A handwritten answer processing method, device, equipment and storage medium

Abstract

This application discloses a handwritten answer processing method, apparatus, device, and storage medium. The technical solution provided by this application acquires an image of handwritten content input on a question interface, performs character recognition processing on the text content in the handwritten content image to obtain handwritten recognized text, matches the handwritten recognized text with questions, determines the target question corresponding to the handwritten recognized text based on the question matching result, and performs answer grading processing on the handwritten recognized text based on the answer to the target question to obtain a grading result. Users do not need to answer questions in a restricted answer area, providing users with a more free answer area and accurately recognizing user answers, effectively improving the user's answering experience.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a handwritten answer processing method, apparatus, device, and storage medium. Background Technology

[0002] With the development of computer technology, learning machines and other practice products offer more and more functions. Users can answer questions on the practice software of these products, and the software can grade the answers based on the user's responses.

[0003] Currently, quiz apps typically restrict users to answering questions in designated areas. This handwritten answering method limits users' freedom of choice, and users may write in the designated area when nervous, resulting in their answers not being recognized correctly and affecting their quiz experience. Summary of the Invention

[0004] This application provides a handwritten answer processing method, apparatus, device, and storage medium to solve the technical problem in related technologies where handwritten answer processing schemes restrict users to answering in designated areas, affecting the user's answering experience. It can provide users with a more free answering area and accurately identify the user's answer content, effectively improving the user's answering experience.

[0005] In a first aspect, embodiments of this application provide a handwritten answer processing method, including:

[0006] The handwritten content image entered on the question interface is obtained, and the text content in the handwritten content image is processed by character recognition to obtain handwritten recognized text;

[0007] The handwritten text is matched with a question, and the target question corresponding to the handwritten text is determined based on the question matching result.

[0008] The handwritten recognition text is graded based on the answer to the target question, and the grading result is obtained.

[0009] In a second aspect, embodiments of this application provide a handwritten answer processing device, including a handwriting recognition module, a question matching module, and a grading processing module, wherein:

[0010] The handwriting recognition module is used to acquire the handwritten content image input on the question interface, and to perform character recognition processing on the text content in the handwritten content image to obtain handwritten recognition text.

[0011] The question matching module is used to match questions based on the handwritten recognition text and determine the target question corresponding to the handwritten recognition text based on the question matching result.

[0012] The grading module is used to grade the handwritten recognition text according to the answer to the target question, and obtain the grading result.

[0013] In a third aspect, embodiments of this application provide a handwritten answer processing device, including: a memory and one or more processors;

[0014] The memory is used to store one or more programs;

[0015] When the one or more programs are executed by the one or more processors, the one or more processors implement the handwritten answer processing method as described in the first aspect.

[0016] In a fourth aspect, embodiments of this application provide a storage medium for storing computer-executable instructions, which, when executed by a computer processor, are used to perform the handwritten answer processing method as described in the first aspect.

[0017] This application embodiment acquires an image of handwritten content input on the question interface, performs text recognition processing on the text content in the handwritten content image to obtain handwritten recognition text, performs question matching based on the handwritten recognition text, determines the target question corresponding to the handwritten recognition text based on the question matching result, and performs answer grading processing on the handwritten recognition text based on the answer to the target question to obtain the grading result. Users do not need to answer questions in a restricted answer area, providing users with a more free answer area, and accurately recognizing user answers, effectively improving the user's answering experience. Attached Figure Description

[0018] Figure 1 This is a flowchart of a handwritten answer processing method provided in an embodiment of this application;

[0019] Figure 2 This is a flowchart of another handwritten answer processing method provided in an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of the structure of a handwritten answer processing device provided in an embodiment of this application;

[0021] Figure 4 This is a schematic diagram of the structure of a handwritten answer processing device provided in an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this application clearer, specific embodiments of this application will be described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely for explaining this application and not for limiting it. It should also be noted that, for ease of description, only the parts relevant to this application are shown in the drawings, not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe operations (or steps) as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but additional steps not included in the drawings may also be present. The above processes can correspond to methods, functions, procedures, subroutines, subroutines, etc.

[0023] Existing question-solving solutions typically restrict users to answering questions within a designated area, limiting their freedom of expression. Under pressure, users may accidentally stray outside the designated area, resulting in incorrect answers and negatively impacting the learning experience and performance evaluation. Furthermore, this single designated-area answering mode cannot accommodate different users' writing habits and answering styles, proving extremely inconvenient for users who habitually jot down their thoughts and highlight key points around the questions. Moreover, related question-solving software focuses solely on the correctness of the answer, ignoring the user's writing and answering habits, which hinders the development of good learning habits and has limited impact on improving users' learning abilities. Therefore, this application provides a handwritten answer processing method to address the technical problems of existing handwritten answer processing solutions that restrict users to answering questions within a designated area, negatively affecting the user's learning experience and limiting their ability to improve learning.

[0024] Figure 1 A flowchart of a handwritten answer processing method provided in an embodiment of this application is given. The handwritten answer processing method provided in this embodiment of the application can be executed by a handwritten answer processing device, which can be implemented by hardware and / or software and integrated into a handwritten answer processing equipment.

[0025] The following description uses a handwritten answer processing device as an example to illustrate the handwritten answer processing method. (Reference) Figure 1 The handwritten answer processing method includes:

[0026] S110: Obtain the image of the handwritten content entered on the question interface, and perform text recognition processing on the text content in the handwritten content image to obtain the handwritten recognized text.

[0027] In one embodiment, the handwriting answer processing device provided in this application is configured with a practice product (e.g., practice software). When the practice software is running, a corresponding question interface can be displayed on the screen of the handwriting answer processing device. One or more questions are displayed on the question interface, and the user can handwrite the answer on the question interface to answer the question.

[0028] For example, an image of the handwritten content entered by the user on the question interface is captured as the handwritten content image. The handwritten content image records the user's handwritten answer to the question (e.g., the question displayed on the current question interface, or the question displayed on other question interfaces). Optionally, when practicing answering questions, multiple questions can be arranged on different question interfaces, and the user can see different questions on different question interfaces.

[0029] After acquiring the handwritten content image, character recognition processing can be performed on the text content in the handwritten content image to obtain handwritten recognized text (editable text information). The handwritten recognized text reflects the text content corresponding to the user's handwritten answer to the question on the question interface. Optionally, character recognition of the text content in the handwritten content image can be performed through one or more combinations of optical character recognition (OCR), handwriting recognition, and neural network recognition (e.g., convolutional neural network (CNN), recurrent neural network (RNN), or a combination thereof).

[0030] S120: Match the handwritten text with the target question and determine the target question corresponding to the handwritten text based on the matching result.

[0031] For example, after recognizing handwritten text from a handwritten content image, question matching can be performed based on the handwritten text to obtain a question matching result, and the target question corresponding to the handwritten text can be determined based on the question matching result. Optionally, question matching can be performed in a preset question database, or it can be performed among the questions displayed on the question interface.

[0032] In one embodiment, when a user answers a question on the question interface, they can handwrite the answer in the answer area corresponding to the target question on the question interface, or they can handwrite the answer in an area outside the answer area of ​​the target question. That is, the handwritten content image can be located in the answer area of ​​the target question or outside the answer area of ​​the target question.

[0033] In one embodiment, when a user answers a question on the question interface, they can either handwrite the answer on the question interface where the target question is located (i.e., the current question interface displays the target question) or handwrite the answer on a question interface other than the target question (i.e., the current question interface does not display the target question). In other words, the handwritten content image can be located on the question interface where the target question is located or outside the question interface where the target question is located.

[0034] S130: Based on the answer to the target question, perform answer correction processing on the handwritten recognition text to obtain the correction result.

[0035] For example, the system obtains the answer (i.e., the correct answer) to the target question, performs grading on the handwritten digitized text based on the answer, and displays the grading result. Users can determine whether they have correctly answered the target question based on the matching result. Specifically, if the handwritten digitized text matches the answer, it is considered correct; otherwise, it is considered incorrect.

[0036] Optionally, whether the handwritten text matches the answer to the question can be determined according to preset judgment rules or by using a trained answer grading model. The answer to the question can be stored locally on the answer analysis device or on a server (e.g., a database).

[0037] The above describes a process where the system acquires an image of handwritten text input on the question interface, performs text recognition processing on the text in the handwritten text image to obtain handwritten recognition text, matches the handwritten recognition text with questions, determines the target question corresponding to the handwritten recognition text based on the matching results, and performs answer grading processing on the handwritten recognition text based on the answer to the target question to obtain the grading result. Users are not required to answer questions in a restricted answer area, providing them with a more free answer area and accurately recognizing user answers, effectively improving the user's answering experience.

[0038] Based on the above embodiments, Figure 2 A flowchart of another handwritten answer processing method provided in an embodiment of this application is given, which is a concretization of the above-described handwritten answer processing method. (Reference) Figure 2 The handwritten answer processing method includes:

[0039] S210: Obtain the image of the handwritten content entered on the question interface, and segment the handwritten content image into multiple character content images.

[0040] S220: Input multiple character content images into the trained text recognition model, and perform character recognition processing on the multiple character content to obtain handwritten recognition text.

[0041] For example, after obtaining the image of the handwritten content entered on the question interface, the handwritten content image is segmented into multiple character content images, wherein each character content image records one or more characters of the text content in the user's handwritten answer.

[0042] In one embodiment, a handwritten content image can be segmented into multiple character content images according to a preset character segmentation algorithm, or it can be segmented into multiple character content images based on a deep learning model, or it can be segmented into multiple character content images based on a combination of character segmentation algorithms and deep learning models. For example, a coarse segmentation of the handwritten content image can be performed using a character segmentation algorithm, and then a deep learning model can be used to perform fine segmentation on the coarse segmentation result, quickly and accurately obtaining multiple character content images. This approach balances segmentation speed and accuracy and is suitable for mobile deployment.

[0043] Optionally, the character segmentation algorithm provided in this application may be a combination of one or more of the projection analysis method, connected component analysis method, and skeletonization segmentation method, and the deep learning model may be a combination of one or more of the CNN-based segmentation model, the end-to-end segmentation and recognition joint model, and the Transformer segmentation model.

[0044] Furthermore, after segmenting the handwritten content image into multiple character content images, these multiple character content images can be input into a trained text recognition model. The text recognition model then performs character recognition processing on the multiple character content images to obtain the handwritten recognized text.

[0045] For example, a text recognition model performs character recognition processing on each character content to obtain the recognized text corresponding to each character content image. Then, based on the position of each recognized character content image within the handwritten content image, the recognized characters are sorted (either row-level or inline-level sorting) to obtain the handwritten recognized text composed of these characters. This scheme effectively improves the accuracy of character recognition and thus the accuracy of answer processing by segmenting the handwritten content image into multiple character content images and then using the text recognition model to perform character recognition processing on the character content.

[0046] S230: Match the handwritten text with the target question and determine the target question corresponding to the handwritten text based on the matching result.

[0047] In one possible embodiment, the handwritten answer processing method provided in this application performs question matching based on the handwritten recognition text and determines the target question corresponding to the handwritten recognition text based on the question matching result. This can be done by: performing question matching on the handwritten recognition text in a question database and determining the target question corresponding to the handwritten recognition text based on the question matching result.

[0048] The question database records the answers to multiple questions. After the handwritten content image is processed by text recognition to obtain the handwritten recognition text, the handwritten recognition text can be matched with questions in the question database, and the question matching result is determined based on the similarity between the handwritten recognition text and the answers to each question.

[0049] For example, the question with the highest similarity can be used as the target question for the handwritten text, or questions with a similarity threshold can be identified, and the question whose position on the question interface is closest to the handwritten content image can be used as the target question for the handwritten text. This application determines the target question by matching the handwritten text with questions in a question database, which can accurately identify the target question and effectively improve the accuracy of handwritten answer processing.

[0050] In one embodiment, the handwritten answer processing method provided in this application performs question matching on the handwritten recognition text in the question database, which may involve: extracting keywords and / or contextual semantic information from the handwritten recognition text; and performing question matching on the question database based on the keywords and / or contextual semantic information.

[0051] For example, keyword extraction and / or contextual semantic information extraction are performed on the handwritten recognition text to obtain the keywords and / or contextual semantic information of the handwritten recognition text. Then, question matching is performed based on the keywords and / or contextual semantic information and the question answers corresponding to the questions recorded in the question database. Finally, the question matching result is determined based on the similarity between the keywords and / or contextual semantic information and the question answers of each question.

[0052] Optionally, keywords and / or contextual semantic information of the handwritten recognition text can be extracted according to preset extraction rules, or keywords and / or contextual semantic information of the handwritten recognition text can be extracted through a trained feature extraction model. This application improves the accuracy of target question matching by matching questions based on keywords and / or contextual semantic information of the handwritten recognition text, thereby eliminating interference from irrelevant information and effectively improving the accuracy of handwritten answer processing.

[0053] In one possible embodiment, the handwritten answer processing method provided in this application performs question matching based on the handwritten recognition text and determines the target question corresponding to the handwritten recognition text based on the question matching result. This can be done by performing question matching based on the question displayed on the question interface and the handwritten recognition text, and determining the target question corresponding to the handwritten recognition text based on the question matching result.

[0054] For example, the questions displayed on the question interface can be determined, and the questions displayed on the question interface can be matched with the handwritten recognition text to determine the question with the highest matching degree with the handwritten recognition text, and this question can be used as the target question corresponding to the handwritten recognition text. For example, the similarity between the handwritten recognition text and the answers to each question displayed on the question interface can be determined, and the question matching result can be determined based on the similarity between the handwritten recognition text and the answers to each question.

[0055] Optionally, the questions matched with the handwritten text can be those displayed on the currently displayed question interface or on other question interfaces. The questions displayed on the question interface can be determined based on questions stored locally on the analysis device, questions recorded on the server, or by performing text recognition on the question interface and matching the recognition results with questions stored locally on the analysis device or recorded on the server.

[0056] In one embodiment, matching the question displayed on the question interface with the handwritten recognition text can involve extracting keywords and / or contextual semantic information from the handwritten recognition text; and then matching the keywords and / or contextual semantic information with the corresponding answer to the question displayed on the question interface. This solution accurately identifies the target question among the questions on the question interface by matching the questions on the question interface with the handwritten recognition text, effectively improving the accuracy of handwritten answer processing.

[0057] S240: Based on the answer to the target question, perform answer correction processing on the handwritten recognition text to obtain the correction result.

[0058] In one possible embodiment, the handwritten answer processing method provided in this application, after acquiring the image of the handwritten content entered on the question interface, may further include:

[0059] S250: Extract writing feature information from handwritten content images.

[0060] S260: Based on writing characteristic information and preset writing habit standards, analyze writing habits to obtain writing habit analysis results.

[0061] For example, after acquiring an image of the handwritten content entered on the question interface, writing feature information is extracted from the handwritten content image based on a preset feature extraction algorithm or a trained feature extraction model. The writing feature information provided by this solution may include one or more combinations of stroke order, writing speed, font size, font spacing, and answer layout.

[0062] Furthermore, based on the determined writing feature information and preset writing habit standards, writing habit analysis is performed to obtain the user's writing habit analysis results. For example, a trained habit analysis model (such as a support vector machine (SVM) model) can be used to perform writing habit analysis based on writing feature information and preset writing habit standards and output the corresponding writing habit analysis results.

[0063] Among them, the results of the writing habit analysis can be used to indicate problems in the user's writing process, such as whether the stroke order is correct, whether the writing speed is too fast or too slow, whether the font is too large or too small, whether the font spacing is too large or too small, and whether the answer layout is reasonable (whether the writing is done in an appropriate area), etc.

[0064] In one embodiment, after analyzing writing habits based on writing feature information and obtaining the results of the writing habit analysis, writing standard suggestions can be generated and displayed based on the results of the writing habit analysis.

[0065] For example, writing style suggestions can be generated based on the results of writing habit analysis, and these suggestions can be displayed through pop-ups, voice prompts, text prompts, and illustrative prompts (such as displaying standard writing habits through pictures). These writing style suggestions can be used to guide users in developing good writing habits. For instance, the results of writing habit analysis can be input into a large language model, which can then generate corresponding writing style suggestions based on the analysis results.

[0066] The above describes a process where the system acquires an image of handwritten content input on the question interface, performs text recognition processing on the text within the image to obtain handwritten recognized text, matches the handwritten recognized text with questions, determines the target question corresponding to the handwritten recognized text based on the matching results, and then performs answer correction processing on the handwritten recognized text based on the answer to the target question to obtain the correction result. Users are not required to answer questions within a restricted answer area, providing them with a more flexible answering space and accurately recognizing user answers, effectively improving the user's question-answering experience. Furthermore, by segmenting the handwritten content image into multiple character content images and then using a text recognition model to perform text recognition processing on the character content to obtain handwritten recognized text, the accuracy of text recognition and the accuracy of answer processing are effectively improved. This application also accurately analyzes users' writing habits based on writing feature information, guides users to standardize their writing habits, and helps users form good learning habits. This application, based on user answering behavior and learning needs, grants users greater freedom in answering questions, aligns more with their natural answering habits, enhances user experience, and supports text recognition of handwritten content images, enabling accurate matching and correction of answers through image and text processing. Furthermore, it guides users to develop good writing and answering habits through handwriting habit analysis, helping to improve learning efficiency and quality, identifying potential user problems and providing targeted suggestions, thus promoting the development of good learning habits.

[0067] Figure 3 A schematic diagram of a handwritten answer processing device according to an embodiment of this application is provided. (Reference) Figure 3 The handwritten answer processing device includes a handwriting recognition module 31, a question matching module 32, and a grading processing module 33.

[0068] The handwriting recognition module 31 is used to acquire the handwritten content image input on the question interface and perform text recognition processing on the text content in the handwritten content image to obtain the handwritten recognition text; the question matching module 32 is used to match questions based on the handwritten recognition text and determine the target question corresponding to the handwritten recognition text based on the question matching result; the grading processing module 33 is used to grade the handwritten recognition text based on the question answer corresponding to the target question and obtain the grading result.

[0069] The above describes a process where the system acquires an image of handwritten text input on the question interface, performs text recognition processing on the text in the handwritten text image to obtain handwritten recognition text, matches the handwritten recognition text with questions, determines the target question corresponding to the handwritten recognition text based on the matching results, and performs answer grading processing on the handwritten recognition text based on the answer to the target question to obtain the grading result. Users are not required to answer questions in a restricted answer area, providing them with a more free answer area and accurately recognizing user answers, effectively improving the user's answering experience.

[0070] In one possible embodiment, the handwriting recognition module 31 performs character recognition processing on the text content in the handwritten content image to obtain handwritten recognized text, including:

[0071] The handwritten content image is segmented into multiple character content images;

[0072] Multiple character content images are input into a trained text recognition model, and the multiple character content is processed to obtain handwritten recognition text.

[0073] In one possible embodiment, the question matching module 32 performs question matching based on the handwritten recognition text, and determines the target question corresponding to the handwritten recognition text based on the question matching result, including:

[0074] The system performs question matching on handwritten digit recognition text in the question database and determines the target question corresponding to the handwritten digit recognition text based on the question matching results.

[0075] In one possible embodiment, the question matching module 32 performs question matching on handwritten digit recognition text in the question database, including:

[0076] Extract keywords and / or contextual semantic information from handwritten text;

[0077] Questions are matched against the question database based on keywords and / or contextual semantic information.

[0078] In one possible embodiment, the question matching module 32 performs question matching based on the handwritten recognition text, and determines the target question corresponding to the handwritten recognition text based on the question matching result, including:

[0079] The system matches the questions displayed on the question interface with the handwritten text and determines the target question corresponding to the handwritten text based on the matching results.

[0080] In one possible embodiment, the handwritten answer processing device further includes a habit analysis module, which is used for:

[0081] The handwriting feature information is extracted from the handwritten content image. The handwriting feature information includes one or more combinations of stroke order, writing speed, font size, font spacing and answer layout.

[0082] Based on writing characteristic information and preset writing habit standards, writing habit analysis is conducted to obtain writing habit analysis results.

[0083] In one possible embodiment, the handwriting answer processing device further includes a standardization suggestion module, which generates and displays writing standardization suggestions based on the analysis results of writing habits.

[0084] It is worth noting that in the above embodiments of the handwritten answer processing device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of this application.

[0085] This application also provides a handwritten answer processing device, which can integrate the handwritten answer processing apparatus provided in this application. Figure 4 This is a schematic diagram of the structure of a handwritten answer processing device provided in an embodiment of this application. (Reference) Figure 4 The handwritten answer processing device includes: an input device 43, an output device 44, a memory 42, and one or more processors 41; the memory 42 is used to store one or more programs; when one or more programs are executed by one or more processors 41, the one or more processors 41 implement the handwritten answer processing method provided in the above embodiments. The input device 43, output device 44, memory 42, and processors 41 can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0086] The memory 42, as a computing device-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as program instructions / modules corresponding to the handwriting answer processing method provided in any embodiment of this application (e.g., the handwriting recognition module 31, question matching module 32, and grading module 33 in the handwriting answer processing device). The memory 42 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on the use of the device. Furthermore, the memory 42 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 42 may further include memory remotely located relative to the processor 41, and these remote memories can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0087] Input device 43 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the device. Output device 44 may include display devices such as a display screen.

[0088] The processor 41 executes various functional applications and data processing of the device by running software programs, instructions and modules stored in the memory 42, thereby realizing the above-mentioned handwritten answer processing method.

[0089] The handwritten answer processing device, equipment, and computer provided above can be used to execute the handwritten answer processing method provided in any of the above embodiments, and have corresponding functions and beneficial effects.

[0090] This application embodiment also provides a storage medium for storing computer-executable instructions. When executed by a computer processor, the computer-executable instructions are used to execute the handwritten answer processing method provided in the above embodiment. The handwritten answer processing method includes: acquiring an image of handwritten content input on a question interface, and performing character recognition processing on the text content in the handwritten content image to obtain handwritten recognition text; matching questions based on the handwritten recognition text, and determining the target question corresponding to the handwritten recognition text based on the question matching result; and performing answer correction processing on the handwritten recognition text based on the question answer corresponding to the target question to obtain a correction result.

[0091] Storage medium – any type of memory device or storage device. The term “storage medium” is intended to include: mounting media, such as CD-ROMs, floppy disks, or magnetic tape devices; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (e.g., hard disks or optical storage); registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. Furthermore, storage media may reside in a first computer system in which a program is executed, or may reside in a different second computer system connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term “storage medium” can include two or more storage media that may reside in different locations (e.g., in different computer systems connected via a network). Storage media may store program instructions (e.g., specifically implemented as a computer program) executable by one or more processors.

[0092] Of course, the computer-executable instructions provided in the embodiments of this application are not limited to the handwriting answer processing method provided above, but can also execute related operations in the handwriting answer processing method provided in any embodiment of this application.

[0093] The handwritten answer processing device, equipment, and storage medium provided in the above embodiments can execute the handwritten answer processing method provided in any embodiment of this application. For technical details not described in detail in the above embodiments, please refer to the handwritten answer processing method provided in any embodiment of this application.

[0094] The above description is merely a preferred embodiment and the technical principles employed in this application. This application is not limited to the specific embodiments provided herein, and various obvious changes, readjustments, and substitutions that can be made by those skilled in the art will not depart from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the claims.

Claims

1. A method for processing handwritten answers, characterized in that, include: The handwritten content image entered on the question interface is obtained, and the text content in the handwritten content image is processed by character recognition to obtain handwritten recognized text; The handwritten text is matched with a question, and the target question corresponding to the handwritten text is determined based on the question matching result. The handwritten recognition text is graded based on the answer to the target question, and the grading result is obtained.

2. The handwritten answer processing method according to claim 1, characterized in that, The step of performing character recognition processing on the text content in the handwritten content image to obtain handwritten recognition text includes: The handwritten content image is segmented into multiple character content images; Multiple character content images are input into a trained text recognition model, and the multiple character content images are processed to obtain handwritten recognition text.

3. The handwritten answer processing method according to claim 1, characterized in that, The step of matching questions based on the handwritten recognition text and determining the target question corresponding to the handwritten recognition text based on the question matching result includes: The handwritten text is matched against a question database, and the target question corresponding to the handwritten text is determined based on the question matching result.

4. The handwritten answer processing method according to claim 3, characterized in that, The step of matching the handwritten recognition text with questions in the question database includes: Extract keywords and / or contextual semantic information from the handwritten text; Question matching is performed in the question database based on the keywords and / or the contextual semantic information.

5. The handwritten answer processing method according to claim 1, characterized in that, The step of matching questions based on the handwritten recognition text and determining the target question corresponding to the handwritten recognition text based on the question matching result includes: The question is matched with the handwritten text displayed on the question interface, and the target question corresponding to the handwritten text is determined based on the question matching result.

6. The handwritten answer processing method according to claim 1, characterized in that, After obtaining the image of the handwritten content input on the question interface, the method further includes: The handwritten content image is used to extract writing feature information, which includes one or more combinations of stroke order, writing speed, font size, font spacing, and answer layout. Based on the writing feature information and the preset writing habit standards, a writing habit analysis is performed to obtain the writing habit analysis results.

7. The handwritten answer processing method according to claim 6, characterized in that, After performing handwriting habit analysis based on the handwriting feature information to obtain the handwriting habit analysis result, the method further includes: Based on the analysis results of the writing habits, writing standard suggestions are generated and displayed.

8. A handwritten answer processing device, characterized in that, It includes a handwriting recognition module, a question matching module, and a grading module, among which: The handwriting recognition module is used to acquire the handwritten content image input on the question interface, and to perform character recognition processing on the text content in the handwritten content image to obtain handwritten recognition text. The question matching module is used to match questions based on the handwritten recognition text and determine the target question corresponding to the handwritten recognition text based on the question matching result. The grading module is used to grade the handwritten recognition text according to the answer to the target question, and obtain the grading result.

9. A handwritten answer processing device, characterized in that, include: Memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the handwritten answer processing method as described in any one of claims 1-7.

10. A storage medium for storing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the handwritten answer processing method as described in any one of claims 1-7.

Images