分组中。题目类型可以包括口算题，填空题，判断题，应用题等，具体可以通过查询题库中与某错题相应的题目，将该相应的题目所预先标注的题目类型确定为该错题的题目类型，或者也可以通过预先训练的分类识别模型来识别错题的题目类型。
进一步的，还可以在题库中搜索与所述错题对应的题目，获得所述错题的标准答案，将所获得的标准答案存储到所述错题数据库中并与所述错题关联。将标准答案与错题对应存储，以便于后期查看。具体的，可以提取所述错题所在试卷的各个题目的特征向量，根据特征向量在题库中搜索对应的试卷，从对应的试卷中确定与所述错题对应的题目，或者直接根据所述错题的特征向量在题库中搜索到对应的题目。
进一步的，还可以根据所述错题对应的考核知识点，从所述错题数据库中选择考核知识点相同或相似的题目进行推送，以便学生对所考核的知识点进行练习。在此，考核知识点相同指属于同一个考核知识点，例如同属于单位换算这个考核知识点；考核知识点相似指属于不同的考核知识点，但是该两个考核知识点非常相近，这点可以根据教学大纲或者教师经验预先予以确定，例如，可以认为口算和脱式计算属于考核知识点相似，由此，若错题对应的考核知识点属于口算，则推送时也可以推送考核知识点属于脱式计算的题目。具体的，在将识别出的错题存储到错题数据库时，可以对各个错题对应的考核知识点进行标注，这样，当将某一错题进行存储时，可以根据该错题对应的知识点从错题数据库中选择类似的错题进行推送。
在实际应用中，所述已批改试卷中还可以包含学生的学号和/或姓名、班级等学生信息，基于学生信息可以建立各个学生的错题本，以便各个学生进行针对性学习。具体的，所述方法还可以识别所述已批改试卷中学生的学号和/或姓名、班级；步骤S104将所述错题的区域存储到错题数据库中，生成错题本，可以包括：将所述错题的区域存储到错题数据库中并与所识别的学号和/或姓名、班级关联，建立所述学生的错题本。
与现有技术相比，本发明在获得已批改试卷的图像之后，根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域，再根据预先 训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题，然后将所述错题的区域存储到错题数据库中，从而生成错题本。应用本发明的方案可以自动根据学生试卷中的错题生成错题本，不需要学生花费时间精力手工抄写错题，提高了生成错题本的效率，减轻了学生的负担。
相应于上述错题本生成方法实施例，本发明提供了一种错题本生成装置，参见图2，该装置可以包括：
获得模块201，用于获得已批改试卷的图像；
第一识别模块202，用于根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域；
第二识别模块203，用于根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题；
生成模块204，用于将所述错题的区域存储到错题数据库中，生成错题本。
可选的，所述第二识别模块203，具体用于：
根据预先训练的错题识别模型，识别各个题目的区域内的批改标识，将批改标识为预设标识的题目确定为批改结果为错误的题目。
可选的，所述第二识别模块203识别各个题目的区域内的批改标识，具体为：
对各个题目的区域对应的图像进行二值化处理，根据像素值范围来识别各个题目的区域内的批改标识。
可选的，所述装置还包括：
第三识别模块，用于根据预先训练的第二区域识别模型，识别所述错题的答案区域和/或批改区域；
所述生成模块204，具体用于：
将所述错题的答案区域和/或批改区域进行遮盖处理，并将遮盖处理后的所述错题的区域存储到错题数据库中。
可选的，所述装置还包括：
第一关联模块，用于根据预先训练的字符识别模型，识别所述错题的答 案区域的字符，将所识别的错误答案存储到所述错题数据库中并与所述错题关联。
可选的，所述装置还包括：
第二关联模块，用于将所述错题对应的试卷ID、错误原因和/或考核知识点存储到所述错题数据库中并与所述错题关联。
可选的，所述生成模块204，具体用于：
识别所述错题的题目类型，将所述错题的区域存储到错题数据库中相应的题目类型分组中。
可选的，所述装置还包括：
第三关联模块，用于在题库中搜索与所述错题对应的题目，获得所述错题的标准答案，将所获得的标准答案存储到所述错题数据库中并与所述错题关联。
可选的，所述装置还包括：
推送模块，用于根据所述错题对应的考核知识点，从所述错题数据库中选择考核知识点相同或相似的题目进行推送。
可选的，所述装置还包括：
第四识别模块，用于识别所述已批改试卷中学生的学号和/或姓名、班级；
所述生成模块204，具体用于：
将所述错题的区域存储到错题数据库中并与所识别的学号和/或姓名、班级关联，建立所述学生的错题本。
本发明还提供了一种电子设备，如图3所示，包括处理器301、通信接口302、存储器303和通信总线304，其中，处理器301，通信接口302，存储器303通过通信总线304完成相互间的通信，
存储器303，用于存放计算机程序；
处理器301，用于执行存储器303上所存放的程序时，实现如下步骤：
获得已批改试卷的图像；
根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的 区域；
根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题；
将所述错题的区域存储到错题数据库中，生成错题本。
关于该方法各个步骤的具体实现以及相关解释内容可以参见上述图1所示的方法实施例，在此不做赘述。
另外，处理器301执行存储器303上所存放的程序而实现的错题本生成方法的其他实现方式，与前述方法实施例部分所提及的实现方式相同，这里也不再赘述。
上述电子设备提到的通信总线可以是外设部件互连标准(Peripheral Component Interconnect，PCI)总线或扩展工业标准结构(Extended Industry Standard Architecture，EISA)总线等。该通信总线可以分为地址总线、数据总线、控制总线等。为便于表示，图中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
通信接口用于上述电子设备与其他设备之间的通信。
存储器可以包括随机存取存储器(Random Access Memory，RAM)，也可以包括非易失性存储器(Non-Volatile Memory，NVM)，例如至少一个磁盘存储器。可选的，存储器还可以是至少一个位于远离前述处理器的存储装置。
上述的处理器可以是通用处理器，包括中央处理器(Central Processing Unit，CPU)、网络处理器(Network Processor，NP)等；还可以是数字信号处理器(Digital Signal Processing，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。
本发明还提供了一种计算机可读存储介质，该计算机可读存储介质内存储有计算机程序，该计算机程序被处理器执行时实现上述的错题本生成方法的方法步骤。
需要说明的是，本说明书中的各个实施例均采用相关的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于装置、电子设备、计算机可读存储介质实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
上述描述仅是对本发明较佳实施例的描述，并非对本发明范围的任何限定，本发明领域的普通技术人员根据上述揭示内容做的任何变更、修饰，均属于权利要求书的保护范围。

Claims (15)

  1. 一种错题本生成方法，其特征在于，所述方法包括：
    获得已批改试卷的图像；
    根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域；
    根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题；以及
    将所述错题的区域存储到错题数据库中，生成错题本。
  2. 如权利要求1所述的错题本生成方法，其特征在于，所述根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，包括：
    根据预先训练的错题识别模型，识别各个题目的区域内的批改标识，将批改标识为预设标识的题目确定为批改结果为错误的题目。
  3. 如权利要求2所述的错题本生成方法，其特征在于，所述识别各个题目的区域内的批改标识，包括：
    对各个题目的区域对应的图像进行二值化处理，根据所述二值化处理得到的像素值范围来识别各个题目的区域内的批改标识。
  4. 如权利要求1所述的错题本生成方法，其特征在于，所述方法还包括：
    根据预先训练的第二区域识别模型，识别所述错题的答案区域和/或批改区域；
    所述将所述错题的区域存储到错题数据库中，包括：
    将所述错题的答案区域和/或批改区域进行遮盖处理，并将遮盖处理后的所述错题的区域存储到错题数据库中。
  5. 如权利要求4所述的错题本生成方法，其特征在于，所述方法还包括：
    根据预先训练的字符识别模型，识别所述错题的答案区域的字符，将所识别的错误答案存储到所述错题数据库中并与所述错题关联。
  6. 如权利要求1所述的错题本生成方法，其特征在于，所述方法还包括：
    将所述错题对应的试卷ID、错误原因和考核知识点中的至少一个存储到 所述错题数据库中并与所述错题关联。
  7. 如权利要求1所述的错题本生成方法，其特征在于，所述方法还包括：识别所述错题的题目类型；
    所述将所述错题的区域存储到错题数据库中，包括：
    将所述错题的区域存储到错题数据库中相应的题目类型分组中。
  8. 如权利要求1所述的错题本生成方法，其特征在于，所述方法还包括：
    在题库中搜索与所述错题对应的题目，获得所述错题的标准答案，将所获得的标准答案存储到所述错题数据库中并与所述错题关联。
  9. 如权利要求1所述的错题本生成方法，其特征在于，所述方法还包括：
    根据所述错题对应的考核知识点，从所述错题数据库中选择考核知识点相同或相似的题目进行推送。
  10. 如权利要求1-9中任一项所述的错题本生成方法，其特征在于，所述方法还包括：
    识别所述已批改试卷中学生的学号和姓名中的至少一个以及班级；
    所述将所述错题的区域存储到错题数据库中，生成错题本，包括：
    将所述错题的区域存储到错题数据库中并与所识别的学号和姓名中的至少一个以及所述班级关联，建立所述学生的错题本。
  11. 一种错题本生成装置，其特征在于，所述装置包括：
    获得模块，用于获得已批改试卷的图像；
    第一识别模块，用于根据预先训练的第一区域识别模型，识别所述已批改试卷中各个题目的区域；
    第二识别模块，用于根据预先训练的错题识别模型，识别所述已批改试卷中批改结果为错误的题目，作为错题；
    生成模块，用于将所述错题的区域存储到错题数据库中，生成错题本。
  12. 如权利要求11所述的错题本生成装置，其特征在于，所述第二识别模块用于：
    根据预先训练的错题识别模型，识别各个题目的区域内的批改标识，将批改标识为预设标识的题目确定为批改结果为错误的题目。
  13. 如权利要求11所述的错题本生成装置，其特征在于，所述装置还包括：
    第三识别模块，用于根据预先训练的第二区域识别模型，识别所述错题的答案区域和/或批改区域；
    所述生成模块用于：
    将所述错题的答案区域和/或批改区域进行遮盖处理，并将遮盖处理后的所述错题的区域存储到错题数据库中。
  14. 一种电子设备，其特征在于，包括处理器、通信接口、存储器和通信总线，其中，所述处理器、所述通信接口和所述存储器通过所述通信总线完成相互间的通信；
    所述存储器用于存放计算机程序；
    所述处理器在执行所述存储器上所存放的所述计算机程序时，用于实现权利要求1-10中任一项所述的方法。
  15. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1-10中任一项所述的方法。
PCT/CN2019/128160 2018-12-26 2019-12-25 一种错题本生成方法及装置 WO2020135472A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/418,238 US11410407B2 (en) 2018-12-26 2019-12-25 Method and device for generating collection of incorrectly-answered questions

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811603491.3 2018-12-26
CN201811603491.3A CN109710590B (zh) 2018-12-26 2018-12-26 一种错题本生成方法及装置

Publications (1)

Publication Number Publication Date
WO2020135472A1 true WO2020135472A1 (zh) 2020-07-02

Family

ID=66257743

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/128160 WO2020135472A1 (zh) 2018-12-26 2019-12-25 一种错题本生成方法及装置

Country Status (3)

Country Link
US (1) US11410407B2 (zh)
CN (1) CN109710590B (zh)
WO (1) WO2020135472A1 (zh)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109710590B (zh) 2018-12-26 2021-05-07 杭州大拿科技股份有限公司 一种错题本生成方法及装置
CN110135339A (zh) * 2019-05-15 2019-08-16 宜春宜联科技有限公司 作业制定方法及装置
CN110147434A (zh) * 2019-05-23 2019-08-20 联想(北京)有限公司 一种信息处理方法和电子设备
CN110348444A (zh) * 2019-05-31 2019-10-18 浙江米猪控股有限公司 基于深度学习的错题收集方法、装置及设备
CN110443732B (zh) * 2019-07-30 2023-05-05 南京乐教信息技术有限公司 作业统分方法、装置、设备和介质
CN110533037B (zh) * 2019-08-30 2022-04-22 联想(北京)有限公司 一种信息处理方法及电子设备
CN110705534B (zh) * 2019-09-17 2022-06-14 浙江工业大学 一种适用于电子助视器的错题本生成方法
CN111586239A (zh) * 2020-05-18 2020-08-25 艾新来 学生专用手机
CN111950557A (zh) * 2020-08-21 2020-11-17 珠海奔图电子有限公司 错题处理方法、图像形成装置及电子设备
CN113205527A (zh) * 2021-04-02 2021-08-03 广州远大信息发展有限公司 一种试卷智能切割方法、系统及存储介质
CN113326010A (zh) * 2021-04-23 2021-08-31 深圳市沃特沃德信息有限公司 错题打印方法、装置和计算机设备
CN113688201A (zh) * 2021-08-06 2021-11-23 上海汉图科技有限公司 作业生成方法及服务端
CN114241043A (zh) * 2022-02-23 2022-03-25 北京世纪好未来教育科技有限公司 题目类别检测方法、训练方法、装置、电子设备和介质
CN114820254A (zh) * 2022-04-15 2022-07-29 光合新知(北京)科技有限公司 一种基于学生对错题的知识掌握策略分析方法和系统
CN116662533B (zh) * 2023-08-02 2023-11-03 广东信聚丰科技股份有限公司 基于ai辅助的题库知识点挖掘方法及授课服务系统
CN117831037A (zh) * 2024-01-04 2024-04-05 北京和气聚力教育科技有限公司 一种答题卡中客观题答题情况的确定方法及装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104794948A (zh) * 2015-04-20 2015-07-22 西安青柠电子信息技术有限公司 自动阅卷系统及其使用方法
CN105095892A (zh) * 2014-05-16 2015-11-25 上海市上海中学 基于图像处理的学生文档管理系统
CN106897749A (zh) * 2017-03-03 2017-06-27 盐城工学院 自动阅卷方法及系统
CN107784264A (zh) * 2017-05-05 2018-03-09 合肥林翔智能科技有限公司 基于图像处理的答卷分析方法及系统
CN108932508A (zh) * 2018-08-13 2018-12-04 杭州大拿科技股份有限公司 一种题目智能识别、批改的方法和系统
CN109710590A (zh) * 2018-12-26 2019-05-03 杭州大拿科技股份有限公司 一种错题本生成方法及装置

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6810232B2 (en) * 2001-03-05 2004-10-26 Ncs Pearson, Inc. Test processing workflow tracking system
US7286793B1 (en) * 2001-05-07 2007-10-23 Miele Frank R Method and apparatus for evaluating educational performance
US6927872B2 (en) * 2001-07-25 2005-08-09 Hewlett-Packard Development Company, L.P. Data acquisition system and method using answer forms
AU2003239936A1 (en) * 2002-05-31 2003-12-19 Vsc Technologies, Llc System for scoring scanned test answer sheets
JP4872214B2 (ja) * 2005-01-19 2012-02-08 富士ゼロックス株式会社 自動採点装置
US20060252023A1 (en) * 2005-05-03 2006-11-09 Lexmark International, Inc. Methods for automatically identifying user selected answers on a test sheet
JP4894189B2 (ja) * 2005-08-02 2012-03-14 富士ゼロックス株式会社 教材処理装置、教材処理方法、教材処理プログラム及び記録媒体
US20070048718A1 (en) * 2005-08-09 2007-03-01 Exam Grader, Llc System and Method for Test Creation, Verification, and Evaluation
US20100159437A1 (en) * 2008-12-19 2010-06-24 Xerox Corporation System and method for recommending educational resources
US8336779B1 (en) * 2007-09-10 2012-12-25 Robert Austin Porter System for automatically reading a response form using a digital camera
US20090226872A1 (en) * 2008-01-16 2009-09-10 Nicholas Langdon Gunther Electronic grading system
CN101916327B (zh) * 2010-07-09 2011-11-09 北京商纳科技有限公司 一种生成错题本的方法及系统
WO2013040088A1 (en) * 2011-09-13 2013-03-21 Monk Akarshala Design Private Limited Learner admission systems and methods in a modular learning system
US20150187219A1 (en) * 2013-12-27 2015-07-02 Cloud Ta Llc Systems and methods for computer-assisted grading of printed tests
CN104268603B (zh) * 2014-09-16 2017-04-12 科大讯飞股份有限公司 用于文字性客观题的智能阅卷方法及系统
KR101812755B1 (ko) * 2016-06-10 2017-12-27 주식회사 아이디엘 한글 철자 검사지의 검사 결과 해석 시스템
CN108564035B (zh) * 2018-04-13 2020-09-25 杭州睿琪软件有限公司 识别单据上记载的信息的方法及系统
US20190370672A1 (en) * 2018-05-30 2019-12-05 Ashley Jean Funderburk Computerized intelligent assessment systems and methods
US10755039B2 (en) * 2018-11-15 2020-08-25 International Business Machines Corporation Extracting structured information from a document containing filled form images
CN109815955B (zh) * 2019-03-04 2021-09-28 杭州大拿科技股份有限公司 题目辅助方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105095892A (zh) * 2014-05-16 2015-11-25 上海市上海中学 基于图像处理的学生文档管理系统
CN104794948A (zh) * 2015-04-20 2015-07-22 西安青柠电子信息技术有限公司 自动阅卷系统及其使用方法
CN106897749A (zh) * 2017-03-03 2017-06-27 盐城工学院 自动阅卷方法及系统
CN107784264A (zh) * 2017-05-05 2018-03-09 合肥林翔智能科技有限公司 基于图像处理的答卷分析方法及系统
CN108932508A (zh) * 2018-08-13 2018-12-04 杭州大拿科技股份有限公司 一种题目智能识别、批改的方法和系统
CN109710590A (zh) * 2018-12-26 2019-05-03 杭州大拿科技股份有限公司 一种错题本生成方法及装置

Also Published As

Publication number Publication date
US11410407B2 (en) 2022-08-09
CN109710590A (zh) 2019-05-03
US20220067416A1 (en) 2022-03-03
CN109710590B (zh) 2021-05-07

Similar Documents

Publication Publication Date Title
WO2020135472A1 (zh) 一种错题本生成方法及装置
WO2020156362A1 (zh) 一种试卷批改方法、装置、电子设备及存储介质
CN109993112B (zh) 一种图片中表格的识别方法及装置
US20210020059A1 (en) Answer evaluation method, answer evaluation system, electronic device, and medium
US11508251B2 (en) Method and system for intelligent identification and correction of questions
CN109670504B (zh) 一种手写答案识别批改方法及装置
CN109583429B (zh) 一种批改试卷中应用题的方法及装置
CN109634961B (zh) 一种试卷样本生成方法、装置、电子设备和存储介质
WO2018006294A1 (zh) 一种基于图像模式识别技术的阅卷系统、装置和方法
CN110110581B (zh) 一种基于人工智能的试卷批改方法和系统
CN110956138B (zh) 一种基于家教设备的辅助学习方法及家教设备
WO2023273583A1 (zh) 阅卷方法、装置、电子设备及存储介质
CN109189895B (zh) 一种针对口算题的题目批改方法及装置
CN107479791A (zh) 关联注释信息确定方法、装置、智能教学设备及存储介质
CN105427696A (zh) 一种对目标题目所作答案进行判别的方法
CN113537801B (zh) 板书处理方法、装置、终端和存储介质
WO2022161293A1 (zh) 图像处理方法及装置、电子设备和存储介质
CN112580503A (zh) 一种作业批改方法、装置、设备及存储介质
CN111008594B (zh) 改错题评阅方法、相关设备及可读存储介质
CN112396047A (zh) 训练样本生成方法、装置、计算机设备和存储介质
US11749128B2 (en) Answer correction method and device
CN112396897A (zh) 一种教学系统
WO2023024898A1 (zh) 题目辅助方法、题目辅助装置和题目辅助系统
Ravikumar et al. Automation of Answer Scripts Evaluation-A Review
CN115512383A (zh) 一种作业批改方法、装置、计算机设备以及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19904054

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19904054

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19904054

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 04/02/2022)

122 Ep: pct application non-entry in european phase

Ref document number: 19904054

Country of ref document: EP

Kind code of ref document: A1