CN115294573A - Job correction method, device, equipment and medium - Google Patents

Abstract

The present disclosure provides a job approval method, an apparatus, a device and a medium, which relate to the field of artificial intelligence, wherein the job approval method comprises: acquiring a response result image of the current operation; performing text recognition on the image of the answer result to obtain an initial answer result; obtaining the question type of the current operation, and adjusting the initial response result based on the question type of the current operation to obtain a target response result; and judging the target response result based on the question type of the current operation and the standard response result of the current operation, and determining the correction result of the current operation. The method and the device can adjust the initial answer result in a more targeted manner, improve the accuracy of the finally obtained target answer result, determine the correction result of the target answer result according to the question type of the current operation and the standard answer result, and improve the adaptability of various question types and the accuracy of the correction result in the correction process.

Description

Translated fromChinese

作业批改方法、装置、设备及介质Homework correction method, device, equipment and medium

技术领域technical field

本公开涉及人工智能领域，尤其涉及作业批改方法、装置、设备及介质。The present disclosure relates to the field of artificial intelligence, and in particular to a method, device, equipment and medium for job correction.

背景技术Background technique

随着人工智能技术的发展，人工智能的应用场景愈加广泛，在教育教学等场景中，也可以使用人工智能技术来辅助教师和/或学生，从而提高教学的效率等。With the development of artificial intelligence technology, the application scenarios of artificial intelligence are becoming more and more extensive. In scenarios such as education and teaching, artificial intelligence technology can also be used to assist teachers and/or students, thereby improving the efficiency of teaching.

相关技术中，可以通过对学生的作答结果进行图像识别从而辅助教师实现作业的批改。但是这种方法识别作答结果的准确性较低，进而导致作业批改的准确性较低。In related technologies, it is possible to assist teachers to correct homework by performing image recognition on students' answering results. However, this method has low accuracy in identifying answer results, which in turn leads to low accuracy in homework correction.

发明内容Contents of the invention

为了解决上述技术问题或者至少部分地解决上述技术问题，本公开提供了一种作业批改方法、装置、设备及介质。In order to solve the above technical problems or at least partly solve the above technical problems, the present disclosure provides a method, device, equipment and medium for correcting assignments.

根据本公开的一方面，提供了一种作业批改方法，包括：获取当前作业的作答结果图像；对作答结果图像进行文本识别，得到初始作答结果；获取当前作业的题型，并基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果；基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果。According to an aspect of the present disclosure, there is provided a homework correction method, including: acquiring the answer result image of the current homework; performing text recognition on the answer result image to obtain the initial answer result; acquiring the question type of the current homework, and based on the current homework The question type adjusts the initial answer result to obtain the target answer result; based on the current homework question type and the current homework standard answer result, the target answer result is judged to determine the current homework correction result.

根据本公开的另一方面，提供了一种作业批改装置，所述装置，包括：图像获取模块，用于获取当前作业的作答结果图像；文本识别模块，用于对作答结果图像进行文本识别，得到初始作答结果；结果优化模块，用于获取当前作业的题型，并基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果；结果判断模块，用于基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果。According to another aspect of the present disclosure, a homework correction device is provided, and the device includes: an image acquisition module for acquiring an answer result image of the current homework; a text recognition module for performing text recognition on the answer result image, Obtain the initial answer result; the result optimization module is used to obtain the question type of the current homework, and adjust the initial answer result based on the question type of the current homework to obtain the target answer result; the result judgment module is used to use the question type based on the current homework Based on the standard answer results of the current homework, the target answer results are judged to determine the correcting results of the current homework.

根据本公开的另一方面，提供了一种电子设备，包括：处理器；以及存储程序的存储器，其中，所述程序包括指令，所述指令在由所述处理器执行时使所述处理器执行上述作业批改方法或者上述作业批改方法。According to another aspect of the present disclosure, there is provided an electronic device comprising: a processor; and a memory storing a program, wherein the program includes instructions which, when executed by the processor, cause the processor to Execute the above-mentioned homework correction method or the above-mentioned homework correction method.

根据本公开的另一方面，提供了一种计算机可读存储介质，所述存储介质存储有计算机程序，所述计算机程序用于执行上述作业批改方法或者上述作业批改方法。According to another aspect of the present disclosure, a computer-readable storage medium is provided, the storage medium stores a computer program, and the computer program is used to execute the above-mentioned job correction method or the above-mentioned job correction method.

本公开实施例中提供的上述作业批改方法及装置，能够获取当前作业的作答结果图像；对作答结果图像进行文本识别，得到初始作答结果；获取当前作业的题型，并基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果；基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果。采用上述技术方案，针对不同题型的特点存在对应的针对初始作答结果的调整处理方法，从而能够更有针对性的对初始作答结果进行调整，提高了最终获得的目标作答结果的准确性，并且根据当前作业的题型以及标准作答结果确定目标作答结果的批改结果，提高了批改过程中针对各种题型的适配性以及批改结果的准确性。The above-mentioned homework correction method and device provided in the embodiments of the present disclosure can obtain the answer result image of the current homework; perform text recognition on the answer result image to obtain the initial answer result; acquire the question type of the current homework, and based on the question type of the current homework Adjust the initial answer result to obtain the target answer result; based on the question type of the current homework and the standard answer result of the current homework, judge the target answer result and determine the correcting result of the current homework. By adopting the above technical solution, there are corresponding adjustment processing methods for the initial answer results according to the characteristics of different question types, so that the initial answer results can be adjusted in a more targeted manner, and the accuracy of the final target answer results is improved, and The correction result of the target answer result is determined according to the question type of the current homework and the standard answer result, which improves the adaptability to various question types and the accuracy of the correction result during the correction process.

应当理解，本部分所描述的内容并非旨在标识本公开的实施例的关键或重要特征，也不用于限制本公开的范围。本公开的其它特征将通过以下的说明书而变得容易理解。It should be understood that what is described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will be readily understood through the following description.

附图说明Description of drawings

此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本公开的实施例，并与说明书一起用于解释本公开的原理。The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description serve to explain the principles of the disclosure.

为了更清楚地说明本公开实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，对于本领域普通技术人员而言，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, In other words, other drawings can also be obtained from these drawings without paying creative labor.

图1为本公开实施例提供的一种作业批改方法的流程示意图；FIG. 1 is a schematic flow diagram of a method for correcting a job provided by an embodiment of the present disclosure;

图2为本公开实施例提供的一种作业的示意图；FIG. 2 is a schematic diagram of an operation provided by an embodiment of the present disclosure;

图3为本公开实施例提供的一种批改页面的示意图；FIG. 3 is a schematic diagram of a correction page provided by an embodiment of the present disclosure;

图4为本公开实施例提供的一种扩大图像采集区域的示意图；FIG. 4 is a schematic diagram of an enlarged image acquisition area provided by an embodiment of the present disclosure;

图5为本公开实施例提供的一种标准作答规范信息的示意图；FIG. 5 is a schematic diagram of standard answer specification information provided by an embodiment of the present disclosure;

图6为本公开实施例提供的另一种作业批改方法的流程示意图；FIG. 6 is a schematic flowchart of another method for correcting assignments provided by an embodiment of the present disclosure;

图7为本公开实施例提供的又一种作业批改方法的流程示意图；FIG. 7 is a schematic flowchart of another method for correcting a job provided by an embodiment of the present disclosure;

图8为本公开实施例提供的再一种作业批改方法的流程示意图；FIG. 8 is a schematic flowchart of another method for correcting a job provided by an embodiment of the present disclosure;

图9为本公开实施例提供的一种阅卷方法的示意图；FIG. 9 is a schematic diagram of a marking method provided by an embodiment of the present disclosure;

图10为本公开实施例提供的一种标准阅卷规范的示意图；FIG. 10 is a schematic diagram of a standard marking specification provided by an embodiment of the present disclosure;

图11为本公开实施例提供的一种作业批改方法的示意图；FIG. 11 is a schematic diagram of a method for correcting assignments provided by an embodiment of the present disclosure;

图12为本公开实施例提供的一种作业批改装置的结构示意图；Fig. 12 is a schematic structural diagram of a job correction device provided by an embodiment of the present disclosure;

图13为本公开实施例提供的一种电子设备的结构示意图。FIG. 13 is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure.

具体实施方式Detailed ways

下面将参照附图更详细地描述本公开的实施例。虽然附图中显示了本公开的某些实施例，然而应当理解的是，本公开可以通过各种形式来实现，而且不应该被解释为限于这里阐述的实施例，相反提供这些实施例是为了更加透彻和完整地理解本公开。应当理解的是，本公开的附图及实施例仅用于示例性作用，并非用于限制本公开的保护范围。Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the drawings, it should be understood that the disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein; A more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for exemplary purposes only, and are not intended to limit the protection scope of the present disclosure.

应当理解，本公开的方法实施方式中记载的各个步骤可以按照不同的顺序执行，和/或并行执行。此外，方法实施方式可以包括附加的步骤和/或省略执行示出的步骤。本公开的范围在此方面不受限制。It should be understood that the various steps described in the method implementations of the present disclosure may be executed in different orders, and/or executed in parallel. Additionally, method embodiments may include additional steps and/or omit performing illustrated steps. The scope of the present disclosure is not limited in this regard.

本公开使用的术语“包括”及其变形是开放性包括，即“包括但不限于”。术语“基于”是“至少部分地基于”。术语“一个实施例”表示“至少一个实施例”；术语“另一实施例”表示“至少一个另外的实施例”；术语“一些实施例”表示“至少一些实施例”。其他术语的相关定义将在下文描述中给出。需要注意，本公开中提及的“第一”、“第二”等概念仅用于对不同的装置、模块或单元进行区分，并非用于限定这些装置、模块或单元所执行的功能的顺序或者相互依存关系。As used in this disclosure, the term "comprise" and its variations are open-ended, ie "including but not limited to". The term "based on" is "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one further embodiment"; the term "some embodiments" means "at least some embodiments." Relevant definitions of other terms will be given in the description below. It should be noted that concepts such as "first" and "second" mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the sequence of functions performed by these devices, modules or units or interdependence.

需要注意，本公开中提及的“一个”、“多个”的修饰是示意性而非限制性的，本领域技术人员应当理解，除非在上下文另有明确指出，否则应该理解为“一个或多个”。It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative and not restrictive, and those skilled in the art should understand that unless the context clearly indicates otherwise, it should be understood as "one or more" multiple".

为了改善以上至少一种问题，本公开实施例提供了一种作业批改方法、装置、设备及介质，为便于理解，以下进行阐述说明。In order to improve at least one of the above problems, embodiments of the present disclosure provide a method, device, device, and medium for job correction, which are described below for ease of understanding.

图1为本公开实施例提供的一种作业批改方法的流程示意图，该方法可以由作业批改装置执行，其中该装置可以采用软件和/或硬件实现，一般可集成在电子设备中。如图1所示，该方法主要包括：FIG. 1 is a schematic flowchart of a method for correcting a job provided by an embodiment of the present disclosure. The method can be executed by a device for correcting a job, wherein the device can be implemented by software and/or hardware, and can generally be integrated into an electronic device. As shown in Figure 1, the method mainly includes:

步骤101、获取当前作业的作答结果图像。Step 101 , acquiring an answer result image of the current assignment.

其中，当前作业为当前进行批改处理的作业，本实施例对该当前作业的题型不做限制，例如，该当前作业的题型可以包括但不限于：填空题、主观题、选择题中的至少一种。本实施例对该当前作业的学科也不限制，例如，该当前作业的学科可以包括但不限于：语文、数学、英语中的至少一种。作答结果图像可以为包括当前作业的作答内容的图像，该作答结果图像的获取可以根据用户书写工具的不同采用不同的方式实现。Wherein, the current assignment is the assignment currently undergoing correction processing. This embodiment does not limit the question type of the current assignment. For example, the question type of the current assignment may include but not limited to: fill in the blank questions, subjective questions, multiple choice questions at least one. This embodiment also does not limit the subject of the current job, for example, the subject of the current job may include but not limited to: at least one of Chinese, mathematics, and English. The answer result image may be an image including the answer content of the current assignment, and the acquisition of the answer result image may be implemented in different ways according to different writing tools of the user.

在本实施例中，用户使用书写工具在作答区域作答，该书写工具有多种，本实施例不做限制。In this embodiment, the user uses a writing tool to answer in the answer area. There are many kinds of writing tools, which are not limited in this embodiment.

在一些实施例中，获取当前作业的作答结果图像，可以包括：获取当前作业的图像，并根据所述当前作业图像中预先设定预先确定的作答区域的区域坐标提取得到所述作答结果图像。In some embodiments, acquiring the answer result image of the current job may include: acquiring the image of the current job, and extracting the answer result image according to area coordinates of a preset predetermined answer area in the current job image.

其中，本实施例中的书写工具可以包括圆珠笔和/或铅笔等不具备书写轨迹采集功能的普通笔。作答区域可以为提供给作答者对题目进行作答的区域，该作答区域可以根据具体的题型进行设置，不同题型可以设置不同的作答区域，本实施例不做限制。当前作业图像包括了作答区域的图像以及作答区域以外的其他图像。Wherein, the writing instrument in this embodiment may include a ballpoint pen and/or a pencil and other ordinary pens that do not have the function of collecting writing traces. The answering area may be an area provided for the respondent to answer questions, and the answering area may be set according to specific question types, and different answering areas may be set for different question types, which is not limited in this embodiment. The current job image includes the image of the answering area and other images outside the answering area.

作答区域的区域坐标为预先设定的表征作答区域位置的参数，图2为本公开实施例提供的一种作业的示意图，如图2所示，若当前作业的题型为选择题，作答结果图像可以为用户在作答区域内填写的作答内容的图像，其中，该作答区域可以设置在题目区域的一侧，例如图2中区域201-205均为作答区域，区域207-211均为题目区域，作答区域和题目区域的形状均可以设置为矩形，每个区域的位置通过矩形四个顶点的坐标表征，可以将该矩形作答区域的顶点坐标设置为作答区域的区域坐标。基于各题目可以建立作答区域和题目区域的对应关系，例如区域207和区域201相对应。可以理解的是，在生成上述作业的过程中可以记录每个题目区域及作答区域在web端生成的作业页面上的坐标，并将其转化成A4纸张中的坐标，进而可以将每个题目区域的坐标关联作答区域的坐标存储到数据库中。如图2所示，在该作业上还可以设置有二维码区域206，该二维码区域206中可以包括当前作业的二维码，通过扫描该二维码能够获取学生识别号（Identity Document，ID）、试卷ID、试卷页面ID等消息中的至少一个。The area coordinates of the answer area are preset parameters representing the position of the answer area. Figure 2 is a schematic diagram of a job provided by an embodiment of the present disclosure. As shown in Figure 2, if the question type of the current job is a multiple-choice question, the answer result The image can be an image of the answer content filled in by the user in the answer area, wherein the answer area can be set on one side of the question area, for example, areas 201-205 in Figure 2 are all answer areas, and areas 207-211 are all question areas , the shape of both the answer area and the question area can be set as a rectangle, the position of each area is represented by the coordinates of the four vertices of the rectangle, and the coordinates of the vertices of the rectangular answer area can be set as the area coordinates of the answer area. Based on each question, a corresponding relationship between the answer area and the question area can be established, for example, thearea 207 corresponds to thearea 201 . It can be understood that in the process of generating the above assignments, the coordinates of each question area and answer area on the homework page generated on the web can be recorded, and converted into coordinates in A4 paper, and then each question area can be The coordinates of the coordinates associated with the answering area are stored in the database. As shown in Figure 2, a two-dimensional code area 206 can also be set on the assignment, which can include the two-dimensional code of the current assignment, and the student identification number (Identity Document) can be obtained by scanning the two-dimensional code. , ID), test paper ID, test paper page ID and other messages.

在本实施例中，当书写工具不具备书写轨迹采集功能，该书写工具对应的当前作业图像的获取方法有多种，本实施例不做限制，例如可以通过具备拍照功能的图像采集设备获取该当前作业图像，该图像采集设备可以包括具备拍照功能的手机和/或数码相机等；或者，可以通过具备扫描功能的扫描设备获取该当前作业图像，该图像扫描设备可以包括双面扫描仪和/或单面扫描仪等，其中，若使用双面扫描仪能够免去了人工翻页的过程，提高了图像获取效率。获取当前作业图像之后，读取当前作业图像中预先设定的作答区域的区域坐标，根据该区域坐标进行图像提取，获得与作答区域对应的作答结果图像，从而缩小了后续图像处理的范围，提高了图像处理的效率以及准确性。In this embodiment, when the writing tool does not have the function of collecting the writing track, there are many ways to obtain the current job image corresponding to the writing tool, which is not limited in this embodiment. The current job image, the image acquisition device may include a mobile phone and/or digital camera with a camera function; or, the current job image may be obtained through a scanning device with a scanning function, and the image scanning device may include a double-sided scanner and/or Or a single-sided scanner, etc., wherein, if a double-sided scanner is used, the process of manually turning pages can be eliminated, and the image acquisition efficiency can be improved. After acquiring the current job image, read the area coordinates of the preset answer area in the current job image, perform image extraction according to the area coordinates, and obtain the answer result image corresponding to the answer area, thereby reducing the scope of subsequent image processing and improving Improve the efficiency and accuracy of image processing.

在一些实施例中，获取当前作业的作答结果图像，可以包括：获取作答工具在作答区域内的轨迹坐标，并根据轨迹坐标生成作答结果图像。In some embodiments, acquiring the answer result image of the current assignment may include: acquiring trajectory coordinates of the answer tool in the answer area, and generating the answer result image according to the trajectory coordinates.

其中，作答工具可以为能够记录自身运动轨迹的笔，该作答工具可以为点阵笔等具有压感功能的笔，若作答工具为点阵笔，该点阵笔可以配合点阵纸张使用，点阵笔在点阵纸张上书写时，产生压感同时点阵笔内的摄像头记录点阵笔的在每个轨迹点的坐标数据以及对应的时间戳，点阵笔获取该轨迹坐标以及时间戳之后，后续作业批改装置可以根据该轨迹坐标以及时间戳进行轨迹拟合，从而生成作答结果图像，通过该点阵笔还可以实现记录作答时长等功能。其中，轨迹坐标可以是用于记录点阵笔的运动路径的参数，可以包括点阵笔在移动过程中每个轨迹点的坐标数据和时间戳，轨迹坐标的数量可以为多个。本实施例对作答结果图像的格式不做限制，例如该作答结果图像的格式可以为位图文档。Wherein, the answering tool can be a pen capable of recording its own motion trajectory, and the answering tool can be a pen with pressure sensitivity such as a dot matrix pen. If the answering tool is a dot matrix pen, the dot matrix pen can be used in conjunction with dot matrix paper. When the matrix pen writes on the dot-matrix paper, it generates pressure sensitivity and the camera in the dot-matrix pen records the coordinate data of each track point of the dot-matrix pen and the corresponding time stamp. After the dot-matrix pen obtains the track coordinates and time stamp , the follow-up homework correction device can perform trajectory fitting according to the trajectory coordinates and time stamps, thereby generating an answer result image, and the dot-matrix pen can also realize functions such as recording the answering time. Wherein, the track coordinates may be parameters for recording the motion path of the dot matrix pen, and may include coordinate data and time stamps of each track point during the movement of the dot matrix pen, and the number of track coordinates may be multiple. This embodiment does not limit the format of the answer result image, for example, the format of the answer result image may be a bitmap file.

在本实施例中，用户可以使用点阵笔等作答工具在作答区域内进行作答，作业批改装置可以根据作答区域的区域坐标提取作答区域内的点阵笔的多个轨迹坐标，基于多个轨迹坐标的时间戳对坐标数据进行轨迹拟合，生成作答结果图像。从而在用户书写的过程中可以采集轨迹坐标并生成作答结果图像，简化了操作流程，并且提取作答区域内的轨迹坐标进行后续的文本识别操作，能够降低卷面上的其他因素（例如，用户求解过程等）对文本识别准确性的影响。In this embodiment, the user can use an answering tool such as a dot-matrix pen to answer in the answering area, and the homework correction device can extract a plurality of track coordinates of the dot-matrix pen in the answering area according to the area coordinates of the answering area, and based on a plurality of tracks The time stamp of the coordinates performs trajectory fitting on the coordinate data to generate an answer result image. In this way, the trajectory coordinates can be collected during the user's writing process and the answer result image can be generated, which simplifies the operation process, and the trajectory coordinates in the answer area can be extracted for subsequent text recognition operations, which can reduce other factors on the paper (for example, the user solves process, etc.) on the accuracy of text recognition.

步骤102、对作答结果图像进行文本识别，得到初始作答结果。Step 102, performing text recognition on the answer result image to obtain an initial answer result.

其中，初始作答结果为对作答结果图像进行文本识别直接获得的作答结果，该初始作答结果的准确性可能较低。Wherein, the initial answer result is an answer result directly obtained by performing text recognition on the answer result image, and the accuracy of the initial answer result may be low.

在一些实施例中，对作答结果图像进行文本识别，得到初始作答结果，包括：采用预先构建的文本识别模型对作答结果图像进行识别，得到初始作答结果。In some embodiments, performing text recognition on the answer result image to obtain the initial answer result includes: using a pre-built text recognition model to recognize the answer result image to obtain the initial answer result.

其中，文本识别模型中包括检测模型和识别模型，检测模型用于提取作答结果图像中的作答位置并输入至识别模型，作答位置为用户的作答内容对应的位置，该检测模型有多种，本实施例对该检测模型不做限制。识别模型用于对作答位置的图像进行文本识别得到初始作答结果，该识别模型有多种，本实施例对该识别模型不做限制，例如该识别模型可以为光学字符识别（Optical Character Recognition，OCR）模型。Among them, the text recognition model includes a detection model and a recognition model. The detection model is used to extract the answer position in the answer result image and input it to the recognition model. The answer position is the position corresponding to the user's answer content. There are many types of detection models. The embodiment does not limit the detection model. The recognition model is used to perform text recognition on the image of the answering position to obtain an initial answer result. There are various recognition models. This embodiment does not limit the recognition model. For example, the recognition model can be Optical Character Recognition (OCR). )Model.

作答区域中具有作答内容的区域为上述作答位置的区域，此外作答区域中还可以包括没有作答内容的空白区域。在本实施例中，通过文本识别模型中的检测模型从作答结果图像中确定作答位置的区域，并将该作答位置输入文本识别模型的识别模型中，之后该识别模型对该作答位置的区域进行文本识别，可以获得初始作答结果。上述方案中，通过先确定作答位置的区域，进一步缩小了图像识别范围，避免了题目等图像对文本识别的干扰，提高了文本识别的针对性。The area with answering content in the answering area is the area of the above-mentioned answering position, and the answering area may also include a blank area without answering content. In this embodiment, the area of the answering position is determined from the answer result image through the detection model in the text recognition model, and the answering position is input into the recognition model of the text recognition model, and then the recognition model performs an operation on the area of the answering position Text recognition, you can get the initial answer results. In the above solution, by first determining the area of the answering position, the scope of image recognition is further narrowed, the interference of images such as questions on text recognition is avoided, and the pertinence of text recognition is improved.

可选的，在当前作业的题型包括单选题且用户多次修改作答内容的情况下，在字符识别模型识别的初始作答结果中可能包括多个字符，该字符识别模型可以将识别到的字符以及各字符的置信度输出，以便于进行后续的初始作答结果优化。Optionally, when the question type of the current job includes multiple choice questions and the user modifies the answer content multiple times, the initial answer result recognized by the character recognition model may include multiple characters, and the character recognition model can convert the recognized Characters and the confidence output of each character, so as to optimize the subsequent initial answer results.

上述通过字符识别模型进行文本识别获得初始作答结果仅为示例，本实施例还可以通过其他方式来实现初始作答结果的获取，具体不限。The above-mentioned obtaining of the initial answer result through text recognition using the character recognition model is only an example. In this embodiment, other methods may also be used to obtain the initial answer result, which is not limited in detail.

步骤103、获取当前作业的题型，并基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果。Step 103: Obtain the question type of the current homework, and adjust the initial answer result based on the question type of the current homework to obtain the target answer result.

其中，当前作业的题型为当前作业所属的类型，在一些实施例中，当前作业的题型包括选择题、填空题或主观题。调整处理可以理解为根据初始作答结果的特征对初始作答结果进行调整，从而使得目标作答结果的准确率更高，不同题型的调整处理方法可以不同，本实施例不做限制。目标作答结果为经过对初始作答结果进行调整处理得到的最终作答结果。Wherein, the question type of the current homework is the type to which the current homework belongs. In some embodiments, the question type of the current homework includes multiple-choice questions, fill-in-the-blank questions or subjective questions. The adjustment processing can be understood as adjusting the initial answer result according to the characteristics of the initial answer result, so that the accuracy of the target answer result is higher. The adjustment processing methods of different question types can be different, and this embodiment does not limit it. The target answer result is the final answer result obtained by adjusting the initial answer result.

本公开实施例中，作业批改装置可以获取当前作业的题型，并根据当前作业的题型确定针对初始作答结果的调整处理方法，进而根据该调整处理方法对当前作业对应的初始作答结果进行调整处理，获得目标作答结果。In the embodiment of the present disclosure, the homework correction device can obtain the question type of the current homework, and determine the adjustment processing method for the initial answer result according to the question type of the current homework, and then adjust the initial answer result corresponding to the current homework according to the adjustment processing method processing to obtain the target answer result.

可以理解地，由于文本识别精确度不足、用户作答不规范等原因，初始作答结果的准确性可能较低，通过对初始作答结果的调整，可以获得准确性较高的目标作答结果，进而提升了作答结果识别的准确性。Understandably, due to insufficient text recognition accuracy and irregular user answers, the accuracy of the initial answer result may be low. By adjusting the initial answer result, a target answer result with higher accuracy can be obtained, thereby improving the Accuracy of answer recognition.

步骤104、基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果。Step 104 , based on the question type of the current homework and the standard answer result of the current homework, judge the target answer result, and determine the correcting result of the current homework.

其中，标准作答结果为当前作业对应的标准答案，该标准作答结果可以为预先设置的。批改结果可以为用于表征目标作答结果和标准作答结果匹配情况的参数，该批改结果的类型有多种，本实施例不做限制，例如，该批改结果的类型可以为二分类类型，即该批改结果包括正确和错误；该批改结果的类型也可以为分数类型，可以在目标作答结果和标准作答结果匹配度越高的情况下，该批改结果分数越高。Wherein, the standard answer result is a standard answer corresponding to the current assignment, and the standard answer result may be preset. The correction result can be a parameter used to characterize the matching between the target answer result and the standard answer result. There are many types of the correction result, which are not limited in this embodiment. For example, the type of the correction result can be a binary classification type, that is, the The correction result includes correctness and error; the type of the correction result may also be a score type, and the higher the matching degree between the target answer result and the standard answer result, the higher the score of the correction result.

以当前作业的题型为选择题、填空题、主观题中的文科主观题以及主观题中的理科主观题为例，说明上述步骤：Taking the question types of the current assignment as multiple-choice questions, filling in the blanks, subjective questions of liberal arts in subjective questions, and subjective questions of science in subjective questions as examples, the above steps are explained:

在一些实施例中，若当前作业的题型为选择题，基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果，包括：将目标作答结果与标准作答结果进行对比；根据对比结果确定当前作业的批改结果。当目标作答结果等于标准作答结果，则当前作业的批改结果为正确；否则当前作业的批改结果为错误。In some embodiments, if the question type of the current homework is a multiple-choice question, based on the question type of the current homework and the standard answer result of the current homework, the target answer result is judged to determine the correcting result of the current homework, including: the target answer result Compare with the standard answer results; determine the correcting result of the current assignment based on the comparison results. When the target answering result is equal to the standard answering result, the correcting result of the current homework is correct; otherwise, the correcting result of the current homework is wrong.

在当前作业的题型为单项选择题的情况下，可以将目标作答结果和标准作答结果进行一致性对比，判断目标作答结果和标准作答结果是否一致，若对比结果为一致，则确定当前作业的批改结果为正确；若对比结果为不一致，则确定当前作业的批改结果为错误。When the question type of the current homework is a multiple-choice question, you can compare the target answer result with the standard answer result to determine whether the target answer result is consistent with the standard answer result. If the comparison results are consistent, determine the current homework result. The correction result is correct; if the comparison result is inconsistent, it is determined that the correction result of the current job is wrong.

在当前作业的题型为多项选择题的情况下，可以将目标作答结果和标准作答结果进行一致性对比，判断目标作答结果和标准作答结果是否一致，若对比结果为一致，则确定当前作业的批改结果为正确；若比对结果为目标作答结果为标准作答结果的子集，则确定当前作业的批改结果为部分正确（即半对半错）；若比对结果为目标作答结果中的部分不属于标准作答结果或者目标作答结果完全不属于标准作答结果，则确定当前作业的批改结果为错误。When the question type of the current homework is a multiple-choice question, you can compare the target answer result with the standard answer result to determine whether the target answer result is consistent with the standard answer result. If the comparison result is consistent, determine the current homework If the result of the comparison is that the target answer result is a subset of the standard answer results, then it is determined that the correction result of the current assignment is partially correct (that is, half right and half wrong); if the comparison result is a subset of the target answer results If part of the answer does not belong to the standard answer result or the target answer result does not belong to the standard answer result at all, it is determined that the correcting result of the current assignment is an error.

在一些实施例中，若当前作业的题型为填空题，基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果，包括：将目标作答结果与标准作答结果进行数据结构的对比，或者字符数量的对比；根据对比结果确定当前作业的批改结果；其中，数据结构用于表征作答结果包括的数据类型，该数据结构可以根据应用场景进行设置，本实施例不做限制，例如该数据结果可以包括数字、符号、数字与符号混合等；字符数量用于表征作答结果包括的字符的数量。In some embodiments, if the question type of the current homework is a fill-in-the-blank question, based on the question type of the current homework and the standard answer result of the current homework, the target answer result is judged to determine the correction result of the current homework, including: the target answer result Compare the data structure with the standard answer result, or compare the number of characters; determine the correction result of the current job according to the comparison result; wherein, the data structure is used to represent the data type included in the answer result, and the data structure can be set according to the application scenario. This embodiment is not limited, for example, the data result may include numbers, symbols, numbers and symbols mixed, etc.; the number of characters is used to represent the number of characters included in the answer result.

在当前作业的题型为填空题的情况下，确定批改结果的方法至少包括以下两种：一种方法中，可以获取目标作答结果的数据结构以及标准作答结果的数据结构，判断目标作答结果的数据结构与标准作答结果的数据结构是否相同，若相同，则当前作业的批改结果为正确；若不相同，则当前作业的批改结果为错误。例如，若标准作答结果的数据结构为数字，目标作答结果的数据结构包括数字以及其他格式，则确定判改结果为错误；若标准作答结果的数据结构包括数字和符合，目标作答结果的数据结构为数字，则确定判改结果为错误。另一种方法中，可以获取目标作答结果的字符数量以及标准作答结果的字符数量，判断目标作答结果与标准作答结果的字符数量是否相同，若相同，则当前作业的批改结果为正确；若不相同，则当前作业的批改结果为错误。In the case that the question type of the current homework is a fill-in-the-blank question, the methods for determining the correction result include at least the following two methods: In one method, the data structure of the target answer result and the data structure of the standard answer result can be obtained, and the data structure of the target answer result can be judged. Whether the data structure is the same as the data structure of the standard answer result, if they are the same, the correcting result of the current assignment is correct; if not, the correcting result of the current assignment is incorrect. For example, if the data structure of the standard answer result is a number, and the data structure of the target answer result includes numbers and other formats, then it is determined that the judgment result is an error; if the data structure of the standard answer result includes numbers and symbols, the data structure of the target answer result is a number, then it is determined that the judgment result is an error. In another method, the number of characters of the target answer result and the number of characters of the standard answer result can be obtained, and it is judged whether the number of characters of the target answer result and the standard answer result are the same. If they are the same, the correcting result of the current assignment is correct; If they are the same, the correction result of the current job is an error.

在一些实施例中，若当前作业的题型为主观题中的文科主观题，基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果，包括：基于预设批改算法识别所述目标作答结果的至少一种批改参数，批改参数包括字数、标题、错别字、语法错误、搭配不当和结构错误中的至少一项；展示标准作答结果以及至少一种批改参数以及当前作业的标准作答结果，获取用户基于标准作答结果以及至少一种批改参数输入的当前作业的批改结果。其中，预设批改算法可以为能够对参数进行识别的算法，本实施例对该预设批改算法的模型结构、具体训练方法等不做限制。批改参数可以为一个或多个维度指导用户对目标作答结果进行修改的参数，本实施例对该批改参数的获取方式不做限制，例如可以将目标作答结果输入参数生成模型中，生成对应的批改参数。批改结果为用户对目标作答结果进行批改之后获得的作答结果。In some embodiments, if the question type of the current homework is a liberal arts subjective question in the subjective questions, based on the question type of the current homework and the standard answer result of the current homework, the target answer result is judged to determine the correcting result of the current homework, including : Identify at least one correction parameter of the target answer result based on a preset correction algorithm, the correction parameter includes at least one of word count, title, typo, grammatical error, improper collocation and structural error; display standard answer results and at least one The correction parameters and the standard answer result of the current assignment are used to obtain the correction result of the current assignment entered by the user based on the standard answer result and at least one correction parameter. Wherein, the preset correction algorithm may be an algorithm capable of identifying parameters, and this embodiment does not limit the model structure, specific training method, etc. of the preset correction algorithm. The correction parameter can be a parameter that guides the user to modify the target answer result in one or more dimensions. This embodiment does not limit the method of obtaining the correction parameter. For example, the target answer result can be input into the parameter generation model to generate a corresponding correction. parameter. The correction result is an answer result obtained after the user corrects the target answer result.

在当前作业的题型为文科主观题的情况下，作业批改装置可以将目标作答结果输入参数生成模型中，参数生成模型分析该目标作答结果，确定目标作答结果的至少一个批改参数，该批改参数包括但不限于：字数、标题、错别字、语法错误、搭配不当和结构错误中的至少一项，并且基于预设批改算法识别上述批改参数。进一步地，将当前作业的标准作答结果以及至少一种批改参数展示给用户，通过标准作答结果能够较使用户直观地理解对应的批改参数。用户基于标准作答结果以及批改参数输入批改结果，获取用户输入的当前作业的批改结果，其中，用户可以为教师。如图3所示，图3为本公开实施例提供的一种批改页面的示意图，图3中展示了一个批改页面300，该批改页面300中可以包括批改参数以及批改参数对应的数字标识①至⑨，批改参数包括总字数、是否存在标题、错别字、语法错误、搭配不当和结构错误，并且在目标作答结果中可以标记有与批改参数对应的数字标识。In the case that the question type of the current job is a subjective question in liberal arts, the homework correction device can input the target answer result into the parameter generation model, and the parameter generation model analyzes the target answer result to determine at least one correction parameter of the target answer result, the correction parameter Including but not limited to: at least one of word count, title, typo, grammatical error, improper collocation and structural error, and identifying the above correction parameters based on a preset correction algorithm. Further, the standard answer result of the current job and at least one correction parameter are displayed to the user, and the user can intuitively understand the corresponding correction parameter through the standard answer result. The user inputs the marking result based on the standard answering result and marking parameters, and obtains the marking result of the current homework input by the user, wherein the user may be a teacher. As shown in FIG. 3, FIG. 3 is a schematic diagram of a correction page provided by an embodiment of the present disclosure. FIG. ⑨, Correction parameters include the total number of words, whether there are titles, typos, grammatical errors, improper collocations, and structural errors, and the target answer results can be marked with numbers corresponding to the correcting parameters.

在一些实施例中，若当前作业的题型为主观题中的理科主观题，基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果，包括：确定目标作答结果中位于最后一位的当前最终结果；将所述目标作答结果的当前最终结果与所述标准作答结果中的标准最终结果进行对比；当对比结果为当前最终结果与标准最终结果相同，则当前作业的批改结果确定为正确；否则当前作业的批改结果确定为错误。In some embodiments, if the question type of the current homework is the subjective question of science in the subjective questions, based on the question type of the current homework and the standard answer result of the current homework, the target answer result is judged, and the correcting result of the current homework is determined, including : determine the current final result at the last place in the target answer result; compare the current final result of the target answer result with the standard final result in the standard answer result; when the comparison result is the current final result and the standard final result If they are the same, the correction result of the current job is determined to be correct; otherwise, the correction result of the current job is determined to be wrong.

其中，目标作答结果中的位于最后一位的当前最终结果过可以理解为该理科主观题在目标作答结果中的计算结果，例如，若当前作业为应用题，目标作答结果中可以包括该应用题的分析、计算过程以及计算结果，该应用题的计算结果为当前最终结果。标准作答结果的标准最终结果可以理解为该理科主观题在标准作答结果中的计算结果。Among them, the current final result at the last place in the target answer result can be understood as the calculation result of the science subjective question in the target answer result. For example, if the current homework is a practical question, the target answer result can include the practical question. The analysis, calculation process and calculation results of the application problem are the current final results. The standard final result of the standard answer result can be understood as the calculation result of the science subjective question in the standard answer result.

在当前作业的题型为理科主观题的情况下，作业批改装置可以获取目标作答结果中位于最后一位的当前最终结果以及标准作答结果中位于最后一位的标准最终结果，判断该当前最终结果与该标准最终结果是否相同，若相同；则当前作业的批改结果为正确；若不相同，则当前作业的批改结果为错误。一种可选的实施方式中，可以提供用于书写当前作业计算结果的结果填写区域，对该结果填写区域的图像进行文本识别获得目标作业结果中的当前最终结果。In the case that the question type of the current homework is a science subjective question, the homework correction device can obtain the current final result that is the last in the target answer result and the standard final result that is the last in the standard answer result, and judge the current final result Whether it is the same as the final result of the standard, if it is the same, the correction result of the current job is correct; if not, the correction result of the current job is wrong. In an optional implementation manner, a result filling area for writing the calculation result of the current operation may be provided, and text recognition is performed on the image of the result filling area to obtain the current final result of the target operation result.

上述方案中，基于题型的特性确定相应的作业批改结果，从而提升了批改结果的准确性，也丰富了本方法能够适配的题型，提高了本方法的适配性。In the above solution, the corresponding homework correction results are determined based on the characteristics of the question types, thereby improving the accuracy of the correction results, enriching the question types that this method can adapt to, and improving the adaptability of this method.

本公开实施例提供的作业批改方案，获取当前作业的作答结果图像；对作答结果图像进行文本识别，得到初始作答结果；获取当前作业的题型，并基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果；基于当前作业的题型和当前作业的标准作答结果，对目标作答结果进行判断，确定当前作业的批改结果。采用上述技术方案，针对不同题型的特点存在对应的针对初始作答结果的调整处理方法，从而能够更有针对性的对初始作答结果进行调整，提高了最终获得的目标作答结果的准确性，并且根据当前作业的题型以及标准作答结果确定目标作答结果的批改结果，提高了批改过程中针对各种题型的适配性以及批改结果的准确性。The homework correction scheme provided by the embodiments of the present disclosure obtains the answer result image of the current homework; performs text recognition on the answer result image to obtain the initial answer result; acquires the question type of the current homework, and performs an initial answer based on the question type of the current homework Adjust the processing to obtain the target answer result; based on the question type of the current homework and the standard answer result of the current homework, judge the target answer result and determine the correcting result of the current homework. By adopting the above technical solution, there are corresponding adjustment processing methods for the initial answer results according to the characteristics of different question types, so that the initial answer results can be adjusted in a more targeted manner, and the accuracy of the final target answer results is improved, and The correction result of the target answer result is determined according to the question type of the current homework and the standard answer result, which improves the adaptability to various question types and the accuracy of the correction result during the correction process.

在一些实施例中，若当前作业的题型是选择题，调整处理包括字符调整处理、数字调整处理、数量调整处理中的至少一种；若当前作业的题型是填空题或主观题，则调整处理包括区域调整处理或单位调整处理。In some embodiments, if the question type of the current assignment is a multiple-choice question, the adjustment process includes at least one of character adjustment processing, number adjustment processing, and quantity adjustment processing; if the question type of the current assignment is a fill-in-the-blank question or subjective question, then The adjustment processing includes area adjustment processing or unit adjustment processing.

其中，字符调整处理可以为根据绑定字符库中的绑定字符与标准结果的对应关系对初始作答结果进行调整；数字调整处理可以为基于初始作答结果包括的数字对初始作答结果进行调整；数量调整处理可以为根据初始作答结果包括的字符数量对初始作答结果进行调整处理。需要说明的是，当前作业的题型是选择题，则对应的上述调整处理方法可以交叉使用，例如可以使用字符调整处理对初始作答结果进行调整之后，使用数字调整处理对字符调整处理获得的调整结果进行进一步的调整。区域调整处理可以为对初始作答结果对应的作答结果图像的区域进行调整；单位调整处理可以为对初始作答结果包括的计量单位的语言格式进行调整。与题型为选择题对应的调整处理方法类似，若当前作业的题型是填空题或主观题对应的上述调整处理方法也可以交叉使用，例如可以使用区域调整处理对初始作答结果进行优化之后，使用数字调整处理对区域调整处理获得的调整结果进行进一步的优化。Among them, the character adjustment process can be to adjust the initial answer result according to the corresponding relationship between the bound characters in the bound character library and the standard result; the number adjustment process can be to adjust the initial answer result based on the numbers included in the initial answer result; The adjustment process may be to adjust the initial answer result according to the number of characters included in the initial answer result. It should be noted that if the question type of the current homework is a multiple-choice question, the corresponding adjustment processing methods above can be used interchangeably. For example, after adjusting the initial answer result by using character adjustment processing, use number adjustment processing to adjust the result obtained by character adjustment processing. The results are further adjusted. The area adjustment process may be to adjust the area of the answer result image corresponding to the initial answer result; the unit adjustment process may be to adjust the language format of the measurement unit included in the initial answer result. Similar to the adjustment processing method corresponding to the multiple-choice question type, if the question type of the current homework is a fill-in-the-blank or subjective question, the above adjustment processing methods corresponding to the above-mentioned adjustment processing methods can also be used interchangeably. The adjustment results obtained by the area adjustment processing are further optimized using digital adjustment processing.

下面，基于字符调整处理、数字调整处理、数量调整处理、区域调整处理、单位调整处理对基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果，进行说明：Next, based on the character adjustment processing, number adjustment processing, quantity adjustment processing, area adjustment processing, and unit adjustment processing, the initial answer result is adjusted based on the question type of the current assignment, and the target answer result is obtained, which will be explained:

在一种实施方式中，基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果，包括：若当前作业的题型是选择题，将初始作答结果在绑定字符库中进行匹配，匹配则得到目标绑定字符，其中，绑定字符库中包括多个预设的绑定字符；基于预设的绑定字符与标准结果的关联关系，确定所述目标绑定字符对应的目标标准结果，将目标标准结果确定为目标作答结果，其中，绑定字符与标准结果的关联关系用于确定多个预设的绑定字符与标准结果之间的对应关系，一个所述标准结果关联至少一个绑定字符。In one embodiment, the initial answer result is adjusted based on the question type of the current homework to obtain the target answer result, including: if the question type of the current homework is a multiple-choice question, matching the initial answer result in the bound character library , the target binding character is obtained by matching, wherein the binding character library includes a plurality of preset binding characters; based on the association relationship between the preset binding character and the standard result, the target corresponding to the target binding character is determined Standard result, determining the target standard result as the target answer result, wherein the association between bound characters and standard results is used to determine the corresponding relationship between multiple preset bound characters and standard results, and one standard result is associated At least one bound character.

其中，预设的绑定字符可以为一个或多个字符构成的具有绑定关系的字符组，绑定字符库可以为包括多个绑定字符的数据库。预设的绑定字符与标准结果的关联关系用于记录绑定字符与标准结果之间的转化对应关系。标准结果为绑定字符对应的转化结果，一个标准结果可以对应一个或多个绑定字符，该标准结果中的字符数量少于或等于绑定字符的字符数量。目标绑定字符可以为绑定字符库中与初始作答结果匹配成功的绑定字符，目标标准结果为根据关联关系确定的与该目标绑定字符匹配的标准结果。Wherein, the preset binding characters may be a character group with a binding relationship composed of one or more characters, and the binding character database may be a database including a plurality of binding characters. The preset association relationship between binding characters and standard results is used to record the conversion correspondence between binding characters and standard results. The standard result is the conversion result corresponding to the bound character. A standard result can correspond to one or more bound characters, and the number of characters in the standard result is less than or equal to the number of bound characters. The target binding character can be a binding character in the binding character library that successfully matches the initial answer result, and the target standard result is a standard result determined according to the association relationship that matches the target binding character.

在本实施例中，在当前作业的题型是选择题时，调整处理可以为字符调整处理，关于该字符调整处理具体说明如下：In this embodiment, when the question type of the current assignment is a multiple-choice question, the adjustment processing can be character adjustment processing, and the specific description of the character adjustment processing is as follows:

用户可以预先建立包括一个或多个预设绑定字符的绑定字符库，并且预先设定绑定字符与标准结果之间的关联关系。在确定初始作答结果之后，可以根据该初始作答结果在绑定字符库中进行查询操作，从而确定与该初始作答结果匹配的目标绑定字符。进一步地，根据该目标绑定字符在绑定字符与标准结果的关联关系进行查询操作，从而确定与该目标绑定字符对应的目标标准结果，将该目标标准结果确定为目标作答结果。可以理解地，由于用户书写的字符不规范等原因（例如，一个字符内部的空隙可能过大），对该字符进行文本识别时，可能将该字符识别为其他字符。本实施例对预设的绑定字符与标准结果的关联关系不做限制，举例而言，绑定字符可以包括多个字符，例如1和3可以为绑定字符，1和7可以为绑定字符，其中1和3对应的标准结果为B，1和7对应的标准结果为D。绑定字符可以包括单个字符，例如，绑定字符和标准结果的对应关系可以包括：H对应A，8对应B，L或e或①对应C，O或0或P或p或d对应D，7或十或+对应T，干对应F，2或√对应对钩字符，义或人对应叉号字符，0或O或@对应不完全正确字符。The user can pre-establish a binding character library including one or more preset binding characters, and preset the association relationship between the binding characters and the standard results. After the initial answer result is determined, a query operation can be performed in the bound character library according to the initial answer result, so as to determine the target bound character matching the initial answer result. Further, a query operation is performed according to the relationship between the bound character and the standard result of the target bound character, so as to determine the target standard result corresponding to the target bound character, and determine the target standard result as the target answer result. Understandably, due to reasons such as irregular characters written by the user (for example, the space inside a character may be too large), when performing text recognition on the character, the character may be recognized as other characters. This embodiment does not limit the association between the preset binding characters and standard results. For example, the binding characters can include multiple characters, for example, 1 and 3 can be binding characters, and 1 and 7 can be binding characters. characters, where the standard result corresponding to 1 and 3 is B, and the standard result corresponding to 1 and 7 is D. Binding characters can include a single character, for example, the correspondence between binding characters and standard results can include: H corresponds to A, 8 corresponds to B, L or e or ① corresponds to C, O or 0 or P or p or d corresponds to D, 7 or ten or + corresponds to T, stem corresponds to F, 2 or √ corresponds to the hook character, justice or person corresponds to the cross character, 0 or O or @ corresponds to the incorrect character.

上述方案中，基于字符调整处理对用户书写不规范造成的文本识别错误进行了调整，从而提高了目标作答结果的准确性。In the above solution, text recognition errors caused by irregular writing of the user are adjusted based on character adjustment processing, thereby improving the accuracy of the target answer result.

在一种实施方式中，基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果，包括：若当前作业的题型是选择题，对初始作答结果进行数字识别；若识别到数字，删除初始作答结果中的数字，得到目标作答结果；或者，当初始作答结果中的数字判定为预设数字且字符数量为2，则删除初始作答结果中的预设数字，得到目标作答结果。其中，预设数字有多种，本实施例不做限制，例如可以将该预设数字设置为1。In one embodiment, the initial answer result is adjusted based on the question type of the current homework to obtain the target answer result, including: if the question type of the current homework is a multiple-choice question, digital recognition is performed on the initial answer result; , delete the number in the initial answer result to obtain the target answer result; or, when the number in the initial answer result is determined to be a preset number and the number of characters is 2, delete the preset number in the initial answer result to obtain the target answer result. There are various preset numbers, which are not limited in this embodiment, for example, the preset number may be set to 1.

在本实施例中，在当前作业的题型是选择题时，调整处理可以为数字调整处理，关于该数字调整处理具体说明如下：In this embodiment, when the question type of the current assignment is a multiple-choice question, the adjustment processing can be digital adjustment processing, and the specific description of the digital adjustment processing is as follows:

在本实施例中，数字调整处理可以包括两种调整处理方法，其中一种调整处理方法包括：判断初始作答结果中包括的各字符是否包括数字，若是则确定初始作答结果中能够识别到数字，则删除该初始作答结果中的数字，获得不包括数字的目标作答结果。示例性的，若当前作业的题型为选项为字母的选择题，该作答结果中不应该包括数字，该数字可能为误识别结果，因而将初始作答结果中的数字删除，获得更准确的目标作答结果。In this embodiment, the number adjustment processing may include two adjustment processing methods, one of which includes: judging whether each character included in the initial answer result includes a number, and if so, determining that the number can be recognized in the initial answer result, The number in the initial answer result is then deleted to obtain a target answer result that does not include numbers. Exemplarily, if the question type of the current assignment is a multiple-choice question whose options are letters, the answer result should not include a number, which may be a misrecognition result, so the number in the initial answer result is deleted to obtain a more accurate goal Answer results.

另一种数字调整处理包括：判断初始作答结果中是否包括预设数字，且初始作答结果中包括的字符数量是否为2，若初始作答结果中包括预设数字且字符数量为2，则在该初始作答结果中删除预设数字，获得不包括预设数字的目标作答结果。可以理解地，在单选题等应用场景中，作答结果中包括的字符数量应该为一个，但是在识别获得作答结果的过程中，作答区域的边框等容易对文本识别形成干扰，从而误识别出预设数字。例如，在初始作答结果中可能将作答区域的边框识别为1，因而当初始作答结果中包括的字符数量为2且包括1时，将初始作答结果中的1删除，获得目标作答结果。Another number adjustment process includes: judging whether the initial answering result includes a preset number, and whether the number of characters included in the initial answering result is 2, if the initial answering result includes a preset number and the number of characters is 2, then in the The preset numbers are deleted from the initial answer results, and the target answer results that do not include the preset numbers are obtained. It is understandable that in application scenarios such as multiple-choice questions, the number of characters included in the answer result should be one, but in the process of recognizing the answer result, the frame of the answer area may easily interfere with the text recognition, thereby misidentifying Preset numbers. For example, the border of the answer area may be identified as 1 in the initial answer result, so when the number of characters included in the initial answer result is 2 and includes 1, 1 in the initial answer result is deleted to obtain the target answer result.

上述方案中，基于数字调整处理对初始作答结果中的数字进行了调整，删除了初始作答结果中的部分或全部数字，获得目标作答结果，从而提高了目标作答结果的准确性。In the above solution, the numbers in the initial answer result are adjusted based on the number adjustment process, some or all of the numbers in the initial answer result are deleted, and the target answer result is obtained, thereby improving the accuracy of the target answer result.

在一种实施方式中，基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果，包括：若当前作业的题型是所述选择题，对初始作答结果进行数字识别；若未识别到数字，则将初始作答结果与预设的多个标准结果进行匹配，基于文本识别模型获取各所述标准结果的识别置信度；将至少两个所述标准结果中识别置信度最高的标准结果确定为目标作答结果。In one embodiment, the initial answer result is adjusted based on the question type of the current homework to obtain the target answer result, including: if the question type of the current homework is the multiple-choice question, digitally identify the initial answer result; If the number is recognized, the initial answer result is matched with a plurality of preset standard results, and the recognition confidence of each standard result is obtained based on the text recognition model; the standard with the highest recognition confidence among at least two of the standard results is The result is determined as the target response result.

其中，标准结果可以为选择题所对应的选项，该标准结果可以根据当前作业的具体选项进行设置，例如若当前作业的选项为A、B、C、D，则标准结果可以包括：A、B、C、D中的至少两个。标准结果的识别置信度表征该标准结果的真实值落在测量结果的周围的概率，该置信度越高则初始作答结果为该标准结果的可能性越大。Wherein, the standard result can be the option corresponding to the multiple-choice question, and the standard result can be set according to the specific options of the current job. For example, if the options of the current job are A, B, C, D, the standard result can include: A, B , C, D at least two. The recognition confidence of the standard result represents the probability that the true value of the standard result falls around the measurement result, and the higher the confidence is, the more likely the initial answer result is the standard result.

在本实施例中，在当前作业的题型是选择题时，调整处理可以为数量调整处理，关于该数量调整处理具体说明如下：In this embodiment, when the question type of the current assignment is a multiple-choice question, the adjustment processing may be quantity adjustment processing, and the specific description of the quantity adjustment processing is as follows:

在本实施例中，可以根据选择题的选项预先设置多个标准结果。对初始作答结果进行数字识别，若未识别到数字，则可以根据初始作答结果对应的识别置信度确定目标作答结果。具体地，作业批改装置可以将初始作答结果中的字符与多个标准结果进行匹配，确定匹配成功的标准结果，匹配成功的标准结果的数量可以为多个，并通过文本识别模型确定各匹配成功的标准作答结果对应的识别置信度，并将该匹配成功的标准结果中置信度最高的标准识别结果确定为目标作答结果。示例性的，学生在作答的过程中，可能先作答C，然后将C划掉，修改作答为A，因而在初始作答结果中可能包括A和C，与该初始作答结果匹配成功的标准作答结果包括A和C，进而根据文本识别模型确定A对应的识别置信度，以及C对应的识别置信度。In this embodiment, multiple standard results can be preset according to the options of the multiple-choice question. Perform number recognition on the initial answer result. If no number is recognized, the target answer result can be determined according to the recognition confidence corresponding to the initial answer result. Specifically, the homework correction device can match the characters in the initial answer result with multiple standard results to determine the standard results that match successfully. The recognition confidence corresponding to the standard answer result of the standard answer result, and the standard recognition result with the highest confidence among the standard results of successful matching is determined as the target answer result. Exemplarily, in the process of answering, the student may first answer C, then cross out C, and modify the answer to A, so the initial answer result may include A and C, and the initial answer result matches the successful standard answer result A and C are included, and then the recognition confidence corresponding to A and the recognition confidence corresponding to C are determined according to the text recognition model.

可以理解地，在一些应用场景中，用户在作答区域内可能使用修改标记对作答内容进行修改，由于修改标记不标准等原因，在进行文本识别时，可能初始作答结果中包括多个标准作答结果，本实施例中在匹配成功的多个标准结果中确定了置信度最高的标准结果为目标作答结果，从而提高了目标作答结果的准确性。Understandably, in some application scenarios, the user may use the modification mark to modify the answer content in the answer area. Due to reasons such as non-standard modification marks, when performing text recognition, the initial answer result may include multiple standard answer results , in this embodiment, the standard result with the highest confidence is determined as the target answer result among multiple standard results that match successfully, thereby improving the accuracy of the target answer result.

上述方案中，使用数量调整处理初始作答结果，排除了置信度较低的标准结果，从而提高了目标作答结果的准确性。In the above scheme, the initial answer results are processed by quantity adjustment, and the standard results with low confidence are excluded, thereby improving the accuracy of the target answer results.

在本实施例中，在当前作业的题型是填空题或主观题时，调整处理可以为区域调整处理，关于该区域调整处理具体说明如下：In this embodiment, when the question type of the current assignment is a fill-in-the-blank question or a subjective question, the adjustment processing can be area adjustment processing, and the specific description of the area adjustment processing is as follows:

在一种实施方式中，基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果，包括：若当前作业的题型为填空题或主观题，对初始作答结果进行文本识别；若未识别到文本，对当前作业的完整图像进行文本识别，得到文本识别结果；当文本识别结果中包括除当前作业的题目文本之外的目标文本，则将该目标文本确定为目标作答结果。In one embodiment, the initial answer result is adjusted based on the question type of the current homework to obtain the target answer result, including: if the question type of the current homework is a fill-in-the-blank question or a subjective question, text recognition is performed on the initial answer result; if If no text is recognized, perform text recognition on the complete image of the current assignment to obtain a text recognition result; when the text recognition result includes target text other than the title text of the current assignment, then determine the target text as the target answer result.

其中，当前作业的完整图像可以为包括所有当前作业题目以及当前作业的作答内容的图像。在一些实施例中，该完整图像可以包括当前作业的题目、当前作业的作答区域、该题目与作答区域之间的空白区域、该题目和作答区域两侧的空白区域。文本识别结果为对当前作业的完整图像进行文本识别获得的识别结果。可以理解地，由于当前作业的完整图像中包括当前作业的题目，因而文本识别结果中可以包括题目文本，该题目文本可以为对当前作业的题目进行文本识别确定的文本。Wherein, the complete image of the current assignment may be an image including all questions of the current assignment and answer contents of the current assignment. In some embodiments, the complete image may include the topic of the current homework, the answer area of the current homework, the blank area between the question and the answer area, and the blank areas on both sides of the question and the answer area. The text recognition result is the recognition result obtained by performing text recognition on the complete image of the current job. Understandably, since the complete image of the current job includes the title of the current job, the text recognition result may include title text, and the title text may be the text determined by text recognition on the title of the current job.

在本实施例中，用户可能在作答区域之外的区域对当前作业进行了作答，因而在初始作答结果中未识别到文本，说明基于当前作答区域无法获得用户作答的内容。因而可以将进行图像采集的图像采集区域进行扩大，扩大为完整图像，并采用文本识别模型对该完整图像进行文本识别，获得对应的文本识别结果，可以理解地，由于完整图像中包括了当前作业的题目以及用户针对该当前作业的作答内容，因而可以将对完整图像识别确定的文本识别结果中的当前作业的题目文本删除，确定作答内容对应的目标文本，将该目标文本作为目标作答结果。如图4所示，区域401为作答区域，区域402为当前作业的完整图像对应的区域，用户将用户作答的内容“A”在作答区域之外，当前作业包括题目1和题目2，则扩大图像采集区域如区域402所示，从而能够识别用户在作答区域之外书写的内容。In this embodiment, the user may have answered the current homework in an area other than the answer area, so no text is recognized in the initial answer result, indicating that the content of the user's answer cannot be obtained based on the current answer area. Therefore, the image acquisition area for image acquisition can be expanded to a complete image, and the text recognition model is used to perform text recognition on the complete image to obtain the corresponding text recognition result. Understandably, since the complete image includes the current job The topic and the user's answer content for the current job, so the topic text of the current job in the text recognition results determined for the complete image recognition can be deleted, the target text corresponding to the answer content can be determined, and the target text can be used as the target answer result. As shown in Figure 4,area 401 is the answer area, andarea 402 is the area corresponding to the complete image of the current assignment. The user puts the content "A" answered by the user outside the answer area. The current assignment includes topic 1 andtopic 2, then expand The image capture area is shown asarea 402, so that the content written by the user outside the answer area can be recognized.

上述方案中，使用区域调整处理方法，当用户在作答区域之外书写作答内容时，通过扩大文本识别的区域能够在完整图像中包括用户的作答内容，从而对当前作业的完整图像进行图像识别，从而获得目标作答结果，提高了获得目标作答结果的灵活性。In the above solution, using the area adjustment processing method, when the user writes the answer content outside the answer area, the user's answer content can be included in the complete image by expanding the text recognition area, so as to perform image recognition on the complete image of the current job, Therefore, the target answer result is obtained, and the flexibility of obtaining the target answer result is improved.

在一种实施方式中，还可以将图像采集区域逐渐进行扩大，具体地，基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果，可以包括：当初始作答结果为空，则将图像采集区域从作答区域开始扩大，直到获取到包括内容的目标作答结果。In one embodiment, the image collection area can also be gradually expanded. Specifically, the initial answer result is adjusted based on the question type of the current assignment to obtain the target answer result, which may include: when the initial answer result is empty, then Expand the image collection area from the answering area until the target answering result including the content is obtained.

在本实施例中，在当前作业的题型是填空题或主观题时，调整处理可以为单位调整处理，关于该单位调整处理具体说明如下：In this embodiment, when the question type of the current assignment is a fill-in-the-blank question or a subjective question, the adjustment processing can be unit adjustment processing, and the specific description of the unit adjustment processing is as follows:

在一种实施方式中，基于当前作业的题型对初始作答结果进行调整处理，得到目标作答结果，包括：若当前作业的题型为填空题或主观题，对初始作答结果进行计量单位识别；若识别到计量单位，提取初始作答结果中的计量单位；对计量单位进行中英文格式的转换，得到包括中英文格式的计量单位的目标作答结果。In one embodiment, the initial answer result is adjusted based on the question type of the current homework to obtain the target answer result, including: if the question type of the current homework is a fill-in-the-blank question or a subjective question, performing unit of measurement identification on the initial answer result; If the unit of measurement is identified, the unit of measurement in the initial answer result is extracted; the unit of measurement is converted between Chinese and English formats to obtain a target answer result including the unit of measurement in both Chinese and English formats.

其中，计量单位为数学或物理等维度的计量值，本实施例对该计量单位不做限制，例如该计量单位可以包括初中数学计量单位、初中物理计量单位、高中数学计量单位、高中物理计量单位中的至少一个。本实施例对该计量单位中包括的语言种类不作限制，举例而言，该计量单位可以为中文语言种类，例如，5厘米；该计量单位可以为英文语言种类，例如，5cm；该计量单位还可以为中英文混合语言种类，例如，5厘米和5cm。Wherein, the unit of measurement is a measurement value in dimensions such as mathematics or physics. This embodiment does not limit the unit of measurement. For example, the unit of measurement may include a junior high school mathematical unit of measurement, a junior high school physical unit of measurement, a high school mathematical unit of measurement, and a high school physical unit of measurement at least one of the This embodiment does not limit the language types included in the measurement unit. For example, the measurement unit can be the Chinese language type, for example, 5 cm; the measurement unit can be the English language type, for example, 5 cm; the measurement unit can also be It can be Chinese and English mixed language types, for example, 5cm and 5cm.

在一种实施方式中，可以预设计量单位，并将预设的计量单位与初始作答结果进行匹配，从而实现初始作答结果的计量单位识别。若在初始作答结果中识别到计量单位，则将初始作答结果中的计量单位进行中英文相互转化，本实施例对该转化过程不作限制，例如，若初始作答结果为5CM，进行中英文转化获得的目标作答结果中包括：5厘米、5cm等。In one embodiment, the unit of measurement can be preset, and the preset unit of measurement can be matched with the initial answer result, so as to realize the identification of the unit of measurement of the initial answer result. If the unit of measurement is identified in the initial answer result, the unit of measurement in the initial answer result is converted between Chinese and English. This embodiment does not limit the conversion process. For example, if the initial answer result is 5CM, perform Chinese-English conversion to obtain The target answer results include: 5 cm, 5 cm, etc.

上述方案中，对初始作答结果中的计量单位进行中英文转化，避免了因为中英文不统一造成的后续批改结果错位。In the above scheme, the units of measurement in the initial answer results are converted into Chinese and English, which avoids the dislocation of subsequent correction results caused by the inconsistency between Chinese and English.

通过上述结果调整处理能够较为明显的提升作答结果的准确性，根据实际测试，初始作答结果的准确性为92.5%，对初始作答结果使用进行调整处理获得的目标作答结果的准确性为97.9%。The above result adjustment process can significantly improve the accuracy of the answer results. According to the actual test, the accuracy of the initial answer results is 92.5%, and the accuracy of the target answer results obtained by adjusting the initial answer results is 97.9%.

在一些实施例中，获取当前作业在作答区域内的作答结果图像之前，所述方法还包括：获取且展示标准作答规范信息，其中标准作答规范信息用于指导用户作答当前作业具体操作的文字表述。该标准作答规范信息可以包括一个或多个维度的规则，本实施例不做限制，例如该标准作答规范可以包括作答区域规则、修改规则、涂改规则、书写规则、作答类型规则中的一个或多个。该标准作答规范信息的形式本实施例也不做限制，例如可以为条例形式和/或图例形式。该标准作答规范信息的展示方式有多种，包括但不限于，通过弹窗展示给用户和/或通过预先印制展示给用户。In some embodiments, before acquiring the answer result image of the current assignment in the answer area, the method further includes: acquiring and displaying standard answer specification information, wherein the standard answer specification information is used to guide the user to answer the textual expression of the specific operation of the current assignment . The standard answer specification information may include rules of one or more dimensions, which is not limited in this embodiment. For example, the standard answer specification may include one or more of answer area rules, modification rules, obliteration rules, writing rules, and answer type rules. indivual. The form of the standard answer specification information is not limited in this embodiment, for example, it may be in the form of regulations and/or diagrams. There are many ways to display the standard answer specification information, including but not limited to, displaying to users through pop-up windows and/or displaying to users through pre-printing.

在本实施例中，获取标准作答规范信息并将该标准作答规范信息展示给用户，用户获取该标准作答规范信息，用户按照该标准作答规范信息进行作答。具体地，一种可选的实施方式中，若标准作答规范信息以条例形式展示，则该标准作答规范信息可以包括但不限于以下条目中的至少一条：（1）做答内容需完全书写在作答区域内，不可超出作答区域，不要和做答区域的边框重叠。（2）如对作答内容进行修改，需使用做答笔划掉需要修改的作答内容（即尽量覆盖需要修改的作答内容，划线至少在三条以上），将更新的作答内容写在划掉的作答内容之后，注意：更新的作答内容也需要在做答区内，不可超出作答区域。（3）若多次进行修改，作答区域内没有足够空余空间填写更新的作答内容，请使用涂改工具（例如，涂改液、涂改带）覆盖需要修改的作答内容，将更新的作答内容写在做答区域内。（4）书写字母（例如，A、B、C、D）时尽量规范、标准，保持书写工整，不可有连笔。（5）若当前题目为单项选择题，做答区域中仅可书写选项字母中的一个，若书写数字、公式等，则当前作业的批改结果为错误。（6）请保持试卷干净整洁，做答区域内无污点、脏污。In this embodiment, the standard answer specification information is obtained and displayed to the user, the user obtains the standard answer specification information, and the user answers according to the standard answer specification information. Specifically, in an optional implementation manner, if the standard answer specification information is displayed in the form of regulations, the standard answer specification information may include but not limited to at least one of the following items: (1) The answer content must be completely written in the In the answer area, do not exceed the answer area, and do not overlap with the border of the answer area. (2) If you want to modify the answer content, you need to use the answer pen to cross out the answer content that needs to be modified (that is, cover the answer content that needs to be modified as much as possible, with at least three underlines), and write the updated answer content on the crossed out answer content After the content, note: the updated answer content also needs to be in the answer area and cannot exceed the answer area. (3) If you have made multiple revisions and there is not enough space in the answer area to fill in the updated answer content, please use correction tools (such as correction fluid, correction tape) to cover the answer content that needs to be modified, and write the updated answer content in the do answer area. (4) When writing letters (for example, A, B, C, D), try to be as standardized and standard as possible, keep the writing neat, and do not have consecutive strokes. (5) If the current question is a multiple-choice question, only one of the option letters can be written in the answer area. If numbers, formulas, etc. are written, the correction result of the current assignment will be an error. (6) Please keep the test paper clean and tidy, and there is no stain or dirt in the answering area.

一种可选的实施方式中，若标准作答规范信息以图例形式展示，则该标准作答规范信息可以包括但不限于图5所示的图例，图5中，区域501展示的为标准作答规范信息中的规范字母书写样式，区域502展示的为标准作答规范信息中的规范修改样式。In an optional embodiment, if the standard answer specification information is displayed in the form of a legend, the standard answer specification information may include but not limited to the legend shown in Figure 5, in Figure 5, thearea 501 shows the standard answer specification information The standard letter writing style in , and thearea 502 shows the standard modification style in the standard answer specification information.

上述方案中，将标准作答规范展示给用户，从而使得用户能够获得该标准作答规范，从而根据作答规范生成当前作业，提高了用户作答的标准程度，基于该作答内容生成目标作答结果，能够从源头上提升目标作答结果的准确性。In the above solution, the standard answer specification is displayed to the user, so that the user can obtain the standard answer specification, thereby generating the current homework according to the answer specification, which improves the standard level of the user's answer, and generates the target answer result based on the answer content, which can be obtained from the source. To improve the accuracy of the target answer results.

接下来通过具体的示例对本公开实施例中的作业批改方法，进行进一步说明。Next, the homework correction method in the embodiments of the present disclosure will be further described through specific examples.

示例性的，图6为本公开实施例提供的另一种作业批改方法的流程示意图，如图6所示，该示例中当前作业的题型为客观题。该方法包括：Exemplarily, FIG. 6 is a schematic flowchart of another homework correction method provided by an embodiment of the present disclosure. As shown in FIG. 6 , the question type of the current homework in this example is an objective question. The method includes:

步骤601，开始。Step 601, start.

步骤602，获取学生作答结果。Step 602, obtaining the student's answer result.

步骤603，根据作答区域对作答结果进行切图，获得作答结果图像。其中，作答区域为矩形框。Step 603: Cut the answer result image according to the answer area to obtain the answer result image. Among them, the answer area is a rectangular frame.

步骤604，对作答结果图像进行文本识别，获得初始作答结果。Step 604, performing text recognition on the answer result image to obtain an initial answer result.

步骤605，判断该初始作答结果是否属于客观题选项，若是，执行步骤607；否则，执行步骤606。Step 605, judging whether the initial answer result belongs to the objective question option, if so, go to step 607; otherwise, go to step 606.

步骤606，对初始作答结果进行调整处理，得到目标作答结果。Step 606, adjust the initial answer result to obtain the target answer result.

步骤607，若存在目标作答结果，对照对目标作答结果和标准作答结果；若不存在目标作答结果，对照初始作答结果和标准作答结果，获得对照结果。Step 607: If there is a target answer result, compare the target answer result with the standard answer result; if there is no target answer result, compare the initial answer result with the standard answer result to obtain a comparison result.

步骤608，根据对照结果生成自动批改结果，并显示该自动批改结果。Step 608, generating an automatic correction result according to the comparison result, and displaying the automatic correction result.

步骤609，结束。Step 609, end.

示例性的，图7为本公开实施例提供的又一种作业批改方法的流程示意图，如图7所示，该示例中当前作业的题型为理科主观题。该方法包括：Exemplarily, FIG. 7 is a schematic flowchart of another homework correction method provided by an embodiment of the present disclosure. As shown in FIG. 7 , the question type of the current homework in this example is a subjective question in science. The method includes:

步骤701，开始。Step 701, start.

步骤702，获取学生作答结果。Step 702, obtaining the student's answer result.

步骤703，根据作答区域对作答结果进行切图，获得作答结果图像。其中，作答区域为虚线框。Step 703: Cut the answer result image according to the answer area to obtain the answer result image. Among them, the answer area is a dotted box.

步骤704，根据检测模型确定作答结果图像中最后一行，对该最后一行进行文本识别，获得初始作答结果。Step 704: Determine the last line in the answer result image according to the detection model, perform text recognition on the last line, and obtain an initial answer result.

步骤705，判断该初始作答结果是否属于客观题选项。若是，执行步骤707；否则，执行步骤706。Step 705, judging whether the initial answer result belongs to the objective question option. If yes, go to step 707; otherwise, go to step 706.

步骤706，对初始作答结果进行调整处理，得到目标作答结果。Step 706, adjust the initial answer result to obtain the target answer result.

步骤707，若存在目标作答结果，对照目标作答结果和标准作答结果；若不存在目标作答结果，对照初始作答结果和标准作答结果，获得对照结果。Step 707: If there is a target answer result, compare the target answer result with the standard answer result; if there is no target answer result, compare the initial answer result with the standard answer result to obtain a comparison result.

步骤708，根据对照结果生成自动批改结果，并显示该自动批改结果。Step 708, generating an automatic correction result according to the comparison result, and displaying the automatic correction result.

步骤709，结束。Step 709, end.

示例性的，图8为本公开实施例提供的再一种作业批改方法的流程示意图，如图8所示，该示例中做欠作业的题型为文科主观题。该方法包括：Exemplarily, FIG. 8 is a schematic flowchart of another homework correction method provided by an embodiment of the present disclosure. As shown in FIG. 8 , in this example, the question type of the homework owed is a subjective question in liberal arts. The method includes:

步骤801，开始。Step 801, start.

步骤802，获取学生作答结果。Step 802, obtaining the student's answer result.

步骤803，根据作答区域对作答结果进行切图，获得作答结果图像。其中，作答区域为虚线框。Step 803: Cut the answer result image according to the answer area to obtain an answer result image. Among them, the answer area is a dotted box.

步骤804，对目标作答结果进行基础校验，获得基础校验结果。Step 804, perform basic verification on the target answer result, and obtain the basic verification result.

步骤805，对目标作答结果进行进阶校验，获得进阶校验结果。Step 805, perform advanced verification on the target answer result, and obtain the advanced verification result.

步骤806，使用校验标识突出显示该校验结果。Step 806, using the verification mark to highlight the verification result.

步骤807，汇总展示该校验结果。Step 807, summarizing and displaying the verification results.

步骤808，获取用户基于该校验结果输入的分数。Step 808, acquiring the score input by the user based on the verification result.

步骤809，结束。Step 809, end.

上述方案中，在生成作业时可以根据学生特点生成个性化作业，获取该个性化作业中的题型以及题目位置，基于题目位置对应的作答区域生成初始作答结果，根据题型确定的对初始结果进行调整处理，从而实现对个性化作业的自动化批改。并且具体说明了对客观题、理科主观题以及文科主观题的批改方法，从而能够根据题型提供针对性的批改方法，从而提高了批改结果的准确性。In the above scheme, when generating homework, personalized homework can be generated according to the characteristics of the students, the question type and question position in the personalized homework can be obtained, and the initial answer result can be generated based on the answer area corresponding to the question position, and the initial answer can be determined according to the question type. Perform adjustment processing to realize automatic correction of personalized assignments. It also specifically explains the correction methods for objective questions, science subjective questions and liberal arts subjective questions, so that targeted correction methods can be provided according to the question types, thereby improving the accuracy of the correction results.

可选的，基于上述实施例，作业批改装置可以在学生作答的过程中生成目标作答结果，根据试卷中坐标位置与web端页面中坐标位置的对应关系，将该目标作答结果存储到数据库中，并通过试卷中的二维码获取并存储对应的学生ID、试卷ID、页码ID。进而能够从该数据库中读取目标作答结果，将该目标作答结果提供给教师，以使教师对该目标作答结果进行阅卷。Optionally, based on the above-mentioned embodiment, the homework correcting device can generate a target answer result during the student's answering process, and store the target answer result in the database according to the correspondence between the coordinate position in the test paper and the coordinate position in the web page, And obtain and store the corresponding student ID, test paper ID, and page number ID through the QR code in the test paper. Furthermore, the target answer result can be read from the database, and the target answer result can be provided to the teacher, so that the teacher can mark the target answer result.

在一些应用场景中，还可以将作答区域设置为教师对题目的批改区域，学生可以使用圆珠笔、签字笔等普通笔作答，教师使用点阵笔在批改区域进行阅卷，如图9所示，图9为本公开实施例提供的一种阅卷方法的示意图，图9展示了提取并识别轨迹，生成初始阅卷结果并优化初始阅卷结果为目标阅卷结果，进而生成个性化作业和/或学情的过程。具体地，首先，获取阅卷区域内的阅卷结果图像，具体地，若阅卷工具为点阵笔，提取点阵笔在阅卷区域内的轨迹坐标，并根据轨迹坐标生成轨迹。进一步地，根据轨迹生成作答结果图形，将该作答结果图像输入到识别模型（例如，OCR模型）接口中，识别模型对作答结果图像进行文本识别，获得初始阅卷结果，并将该初始阅卷结果落库存储。进一步地，基于当前作业的题型对初始阅卷结果进行优化，得到目标阅卷结果。最后，基于当前作业的题型和作答者信息，确定目标阅卷结果对应的分析结果，该分析结果包括分数、个性化作业学情分析中的至少一个。其中个性化作业能够基于学生的错题推荐针对性的习题；学情分析中包括个人学情分析和/班级学情分析，其中个人学情分析包括个人的做题状况或学习曲线等，班级学情分析包括班级的答题完成率、学习曲线等，从而能够辅助教师因材施教。通过阅卷调整处理能够较为明显的提升阅卷结果的准确性，In some application scenarios, the answering area can also be set as the teacher’s marking area for the questions. Students can use ordinary pens such as ballpoint pens and signature pens to answer, and the teacher uses dot matrix pens to mark the papers in the marking area, as shown in Figure 9. 9 is a schematic diagram of a marking method provided by an embodiment of the present disclosure. FIG. 9 shows the process of extracting and identifying trajectories, generating initial marking results and optimizing the initial marking results to target marking results, and then generating personalized assignments and/or academic conditions . Specifically, firstly, obtain the marking result image in the marking area, specifically, if the marking tool is a dot-matrix pen, extract the trajectory coordinates of the dot-matrix pen in the marking area, and generate a trajectory according to the trajectory coordinates. Further, generate an answer result graphic according to the trajectory, input the answer result image into the interface of the recognition model (for example, OCR model), the recognition model performs text recognition on the answer result image, obtains the initial marking result, and puts the initial marking result into the library storage. Further, based on the question type of the current assignment, the initial marking result is optimized to obtain the target marking result. Finally, based on the question type of the current assignment and the information of the respondent, the analysis result corresponding to the target marking result is determined, and the analysis result includes at least one of scores and personalized assignment analysis. Among them, personalized homework can recommend targeted exercises based on students' wrong questions; learning situation analysis includes personal learning situation analysis and/class learning situation analysis, and personal learning situation analysis includes personal problem-solving status or learning curve, etc. Situational analysis includes the class's answer completion rate, learning curve, etc., so as to assist teachers to teach in accordance with their aptitude. The accuracy of marking results can be significantly improved through marking adjustment processing.

根据实际测试，初始阅卷结果的准确性为91.2%，对初始阅卷结果进行调整处理获得的目标阅卷结果的准确性为93%。According to the actual test, the accuracy of the initial marking results is 91.2%, and the accuracy of the target marking results obtained by adjusting the initial marking results is 93%.

一种可选的实施方式中，若目标作答结果包括数学公式或书写步骤等，作业批改装置可以将目标作答结果和标准作答结果呈现给教师，教师点击对、错、半对选项生成目标阅卷结果。In an optional implementation, if the target answer results include mathematical formulas or writing steps, etc., the homework correction device can present the target answer results and standard answer results to the teacher, and the teacher clicks on the right, wrong, and half-right options to generate the target marking results .

上述方案中，通过教师的批改轨迹能够确定学生的作答结果是否正确，从而避免了收集学生的作答内容，通过自动化的方法获取分析结果，提高了分析结果的生成效率。In the above solution, it is possible to determine whether the student's answer is correct through the teacher's correction track, thereby avoiding the collection of the student's answer, and obtaining the analysis result through an automated method, which improves the generation efficiency of the analysis result.

一种可选的实施方式中，还可以展示标准阅卷规范信息给教师，以获取按照标准阅卷规范信息的阅卷结果图像。本实施例对该标准阅卷规则的形式不做限制，例如可以为条例形式和/或图例形式。若标准作答规范以条例形式展示，则该标准作答规范信息可以包括但不限于以下条目中的至少一条：（1）请使用：对、错、半对三种符号进行判题，其余符号不支持识别；（2）对，错，半对均需书写在判题区内，不可超出阅卷区域；（3）错符号不要有连笔，半对符号中对钩上的点划需要书写清晰，和对符号清晰区分；（4）使用点阵笔阅卷时力度尽量均匀，阅卷速度不宜过快；（5）对符号要有弯曲度，不要和横线相似；（6）请保持试卷干净整洁，阅卷区域内无污点，遮挡。In an optional implementation manner, the standard marking specification information can also be displayed to the teacher, so as to obtain the marking result image according to the standard marking specification information. This embodiment does not limit the form of the standard marking rule, for example, it may be in the form of regulations and/or legends. If the standard answer specification is displayed in the form of regulations, the standard answer specification information may include but not limited to at least one of the following items: (1) Please use the three symbols: true, false, and half-pair to judge the question, and other symbols do not support Recognition; (2) Right, wrong, and half-pairs must be written in the judgment area, and must not exceed the marking area; (3) Wrong symbols should not have consecutive strokes, and the dots and dashes on the hooks in the half-pair symbols need to be written clearly, and Clearly distinguish the symbols; (4) Try to use a dot matrix pen to mark the test paper as evenly as possible, and the speed of the test should not be too fast; (5) The symbols should be curved and not similar to horizontal lines; (6) Please keep the test paper clean and tidy, and mark the test paper No smudges, occlusions in the area.

一种可选的实施方式中，若标准阅卷规范以图例形式展示，则该标准作答规范可以包括但不限于图10所示的图例，图10中，区域1001展示的为标准阅卷规范，区域1002展示的为不标准阅卷规范。图10中展示了规范的对符号、错符号、半对符号以及不规范的对符号、错符号、半对符号。In an optional embodiment, if the standard marking specification is displayed in the form of a legend, the standard answer specification may include but not limited to the legend shown in Figure 10. In Figure 10, thearea 1001 shows the standard marking specification, and thearea 1002 What is shown is the non-standard marking specification. Figure 10 shows the correct symbols, wrong symbols, and half pairs of symbols and the irregular symbols, wrong symbols, and half pairs of symbols.

本公开实施例提供的作业批改方案，具有丰富的图像采集手段，通过标准作答规范提高了当前作业的标准性，并且提示用户在作答区域内作答，从而提升了采集获取的作答结果图像的全面性。通过点阵笔能够实现用户书写及时获取轨迹，在收集数据的同时不会影响用户的作答体验，同时也降低了用户上传图像数据的负担。The homework correction scheme provided by the embodiments of the present disclosure has rich image collection means, improves the standardization of the current homework through the standard answer specification, and prompts the user to answer in the answer area, thereby improving the comprehensiveness of the image of the answer result collected and acquired . Through the dot-matrix pen, the user can obtain the track in time when writing, which will not affect the user's answering experience while collecting data, and also reduce the burden on the user to upload image data.

接下来通过一个具体的示例对本公开实施例中的作业批改方法，进行进一步说明，该示例中学生、教师、家长三方可以参与该作业批改过程。示例性的，图11为本公开实施例提供的一种作业批改方法的示意图，如图11所示。步骤1101，学生登录学习系统。步骤1102，学生打印试卷。打印试卷之后可以采用点阵笔或者普通笔作答，若采用点阵笔作答执行步骤1103；若采用普通笔作答执行步骤1104。步骤1103，学生使用点阵笔作答。步骤1104，学生采用普通笔作答。步骤1105，学生使用扫描仪扫描作答结果。步骤1106，云端系统自动化批改客观题，获得客观题批改结果。步骤1107，教师审核客观题批改结果。步骤1108，教师批改主观题。步骤1109，云端系统整合批改结果。步骤1110，云端系统生成学情报告。步骤1111，云端系统生成错题本。步骤1112，教师根据批阅结果在试卷上进行留痕打印。步骤1113，学生获得批改结果。需要说明的是，上述过程中，家长可以实时查看点阵笔作答结果、美学机扫描结果、学情报告、错题本等。Next, a specific example is used to further illustrate the homework correction method in the embodiment of the present disclosure. In this example, students, teachers, and parents can participate in the homework correction process. Exemplarily, FIG. 11 is a schematic diagram of an assignment correction method provided by an embodiment of the present disclosure, as shown in FIG. 11 .Step 1101, students log into the learning system.Step 1102, students print test papers. After printing the test paper, dot-matrix pens or ordinary pens can be used to answer. If dot-matrix pens are used to answer, go tostep 1103; if ordinary pens are used to answer, go tostep 1104.Step 1103, students use dot-matrix pens to answer.Step 1104, students answer with ordinary pens.Step 1105, the students use the scanner to scan the answer results.Step 1106, the cloud system automatically corrects the objective questions, and obtains the correction results of the objective questions.Step 1107, the teacher reviews the correction results of the objective questions.Step 1108, the teacher corrects the subjective questions.Step 1109, the cloud system integrates the correction results.Step 1110, the cloud system generates a student report.Step 1111, the cloud system generates an error book. Instep 1112, the teacher prints marks on the test paper according to the marking results.Step 1113, the student obtains the correction result. It should be noted that during the above process, parents can view the answer results of the dot matrix pen, the scanning results of the aesthetic machine, the learning report, the wrong question book, etc. in real time.

上述方案中，实现了客观题的自动化批改，从而降低了老师的工作强度，并且学生的答题和老师的批改可以并行进行，家长也可以实时查看学生的作答情况。In the above scheme, the automatic correction of objective questions is realized, thereby reducing the workload of the teacher, and the students' answers and the teacher's corrections can be carried out in parallel, and parents can also check the students' answers in real time.

对应于前述作业批改方法，本公开实施例还提供了一种作业批改装置，图12为本公开实施例提供的一种作业批改装置的结构示意图，该装置可由软件和/或硬件实现，一般可集成在电子设备中。如图12所示，作业批改装置1200包括：Corresponding to the aforementioned homework correction method, an embodiment of the present disclosure also provides a homework correction device. FIG. 12 is a schematic structural diagram of a homework correction device provided by an embodiment of the present disclosure. The device can be implemented by software and/or hardware. integrated in electronic equipment. As shown in Figure 12, thehomework correcting device 1200 includes:

图像获取模块1201，用于获取当前作业的作答结果图像；Animage acquisition module 1201, configured to acquire an answer result image of the current job;

文本识别模块1202，用于对所述作答结果图像进行文本识别，得到初始作答结果；Atext recognition module 1202, configured to perform text recognition on the answer result image to obtain an initial answer result;

结果优化模块1203，用于获取所述当前作业的题型，并基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果；Theresult optimization module 1203 is configured to obtain the question type of the current assignment, and adjust the initial answer result based on the question type of the current assignment to obtain a target answer result;

结果判断模块1204，用于基于所述当前作业的题型和所述当前作业的标准作答结果，对所述目标作答结果进行判断，确定所述当前作业的批改结果。Theresult judging module 1204 is configured to judge the target answer result based on the question type of the current homework and the standard answer result of the current homework, and determine the correcting result of the current homework.

在一些实施方式中，图像获取模块1201，用于：In some implementations, theimage acquisition module 1201 is configured to:

获取所述当前作业图像，并根据所述当前作业图像中预先设定作答区域的区域坐标提取得到所述作答结果图像。The current job image is acquired, and the answer result image is extracted according to area coordinates of a preset answer area in the current job image.

在一些实施方式中，图像获取模块1201，用于：In some implementations, theimage acquisition module 1201 is configured to:

获取作答工具在作答区域内的轨迹坐标，并根据所述轨迹坐标生成所述作答结果图像。The track coordinates of the answering tool in the answering area are acquired, and the answer result image is generated according to the track coordinates.

在一些实施方式中，文本识别模块1202，用于：In some implementations, thetext recognition module 1202 is configured to:

采用预先构建的文本识别模型对所述作答结果图像进行识别，得到所述初始作答结果，其中，所述文本识别模型中包括检测模型和识别模型，所述检测模型用于提取所述作答结果图像中的作答位置并输入至所述识别模型，所述识别模型用于对所述作答位置的图像进行文本识别得到所述初始作答结果。Using a pre-built text recognition model to recognize the answer result image to obtain the initial answer result, wherein the text recognition model includes a detection model and a recognition model, and the detection model is used to extract the answer result image The answer position in and input to the recognition model, and the recognition model is used to perform text recognition on the image of the answer position to obtain the initial answer result.

在一些实施方式中，所述当前作业的题型包括选择题、填空题或主观题。In some embodiments, the question types of the current assignment include multiple choice questions, fill-in-the-blank questions or subjective questions.

在一些实施方式中，若所述当前作业的题型是所述选择题，所述调整处理包括字符调整处理、数字调整处理、数量调整处理中的至少一种；若所述当前作业的题型是所述填空题或所述主观题，则所述调整处理包括区域调整处理或单位调整处理。In some embodiments, if the question type of the current assignment is the multiple-choice question, the adjustment process includes at least one of character adjustment processing, number adjustment processing, and quantity adjustment processing; if the question type of the current assignment If it is the filling-in-the-blank question or the subjective question, the adjustment processing includes area adjustment processing or unit adjustment processing.

在一些实施方式中，结果优化模块1203，用于：In some embodiments, theresult optimization module 1203 is used for:

若所述当前作业的题型是所述选择题，将所述初始作答结果在绑定字符库中进行匹配，匹配则得到目标绑定字符，其中，所述绑定字符库中包括多个预设的绑定字符；If the question type of the current job is the multiple-choice question, the initial answer result is matched in the bound character library, and the target bound character is obtained after matching, wherein the bound character library includes a plurality of preset The binding character set;

基于预设的绑定字符与标准结果的关联关系，确定所述目标绑定字符对应的目标标准结果，将所述目标标准结果确定为所述目标作答结果，其中，所述绑定字符与标准结果的关联关系用于确定所述多个预设的绑定字符与标准结果之间的对应关系，一个所述标准结果关联至少一个所述绑定字符。Based on the preset association relationship between bound characters and standard results, determine the target standard results corresponding to the target bound characters, and determine the target standard results as the target answer results, wherein the bound characters and standard The association relationship of the results is used to determine the correspondence between the plurality of preset bound characters and standard results, and one standard result is associated with at least one bound character.

在一些实施方式中，结果优化模块1203，用于：In some embodiments, theresult optimization module 1203 is used for:

若所述当前作业的题型是所述选择题，对所述初始作答结果进行数字识别；If the question type of the current assignment is the multiple-choice question, digitally identify the initial answer result;

若识别到数字，删除所述初始作答结果中的数字，得到所述目标作答结果；If the number is identified, delete the number in the initial answer result to obtain the target answer result;

或者，当所述初始作答结果中的数字判定为预设数字且字符数量为2，则删除所述初始作答结果中的所述预设数字，得到所述目标作答结果。Alternatively, when the number in the initial answer result is determined to be a preset number and the number of characters is 2, the preset number in the initial answer result is deleted to obtain the target answer result.

在一些实施方式中，结果优化模块1203，用于：In some embodiments, theresult optimization module 1203 is used for:

若所述当前作业的题型是所述选择题，对所述初始作答结果进行数字识别；If the question type of the current assignment is the multiple-choice question, digitally identify the initial answer result;

若未识别到数字，则将所述初始作答结果与预设的多个标准结果进行匹配，基于文本识别模型获取各所述标准结果的识别置信度；If the number is not recognized, matching the initial answer result with a plurality of preset standard results, and obtaining the recognition confidence of each standard result based on the text recognition model;

将至少两个所述标准结果中识别置信度最高的标准结果确定为所述目标作答结果。Determining the standard result with the highest recognition confidence among the at least two standard results as the target answer result.

在一些实施方式中，结果优化模块1203，用于：In some embodiments, theresult optimization module 1203 is used for:

若所述当前作业的题型为所述填空题或所述主观题，对所述初始作答结果进行文本识别；If the question type of the current assignment is the filling-in-the-blank question or the subjective question, perform text recognition on the initial answer result;

若未识别到文本，对所述当前作业的完整图像进行文本识别，得到文本识别结果；If no text is recognized, perform text recognition on the complete image of the current job to obtain a text recognition result;

当所述文本识别结果中包括除所述当前作业的题目文本之外的目标文本，则将该目标文本确定为目标作答结果。When the text recognition result includes target text other than the topic text of the current assignment, the target text is determined as the target answer result.

在一些实施方式中，结果优化模块1203，用于：In some embodiments, theresult optimization module 1203 is used for:

若所述当前作业的题型为所述填空题或所述主观题，对所述初始作答结果进行计量单位识别；If the question type of the current assignment is the filling-in-the-blank question or the subjective question, performing unit of measurement identification on the initial answer result;

若识别到计量单位，提取所述初始作答结果中的计量单位；If the unit of measurement is identified, extract the unit of measurement in the initial answer result;

对所述计量单位进行中英文格式的转换，得到包括中英文格式的计量单位的目标作答结果。The Chinese-English format conversion is performed on the measurement unit, and the target answer result including the measurement unit in the Chinese-English format is obtained.

在一些实施方式中，若所述当前作业的题型为选择题，结果判断模块1204，用于：In some implementations, if the question type of the current assignment is a multiple-choice question, theresult judging module 1204 is configured to:

将所述目标作答结果与所述标准作答结果进行对比；comparing the target answer result with the standard answer result;

根据对比结果确定所述当前作业的批改结果。The correction result of the current job is determined according to the comparison result.

在一些实施方式中，若所述当前作业的题型为填空题，结果判断模块1204，用于：In some implementations, if the question type of the current assignment is a fill-in-the-blank question, theresult judging module 1204 is configured to:

将所述目标作答结果与所述标准作答结果进行数据结构的对比，或者字符数量的对比；Comparing the data structure of the target answer result with the standard answer result, or comparing the number of characters;

根据对比结果确定所述当前作业的批改结果；其中，所述数据结构用于表征作答结果包括的数据类型；字符数量用于表征作答结果包括的字符的数量。Determine the correction result of the current job according to the comparison result; wherein, the data structure is used to represent the data type included in the answer result; the number of characters is used to represent the number of characters included in the answer result.

在一些实施方式中，若所述当前作业的题型为主观题中的文科主观题，结果判断模块1204，用于：In some implementations, if the question type of the current assignment is a subjective question in liberal arts, theresult judging module 1204 is used to:

基于预设批改算法识别所述目标作答结果的至少一种批改参数，所述批改参数包括字数、标题、错别字、语法错误、搭配不当和结构错误中的至少一项；Identifying at least one correction parameter of the target answer result based on a preset correction algorithm, the correction parameter including at least one of word count, title, typo, grammatical error, improper collocation and structural error;

展示所述至少一种批改参数以及所述当前作业的标准作答结果，获取用户基于所述标准作答结果以及所述至少一种批改参数输入的所述当前作业的批改结果。The at least one correction parameter and the standard answer result of the current assignment are displayed, and the correction result of the current assignment entered by the user based on the standard answer result and the at least one correction parameter is acquired.

在一些实施方式中，若所述当前作业的题型为主观题中的理科主观题，结果判断模块1204，用于：In some implementations, if the question type of the current assignment is a subjective question in science in the subjective questions, theresult judging module 1204 is used to:

确定所述目标作答结果中位于最后一位的当前最终结果；determining the current final result at the last place among the target answer results;

将所述目标作答结果的当前最终结果与所述标准作答结果中的标准最终结果进行对比；comparing the current final result of the target answer result with the standard final result in the standard answer result;

当对比结果为所述当前最终结果与所述标准最终结果相同，则所述当前作业的批改结果确定为正确；否则所述当前作业的批改结果确定为错误。When the comparison result is that the current final result is the same as the standard final result, the correction result of the current job is determined to be correct; otherwise, the correction result of the current job is determined to be wrong.

在一些实施方式中，所述方法还包括：In some embodiments, the method also includes:

展示模块，用于在获取当前作业的作答结果图像之前，获取且展示标准作答规范信息，其中，所述标准作答规范信息用于指导用户作答当前作业具体操作的文字表述。The display module is used to obtain and display standard answer specification information before obtaining the answer result image of the current job, wherein the standard answer specification information is used to guide the user to answer the text expression of the specific operation of the current job.

本公开实施例所提供的作业批改装置可执行本公开任意实施例所提供的作业批改方法，具备执行方法相应的功能模块和有益效果。The job correction device provided by the embodiments of the present disclosure can execute the job correction method provided by any embodiment of the present disclosure, and has corresponding functional modules and beneficial effects for executing the method.

所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置实施例的具体工作过程，可以参考方法实施例中的对应过程，在此不再赘述。Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process of the device embodiment described above can refer to the corresponding process in the method embodiment, and details are not repeated here.

本公开实施方式中的多个装置之间所交互的消息或者信息的名称仅用于说明性的目的，而并不是用于对这些消息或信息的范围进行限制。The names of messages or information exchanged between multiple devices in the embodiments of the present disclosure are used for illustrative purposes only, and are not used to limit the scope of these messages or information.

本公开示例性实施例还提供一种电子设备，包括：至少一个处理器；以及与至少一个处理器通信连接的存储器。所述存储器存储有能够被所述至少一个处理器执行的计算机程序，所述计算机程序在被所述至少一个处理器执行时用于使所述电子设备执行根据本公开实施例的方法。Exemplary embodiments of the present disclosure also provide an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores a computer program executable by the at least one processor, and when executed by the at least one processor, the computer program is used to cause the electronic device to execute the method according to the embodiment of the present disclosure.

本公开示例性实施例还提供一种存储有计算机程序的非瞬时计算机可读存储介质，其中，所述计算机程序在被计算机的处理器执行时用于使所述计算机执行根据本公开实施例的方法。Exemplary embodiments of the present disclosure also provide a non-transitory computer-readable storage medium storing a computer program, wherein, when the computer program is executed by a processor of a computer, the computer is used to cause the computer to execute the method.

本公开示例性实施例还提供一种计算机程序产品，包括计算机程序，其中，所述计算机程序在被计算机的处理器执行时用于使所述计算机执行根据本公开实施例的方法。Exemplary embodiments of the present disclosure also provide a computer program product, including a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to execute the method according to the embodiments of the present disclosure.

所述计算机程序产品可以以一种或多种程序设计语言的任意组合来编写用于执行本公开实施例操作的程序代码，所述程序设计语言包括面向对象的程序设计语言，诸如Java、C++等，还包括常规的过程式程序设计语言，诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户计算设备上部分在远程计算设备上执行、或者完全在远程计算设备或服务器上执行。The computer program product can be written in any combination of one or more programming languages to execute the program codes for performing the operations of the embodiments of the present disclosure, and the programming languages include object-oriented programming languages, such as Java, C++, etc. , also includes conventional procedural programming languages, such as the "C" 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 to execute.

此外，本公开的实施例还可以是计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令在被处理器运行时使得所述处理器执行本公开实施例所提供的XYZ方法。所述计算机可读存储介质可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以包括但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。可读存储介质的更具体的例子（非穷举的列表）包括：具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器（RAM）、只读存储器（ROM）、可擦式可编程只读存储器（EPROM或闪存）、光纤、便携式紧凑盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。In addition, the embodiments of the present disclosure may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the XYZ method provided by the embodiments of the present disclosure. . The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, but not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or devices, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: electrical connection with one or more conductors, portable disk, hard disk, random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

参考图13，现将描述可以作为本公开的服务器或客户端的电子设备1300的结构框图，其是可以应用于本公开的各方面的硬件设备的示例。电子设备旨在表示各种形式的数字电子的计算机设备，诸如，膝上型计算机、台式计算机、工作台、个人数字助理、服务器、刀片式服务器、大型计算机、和其它适合的计算机。电子设备还可以表示各种形式的移动装置，诸如，个人数字处理、蜂窝电话、智能电话、可穿戴设备和其它类似的计算装置。本文所示的部件、它们的连接和关系、以及它们的功能仅仅作为示例，并且不意在限制本文中描述的和/或者要求的本公开的实现。Referring to FIG. 13 , a structural block diagram of anelectronic device 1300 that can serve as a server or a client of the present disclosure, which is an example of a hardware device that can be applied to various aspects of the present disclosure, will now be described. Electronic device is intended to mean various forms of digital electronic computing equipment, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

如图13所示，电子设备1300包括计算单元1301，其可以根据存储在只读存储器（ROM）1302中的计算机程序或者从存储单元1308加载到随机访问存储器（RAM）1303中的计算机程序，来执行各种适当的动作和处理。在RAM 1303中，还可存储设备1300操作所需的各种程序和数据。计算单元1301、ROM 1302以及RAM 1303通过总线1304彼此相连。输入/输出（I/O）接口1305也连接至总线1304。As shown in FIG. 13 , anelectronic device 1300 includes acomputing unit 1301 , which can calculate the Various appropriate actions and processes are performed. In theRAM 1303, various programs and data necessary for the operation of thedevice 1300 can also be stored. Thecomputing unit 1301 ,ROM 1302 , andRAM 1303 are connected to each other through abus 1304 . An input/output (I/O)interface 1305 is also connected to thebus 1304 .

电子设备1300中的多个部件连接至I/O接口1305，包括：输入单元1306、输出单元1307、存储单元1308以及通信单元1309。输入单元1306可以是能向电子设备1300输入信息的任何类型的设备，输入单元1306可以接收输入的数字或字符信息，以及产生与电子设备的用户设置和/或功能控制有关的键信号输入。输出单元1307可以是能呈现信息的任何类型的设备，并且可以包括但不限于显示器、扬声器、视频/音频输出终端、振动器和/或打印机。存储单元1308可以包括但不限于磁盘、光盘。通信单元1309允许电子设备1300通过诸如因特网的计算机网络和/或各种电信网络与其他设备交换信息/数据，并且可以包括但不限于调制解调器、网卡、红外通信设备、无线通信收发机和/或芯片组，例如蓝牙TM设备、WiFi设备、WiMax设备、蜂窝通信设备和/或类似物。Multiple components in theelectronic device 1300 are connected to the I/O interface 1305 , including: aninput unit 1306 , anoutput unit 1307 , astorage unit 1308 and acommunication unit 1309 . Theinput unit 1306 can be any type of device capable of inputting information to theelectronic device 1300, and theinput unit 1306 can receive input numeric or character information, and generate key signal input related to user settings and/or function control of the electronic device. Theoutput unit 1307 may be any type of device capable of presenting information, and may include, but is not limited to, a display, a speaker, a video/audio output terminal, a vibrator, and/or a printer. Thestorage unit 1308 may include, but is not limited to, a magnetic disk and an optical disk. Thecommunication unit 1309 allows theelectronic device 1300 to exchange information/data with other devices through a computer network such as the Internet and/or various telecommunication networks, and may include but not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver and/or a chip Groups, such as Bluetooth™ devices, WiFi devices, WiMax devices, cellular communication devices, and/or the like.

计算单元1301可以是各种具有处理和计算能力的通用和/或专用处理组件。计算单元1301的一些示例包括但不限于中央处理单元（CPU）、图形处理单元（GPU）、各种专用的人工智能（AI）计算芯片、各种运行机器学习模型算法的计算单元、数字信号处理器（DSP）、以及任何适当的处理器、控制器、微控制器等。计算单元1301执行上文所描述的各个方法和处理。例如，在一些实施例中，作业批改方法均可被实现为计算机软件程序，其被有形地包含于机器可读介质，例如存储单元1308。在一些实施例中，计算机程序的部分或者全部可以经由ROM 1302和/或通信单元1309而被载入和/或安装到电子设备1300上。在一些实施例中，计算单元1301可以通过其他任何适当的方式（例如，借助于固件）而被配置为执行作业批改方法。Thecomputing unit 1301 may be various general-purpose and/or special-purpose processing components having processing and computing capabilities. Some examples ofcomputing units 1301 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. Thecalculation unit 1301 executes the various methods and processes described above. For example, in some embodiments, the job correction method can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as thestorage unit 1308 . In some embodiments, part or all of the computer program may be loaded and/or installed on theelectronic device 1300 via theROM 1302 and/or thecommunication unit 1309 . In some embodiments, thecomputing unit 1301 may be configured in any other appropriate way (for example, by means of firmware) to execute the job correction method.

用于实施本公开的方法的程序代码可以采用一个或多个编程语言的任何组合来编写。这些程序代码可以提供给通用计算机、专用计算机或其他可编程数据处理装置的处理器或控制器，使得程序代码当由处理器或控制器执行时使流程图和/或框图中所规定的功能/操作被实施。程序代码可以完全在机器上执行、部分地在机器上执行，作为独立软件包部分地在机器上执行且部分地在远程机器上执行或完全在远程机器或服务器上执行。Program codes for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special purpose computer, or other programmable data processing devices, so that the program codes, when executed by the processor or controller, make the functions/functions specified in the flow diagrams and/or block diagrams Action is implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

在本公开的上下文中，机器可读介质可以是有形的介质，其可以包含或存储以供指令执行系统、装置或设备使用或与指令执行系统、装置或设备结合地使用的程序。机器可读介质可以是机器可读信号介质或机器可读储存介质。机器可读介质可以包括但不限于电子的、磁性的、光学的、电磁的、红外的、或半导体系统、装置或设备，或者上述内容的任何合适组合。机器可读存储介质的更具体示例会包括基于一个或多个线的电气连接、便携式计算机盘、硬盘、随机存取存储器（RAM）、只读存储器（ROM）、可擦除可编程只读存储器（EPROM或快闪存储器）、光纤、便捷式紧凑盘只读存储器（CD-ROM）、光学储存设备、磁储存设备、或上述内容的任何合适组合。In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer disks, hard disks, Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable Read Only Memory (EPROM or flash memory), fiber optics, compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

如本公开使用的，术语“机器可读介质”和“计算机可读介质”指的是用于将机器指令和/或数据提供给可编程处理器的任何计算机程序产品、设备、和/或装置(例如，磁盘、光盘、存储器、可编程逻辑装置(PLD))，包括，接收作为机器可读信号的机器指令的机器可读介质。术语“机器可读信号”指的是用于将机器指令和/或数据提供给可编程处理器的任何信号。As used in this disclosure, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or means for providing machine instructions and/or data to a programmable processor (eg, magnetic disk, optical disk, memory, programmable logic device (PLD)), including machine-readable media that receive machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.

为了提供与用户的交互，可以在计算机上实施此处描述的系统和技术，该计算机具有：用于向用户显示信息的显示装置（例如，CRT（阴极射线管）或者LCD（液晶显示器）监视器）；以及键盘和指向装置（例如，鼠标或者轨迹球），用户可以通过该键盘和该指向装置来将输入提供给计算机。其它种类的装置还可以用于提供与用户的交互；例如，提供给用户的反馈可以是任何形式的传感反馈（例如，视觉反馈、听觉反馈、或者触觉反馈）；并且可以用任何形式（包括声输入、语音输入或者、触觉输入）来接收来自用户的输入。To provide for interaction with the user, the systems and techniques described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user ); and a keyboard and pointing device (eg, a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and may be in any form (including Acoustic input, voice input, or tactile input) to receive input from the user.

可以将此处描述的系统和技术实施在包括后台部件的计算系统（例如，作为数据服务器）、或者包括中间件部件的计算系统（例如，应用服务器）、或者包括前端部件的计算系统（例如，具有图形用户界面或者网络浏览器的用户计算机，用户可以通过该图形用户界面或者该网络浏览器来与此处描述的系统和技术的实施方式交互）、或者包括这种后台部件、中间件部件、或者前端部件的任何组合的计算系统中。可以通过任何形式或者介质的数字数据通信（例如，通信网络）来将系统的部件相互连接。通信网络的示例包括：局域网（LAN）、广域网（WAN）和互联网。The systems and techniques described herein can be implemented on a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system can be interconnected by any form or medium of digital data communication (eg, a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

计算机系统可以包括客户端和服务器。客户端和服务器一般远离彼此并且通常通过通信网络进行交互。通过在相应的计算机上运行并且彼此具有客户端-服务器关系的计算机程序来产生客户端和服务器的关系。A computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other.

需要说明的是，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。It should be noted that in this article, relative terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply these No such actual relationship or order exists between entities or operations. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

以上所述仅是本公开的具体实施方式，使本领域技术人员能够理解或实现本公开。对这些实施例的多种修改对本领域的技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本公开的精神或范围的情况下，在其它实施例中实现。因此，本公开将不会被限制于本文所述的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。The above descriptions are only specific implementation manners of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (19)

Translated fromChinese

1.一种作业批改方法，包括：1. A method for correcting homework, comprising:获取当前作业的作答结果图像；Obtain the answer result image of the current assignment;对所述作答结果图像进行文本识别，得到初始作答结果；performing text recognition on the answer result image to obtain an initial answer result;获取所述当前作业的题型，并基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果；Obtaining the question type of the current assignment, and adjusting the initial answer result based on the question type of the current assignment to obtain a target answer result;基于所述当前作业的题型和所述当前作业的标准作答结果，对所述目标作答结果进行判断，确定所述当前作业的批改结果。Based on the question type of the current homework and the standard answer result of the current homework, the target answer result is judged to determine the correcting result of the current homework.2.如权利要求1所述的作业批改方法，其中，获取当前作业的作答结果图像，包括：2. The operation correction method as claimed in claim 1, wherein obtaining the answer result image of the current operation comprises:获取所述当前作业图像，并根据所述当前作业图像中预先设定作答区域的区域坐标提取得到所述作答结果图像。The current job image is acquired, and the answer result image is extracted according to area coordinates of a preset answer area in the current job image.3.如权利要求1所述的作业批改方法，其中，获取当前作业的作答结果图像，包括：3. The operation correction method as claimed in claim 1, wherein obtaining the answer result image of the current operation comprises:获取作答工具在作答区域内的轨迹坐标，并根据所述轨迹坐标生成所述作答结果图像。The track coordinates of the answering tool in the answering area are acquired, and the answer result image is generated according to the track coordinates.4.如权利要求1所述的作业批改方法，其中，对所述作答结果图像进行文本识别，得到初始作答结果，包括：4. The operation correcting method as claimed in claim 1, wherein, carrying out text recognition to described answer result image, obtains initial answer result, comprises:采用预先构建的文本识别模型对所述作答结果图像进行识别，得到所述初始作答结果，其中，所述文本识别模型中包括检测模型和识别模型，所述检测模型用于提取所述作答结果图像中的作答位置并输入至所述识别模型，所述识别模型用于对所述作答位置的图像进行文本识别得到所述初始作答结果。Using a pre-built text recognition model to recognize the answer result image to obtain the initial answer result, wherein the text recognition model includes a detection model and a recognition model, and the detection model is used to extract the answer result image The answer position in and input to the recognition model, and the recognition model is used to perform text recognition on the image of the answer position to obtain the initial answer result.5.如权利要求1所述的作业批改方法，其中，所述当前作业的题型包括选择题、填空题或主观题。5. The method for correcting homework according to claim 1, wherein the question types of the current homework include multiple-choice questions, fill-in-the-blank questions or subjective questions.6.如权利要求5所述的作业批改方法，其中，若所述当前作业的题型是所述选择题，所述调整处理包括字符调整处理、数字调整处理、数量调整处理中的至少一种；若所述当前作业的题型是所述填空题或所述主观题，则所述调整处理包括区域调整处理或单位调整处理。6. The operation correction method as claimed in claim 5, wherein, if the question type of the current operation is the multiple-choice question, the adjustment processing includes at least one of character adjustment processing, number adjustment processing, and quantity adjustment processing ; If the question type of the current assignment is the filling-in-the-blank question or the subjective question, the adjustment processing includes area adjustment processing or unit adjustment processing.7.如权利要求6所述的作业批改方法，其中，基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果，包括：7. The operation correction method as claimed in claim 6, wherein, based on the question type of the current operation, the initial answer result is adjusted to obtain the target answer result, including:若所述当前作业的题型是所述选择题，将所述初始作答结果在绑定字符库中进行匹配，匹配则得到目标绑定字符，其中，所述绑定字符库中包括多个预设的绑定字符；If the question type of the current job is the multiple-choice question, the initial answer result is matched in the bound character library, and the target bound character is obtained after matching, wherein the bound character library includes a plurality of preset The binding character set;基于预设的绑定字符与标准结果的关联关系，确定所述目标绑定字符对应的目标标准结果，将所述目标标准结果确定为所述目标作答结果，其中，所述绑定字符与标准结果的关联关系用于确定所述多个预设的绑定字符与标准结果之间的对应关系，一个所述标准结果关联至少一个所述绑定字符。Based on the preset association relationship between bound characters and standard results, determine the target standard results corresponding to the target bound characters, and determine the target standard results as the target answer results, wherein the bound characters and standard The association relationship of the results is used to determine the correspondence between the plurality of preset bound characters and standard results, and one standard result is associated with at least one bound character.8.如根据权利要求6所述的作业批改方法，其中，基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果，包括：8. The operation correction method according to claim 6, wherein, based on the question type of the current operation, the initial answer result is adjusted to obtain the target answer result, including:若所述当前作业的题型是所述选择题，对所述初始作答结果进行数字识别；If the question type of the current assignment is the multiple-choice question, digitally identify the initial answer result;若识别到数字，删除所述初始作答结果中的数字，得到所述目标作答结果；If the number is identified, delete the number in the initial answer result to obtain the target answer result;或者，若所述初始作答结果中的数字判定为预设数字且字符数量为2，则删除所述初始作答结果中的所述预设数字，得到所述目标作答结果。Alternatively, if the number in the initial answer result is determined to be a preset number and the number of characters is 2, then delete the preset number in the initial answer result to obtain the target answer result.9.如权利要求6所述的作业批改方法，其中，基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果，包括：9. The operation correction method as claimed in claim 6, wherein, based on the question type of the current operation, the initial answer result is adjusted to obtain the target answer result, including:若所述当前作业的题型是所述选择题，对所述初始作答结果进行数字识别；If the question type of the current assignment is the multiple-choice question, digitally identify the initial answer result;若未识别到数字，则将所述初始作答结果与预设的多个标准结果进行匹配，基于文本识别模型获取各所述标准结果的识别置信度；If the number is not recognized, matching the initial answer result with a plurality of preset standard results, and obtaining the recognition confidence of each standard result based on the text recognition model;将至少两个所述标准结果中识别置信度最高的标准结果确定为所述目标作答结果。Determining the standard result with the highest recognition confidence among the at least two standard results as the target answer result.10.如权利要求6所述的作业批改方法，其中，基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果，包括：10. The operation correction method as claimed in claim 6, wherein, based on the question type of the current operation, the initial answer result is adjusted to obtain the target answer result, including:若所述当前作业的题型为所述填空题或所述主观题，对所述初始作答结果进行文本识别；If the question type of the current assignment is the filling-in-the-blank question or the subjective question, perform text recognition on the initial answer result;若未识别到文本，对所述当前作业的完整图像进行文本识别，得到文本识别结果；If no text is recognized, perform text recognition on the complete image of the current job to obtain a text recognition result;若所述文本识别结果中包括除所述当前作业的题目文本之外的目标文本，则将该目标文本确定为目标作答结果。If the text recognition result includes target text other than the topic text of the current assignment, then determine the target text as the target answer result.11.如权利要求6所述的作业批改方法，其中，基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果，包括：11. The operation correction method as claimed in claim 6, wherein, based on the question type of the current operation, the initial answer result is adjusted to obtain the target answer result, including:若所述当前作业的题型为所述填空题或所述主观题，对所述初始作答结果进行计量单位识别；If the question type of the current assignment is the filling-in-the-blank question or the subjective question, performing unit of measurement identification on the initial answer result;若识别到计量单位，提取所述初始作答结果中的计量单位；If the unit of measurement is identified, extract the unit of measurement in the initial answer result;对所述计量单位进行中英文格式的转换，得到包括中英文格式的计量单位的目标作答结果。The Chinese-English format conversion is performed on the measurement unit, and the target answer result including the measurement unit in the Chinese-English format is obtained.12.如权利要求1所述的作业批改方法，其中，若所述当前作业的题型为选择题，基于所述当前作业的题型和所述当前作业的标准作答结果，对所述目标作答结果进行判断，确定所述当前作业的批改结果，包括：12. The operation correction method as claimed in claim 1, wherein, if the question type of the current operation is a multiple-choice question, answer the target based on the question type of the current operation and the standard answer result of the current operation Results are judged to determine the correction result of the current job, including:将所述目标作答结果与所述标准作答结果进行对比；comparing the target answer result with the standard answer result;根据对比结果确定所述当前作业的批改结果。The correction result of the current job is determined according to the comparison result.13.如权利要求1所述的作业批改方法，其中，若所述当前作业的题型为填空题，基于所述当前作业的题型和所述当前作业的标准作答结果，对所述目标作答结果进行判断，确定所述当前作业的批改结果，包括：13. The operation correction method as claimed in claim 1, wherein, if the question type of the current operation is a fill-in-the-blank question, answer the target based on the question type of the current operation and the standard answer result of the current operation Results are judged to determine the correction result of the current job, including:将所述目标作答结果与所述标准作答结果进行数据结构的对比，或者字符数量的对比；Comparing the data structure of the target answer result with the standard answer result, or comparing the number of characters;根据对比结果确定所述当前作业的批改结果；其中，所述数据结构用于表征作答结果包括的数据类型；字符数量用于表征作答结果包括的字符的数量。Determine the correction result of the current job according to the comparison result; wherein, the data structure is used to represent the data type included in the answer result; the number of characters is used to represent the number of characters included in the answer result.14.如权利要求1所述的作业批改方法，其中，若所述当前作业的题型为主观题中的文科主观题，基于所述当前作业的题型和所述当前作业的标准作答结果，对所述目标作答结果进行判断，确定所述当前作业的批改结果，包括：14. The operation correction method as claimed in claim 1, wherein, if the question type of the current operation is a liberal arts subjective question in the subjective question, based on the standard answer result of the question type of the current operation and the current operation, Judging the target answer result and determining the correction result of the current job, including:基于预设批改算法识别所述目标作答结果的至少一种批改参数，所述批改参数包括字数、标题、错别字、语法错误、搭配不当和结构错误中的至少一项；Identifying at least one correction parameter of the target answer result based on a preset correction algorithm, the correction parameter including at least one of word count, title, typo, grammatical error, improper collocation and structural error;展示所述至少一种批改参数以及所述当前作业的标准作答结果，获取用户基于所述标准作答结果以及所述至少一种批改参数输入的所述当前作业的批改结果。The at least one correction parameter and the standard answer result of the current assignment are displayed, and the correction result of the current assignment entered by the user based on the standard answer result and the at least one correction parameter is acquired.15.如权利要求1所述的作业批改方法，其中，若所述当前作业的题型为主观题中的理科主观题，基于所述当前作业的题型和所述当前作业的标准作答结果，对所述目标作答结果进行判断，确定所述当前作业的批改结果，包括：15. The operation correction method as claimed in claim 1, wherein, if the question type of the current operation is a science subjective question in the subjective question, based on the question type of the current operation and the standard answer result of the current operation, Judging the target answer result and determining the correction result of the current job, including:确定所述目标作答结果中位于最后一位的当前最终结果；determining the current final result at the last place among the target answer results;将所述目标作答结果的当前最终结果与所述标准作答结果中的标准最终结果进行对比；comparing the current final result of the target answer result with the standard final result in the standard answer result;若对比结果为所述当前最终结果与所述标准最终结果相同，则所述当前作业的批改结果确定为正确；否则所述当前作业的批改结果确定为错误。If the comparison result is that the current final result is the same as the standard final result, then the correction result of the current job is determined to be correct; otherwise, the correction result of the current job is determined to be wrong.16.如权利要求1所述的作业批改方法，其中，在获取当前作业的作答结果图像之前，所述方法还包括：16. The operation correction method as claimed in claim 1, wherein, before obtaining the answer result image of the current operation, the method further comprises:获取且展示标准作答规范信息，其中，所述标准作答规范信息用于指导用户作答当前作业具体操作的文字表述。Acquiring and displaying standard answer specification information, wherein the standard answer specification information is used to guide the user to answer the textual expression of the specific operation of the current job.17.一种作业批改装置，包括：17. A homework correction device, comprising:图像获取模块，用于获取当前作业的作答结果图像；An image acquisition module, configured to acquire an answer result image of the current job;文本识别模块，用于对所述作答结果图像进行文本识别，得到初始作答结果；A text recognition module, configured to perform text recognition on the answer result image to obtain an initial answer result;结果优化模块，用于获取所述当前作业的题型，并基于所述当前作业的题型对所述初始作答结果进行调整处理，得到目标作答结果；A result optimization module, configured to obtain the question type of the current homework, and adjust the initial answer result based on the question type of the current homework to obtain a target answer result;结果判断模块，用于基于所述当前作业的题型和所述当前作业的标准作答结果，对所述目标作答结果进行判断，确定所述当前作业的批改结果。A result judging module, configured to judge the target answer result based on the question type of the current homework and the standard answer result of the current homework, and determine the correcting result of the current homework.18.一种电子设备，包括：18. An electronic device comprising:处理器；以及processor; and存储程序的存储器，memory for storing programs,其中，所述程序包括指令，所述指令在由所述处理器执行时使所述处理器执行根据权利要求1-16中任一项所述的方法。Wherein, the program includes instructions which, when executed by the processor, cause the processor to perform the method according to any one of claims 1-16.19.一种存储有计算机指令的非瞬时计算机可读存储介质，其中，所述计算机指令用于使所述计算机执行根据权利要求1-16中任一项所述的方法。19. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to any one of claims 1-16.

Images